Environmental Science and Pollution Research (2022) 29:28166–28179

https://doi.org/10.1007/s11356-021-18467-2

RESEARCH ARTICLE

Carbon performance, company financial performance, financial

value, and transmission channel: an analysis of South African listed
companies
Fortune Ganda1

Received: 18 November 2021 / Accepted: 29 December 2021 / Published online: 5 January 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022

Abstract
This article examines the influence of carbon performance on corporate financial performance and company financial value
among South African listed firms for the period 2014 to 2018 using a two-step GMM panel process. The short-run findings
show that carbon performance develops a positive and significant association with return on assets, firm value, and Tobin’s
Q. In the long run, the relationship between carbon performance and return on assets as well as firm value is significantly
negative; however, the link with Tobin’s Q remains positively significant. Where carbon performance is employed as the
dependent parameter, a positive, significant relationship is established with return on assets, firm value, and Tobin’s Q in
both the short and long run. The findings also demonstrate that carbon performance is a transmission channel whereby the
debt-to-equity ratio, interest cover ratio, price to cash flow ratio, and current ratio improve corporate financial performance
and firm value in the long run. In the short run, the regression analysis frameworks produce mixed findings on whether carbon
performance is a transmission channel. Policy recommendations are made based on the findings.

Keywords Carbon performance · Return on assets · Tobin’s Q · Firm value

Introduction social responsibility. Since the adoption of the Rio Declara-

tion on Environment and Development, sustainability issues
Profit-oriented companies seek to optimise shareholder value have been high on the agenda of governments, companies,
by producing commodities whose sale to consumers yields and individuals. Many business sectors produce emissions as
significant financial gains (Abban and Hasan 2021; Ganda well as waste that is detrimental to the natural environment
et al. 2015). However, from a social, economic, and politi- and can cause irreversible damage to natural resources (Velte
cal perspective, modern companies also have responsibilities et al. 2020). Companies should thus be accountable and adopt
to society. They are thus called on to respond to the needs initiatives that preserve the environment.
of diverse parties who influence and are influenced by their Efforts to reduce environmental degradation triggered by
business practices. These stakeholders could be internal (for firm practices are indicators of company environmental and/or
instance, investors and employees) and external (for example, green performance. Worldwide, governments and other bodies
suppliers, clients, the local community, government, and envi- have called on companies to reduce the negative environmen-
ronmental interest groups). Carbon performance aimed at mit- tal effects of their operations and to promote sustainability
igating carbon emissions is an imperative aspect of corporate (Lin et al. 2019; Zhang and Vigne 2021). This includes reduc-
ing and/or eliminating carbon emissions as well as producing
green commodities (Ganda 2021a, b). Furthermore, firms
Responsible Editor: Nicholas Apergis
are required to adhere to green regulations and incorporate
* Fortune Ganda environmentally compatible technologies which reduce their
fganda@wsu.ac.za; Fochi555@yahoo.com; ecological footprint. However, compliance with such require-
fortune.ganda@gmail.com ments appears to be more complex in emerging economies
1 like South Africa than in developed economies (Ganda 2019).
Department of Accounting, Faculty of Management
Sciences, Walter Sisulu University, Butterworth Campus, Carbon performance’s impact on financial performance and
Private Bag X3182, Butterworth 4980, South Africa firm value is thus likely to differ in different contexts.

13
Environ Sci Pollut Res (2022) 29:28166–28179 28167

This study focuses on South Africa for numerous rea- the relationship between carbon performance and financial
sons. The country has recorded overexploitation of natural value, and vice versa. To the best of the author’s knowledge,
resources as well as increased carbon emissions in the past this has not been done previously in an African company set-
few decades. South Africa is one of the most industrialised ting. Finally, few studies have been conducted on this rela-
economies on the African continent. Carbon Brief (2018) tionship in emerging economies. The article thus offers new
highlights that, due to over-reliance on coal, it is the 14th insights into sustainable development within these countries.
largest emitter of emissions in the world. The report also The remainder of the article is arranged as follows: the
notes that 45% of South Africans regard climate change Literature review section reviews the recent literature on the
as a serious problem for the country’s development. The topic, while the Methodology section focuses on the data
South African Department of Environmental Affairs (2014) and econometric procedure adopted. The Results and dis-
observes that energy firms (solid and liquid fuels), road cussion section presents and discusses the results, and the
transport, manufacturing, and construction (solid fuels) Implication of the study section considers their implications.
are responsible for 95% of greenhouse gas emissions. The The Conclusion section concludes the article.
Climate Action Tracker (CAT) (2020) explains that the
approval of the Integrated Resource Plan (IRP) in October
2019 paved the way for a carbon-intense recovery process Literature review
that promotes a shift from non-renewable resources towards
green energy sources. Moreover, the carbon tax of R46 per Previous studies on the relationship between carbon per-
tonne of carbon dioxide equivalent (tCO2e) that was intro- formance and corporate financial performance and/or the
duced in 2019 is an important emission reduction strategy. financial value used diverse samples and methodologies.
This tax, which was increased by 5.6% in 2020 (ESI Africa However, they produced mixed findings, and consensus has
2020), aims to encourage firms to adopt emission reduction not yet been reached on the nature of this relationship.
policies and adopt green technologies. Ashraf et al.’s (2020) analysis of South Asian cement manu-
It is against this background that this article focuses on facturing firms demonstrates that financial slack generates a
South African CDP companies listed on the Johannesburg positive relationship with carbon performance and that this
stock exchange (JSE) over the period 2014–2018 to firstly link is moderated through carbon prices in a negative direction.
analyse the influence of firm carbon performance on corporate However, company density was found to positively moderate
financial performance and/or firm value in both short- and this association. Okafor et al.’s (2021) research on the top 100
long-run settings. Secondly, it examines the impact of com- US tech companies listed on the S&P 500 highlights that firms
pany corporate financial performance and/or firm value on that spend more on socially responsible initiatives achieve both
carbon performance. Thirdly, the article investigates if, at the increased revenue and profit. Similarly, Trinks et al.’s (2020)
corporate level, carbon performance is a transmission chan- survey of 1572 international companies found that carbon-effi-
nel whereby other variables (in this article, the debt-to-equity cient firms achieve superior financial performance. Wang et al.
ratio, interest cover ratio, price to cash flow ratio, and current (2020) examined 289 Chinese companies and concluded that
ratio) promote corporate financial performance and/or firm environmental information reporting positively influences cor-
value. It is anticipated that the findings will unveil vital policy porate financial performance directly and indirectly (through
concerns that need to be addressed at corporate, industrial, analyst coverage, the number of reports, and several analysts).
and national levels. In turn, this will enhance the develop- A survey of 201 quoted small and medium enterprises
ment of green programmes and policies and green investment (SMEs) in the UK (Boakye et al. 2021) found that SMEs
in companies in emerging economies such as South Africa. that optimise environmental management achieve improved
This article adds to the existing body of knowledge in financial performance. Velte et al.’s (2020) quantitative
a number of respects. Previous empirical research mainly review of 73 previous studies demonstrates that carbon per-
focused on the relationship between carbon performance formance improves financial performance, that carbon per-
and financial performance. This article expands this anal- formance and carbon disclosure are positively associated,
ysis by including several constructs. Furthermore, both and that both are positively affected by the composition of
short-run and long-run scenarios are examined. A single the corporate board. Zhang and Vigne (2021) highlight that a
period analysis does not add much value to a firm strategic financing-emission reduction policy punishes firms respon-
decision-making. The article examines whether carbon per- sible for high levels of pollution and that such firms also suf-
formance is a transmission channel (in the short and long fer low levels of total factor productivity, sales growth, and
term), whereby other control variables at corporate settings firm profitability. Abban and Hasan’s (2021) analysis estab-
promote financial performance. Furthermore, it adds another lishes that improved environmental performance enhances
important variable, corporate financial value as a dependent financial performance and points to bi-directional causality
variable as well as an independent variable to investigate between the former and the latter.

13
28168 Environ Sci Pollut Res (2022) 29:28166–28179

Lin et al. (2019) evaluated 163 international automotive Methodology

companies and determined that a green innovation strategy
(GIS) positively affects firm financial performance. Fernández- This section presents the data and econometric procedure adopted.
Cuesta et al.’s (2019) investigation of 428 listed firms in 16
European countries shows that carbon risk alongside capital Data
expenditure is the major causes of financial debt. The positive
effect of carbon emissions on debt (driven by the role of The sample consisted of 107 CDP South African companies
emissions) was minimised by the companies’ environmental listed on the Johannesburg Stock Exchange Socially Respon-
performance. Tzouvanas et al. (2020) surveyed 288 sible Index (JSE SRI) for 5 years (2014–2018). Purposive
European manufacturing firms and concluded that improved sampling was employed as these firms’ data is always avail-
environmental performance results in firm financial gain. able. In this regard, these firms possess sound characteristics
Therefore, the association between environmental and financial in terms of sustainability as they regularly comply with the
performance is positive, although the magnitude is different in JSE SRI triple bottom specifications (JSE 2021). This sam-
different quantile ranges. Finally, Agyabeng-Mensah et al.’s ple of firms also engages in CDP climate change initiatives
(2020) research on 240 firms in Ghana supports the notion that (CDP 2018). Table 1 describes the variables.
integration of green logistics management has less impact in
improving communities’ social wellbeing and health, while Econometric procedure
it does improve corporate financial performance (through
environmental and market initiatives). This research takes into account a diverse range of 4 panel
In view of the review above, evidence confirm the valid- frameworks. The multiple linear regression framework for
ity of support of both sides about the influence of carbon individual i = 1, 2, 3… N who is scrutinised at distinct periods
performance on firm performance. In light of this complex- t = 1, 2, 3…T. Suppose this regression model is founded on:
ity, a company’s participation and/or nonengagement is, to
a greater extent, the management concern. This research yit = θ + ai + βx�it + … … ⋯ + uit (1)
examines this debate from an emerging economy perspec-
In this context, yit is the dependent variable, xit refers
�

tive by evaluating specific time periods (short run and long

to the K-dimensional row vector independent variables
run); hence, it is comprehensive in nature. Moreover, carbon
considered to be varying with time, and θ is the intercept.
performance transmission channel tendencies in promoting
Furthermore, ai is an individual particular impact. β is the
financial performance are also analysed which further give
K-dimensional column vector of variables. uit is the error
manager’s insights on whether this variable is a medium,
term of the regression model.
hence aiding firm decision making and policy development.

Table 1  Variables employed

Variable Definition Unit Source

ROA Return on assets Computed value by INET BFA INET BFA database
FIV Firm value The proxy for firm value is computed as follows: common equity market INET BFA database
value = number of shares outstanding × price per share of the company’s
common stock at a specific year-end. Reference includes Matsumura et al.
(2014)
Tobin’s Q Tobin’s Q Tobin’s Q = (FMV + LV + LL) ÷ TA, where FMV is the market value of the INET BFA database
company’s outstanding shares, LV refers to the liquidating value of the
company’s outstanding preferred shares, debt demotes the value of long-
term liabilities, and TA is the book value of the company’s total assets.
Reference sources include Lahouel et al. (2021) and Chung and Pruitt
(1994)
CARPER Carbon performance Carbon performance = total carbon emissions at year-end (tonnes Company sustainability reports
­CO2) ÷ total company sales at year-end. Reference sources include Luo and INET BFA database
Tang (2014); Sutantoputra et al. (2012) and Clarkson et al. (2008)
DE Debt to equity ratio Computed value by INET BFA INET BFA database
IC Interest cover ratio Computed value by INET BFA INET BFA database
PCF Price cash flow ratio Computed value by INET BFA INET BFA database
CR Current ratio Computed value by INET BFA INET BFA database

In this case, ROA, FIV, and TOB are dependent variables. CARPER is the independent variable. DE, IC, PCF, and CR are control variables

13
Environ Sci Pollut Res (2022) 29:28166–28179 28169

This first panelised model is based on the fixed effects yit = θ + ai + σyit−1 + β1 x�it + … … ⋯ + uit (4)
(FE) framework. It assumes the existence of a mutual rela-
tionship of unknown form involving the entity’s error term In the context, we perceive that some components of xit are
′

and independent factors. Moreover, the entity’s error term endogenous. Vector zi refers to instruments comprising of the
along with individual-particular impact are not supposed to parts of x which are exogenous along with more factors that
be mutually associated with the other parameters. are uncorrelated with ui. θ is the intercept of the regression,
The second model specification employed is the ran- while ai is the individual-particular effect. In the GMM model
dom effects (RE) model. Competing with the FE model, approach, the moment conditions are identified as follows:
the RE framework assumes that the individual-particular
E zit uit (β) = E zit (yit − (β)x�it = 0 (5)
{ } { }
effect is a random parameter that is not mutually related
to the independent variables. Nevertheless, in cases where
In this situation, the instruments employed are x (i, t−1),
the individual-particular effect is, in fact, unrelated, the RE
x (i, t−2), x (i, t−3) … (i,1). GMM estimators select values that
framework is preferable to the FE framework. This decision
reduce the quadratic framework of the regression. This tech-
is normally achieved by deploying the Hausman test that is
nique is effective in addressing over-identified configurations
only arguable under homoscedasticity and cannot take fixed
of the model. In addition, the GMM method minimises the
effects into account.
method of moments in cases where the number of variables
The instrumental variables (IV) framework is the third
is equivalent to the number of moment conditions.
model. It states that a selected number of exogenous
explanatory variables are assumed to be endogenous. In
this case, they could be mutually related to the error. The
Results and discussion
IV estimation process is known to considerably mitigate
endogeneity effects. However, the procedure also violates
This section presents the descriptive statistics and the cor-
the zero conditional mean assumption through endogenous
relation matrix of the variables employed, as well as the
variables. This may arise due to estimation error in the
two-step GMM findings.
explanatory variables alongside omitted-variable bias. While
the IV approach has drawbacks, it directs channels to acquire
Descriptive statistics
more compatible variable estimates and results. Thus, the IV
model is expressed as
Summarised data on the sampled companies is presented in
yit = 𝛽1 xit1� + 𝛽2 xit2� + u1it (2) Table 2 below. Amongst the dependent variables, firm value
(FIV) appears to have the highest mean at 22.3 (the second is
�
ROA at 6.88 and the last is Tobin’s Q at 0.65). Thus, for every
xit1 = 𝛼1 xit1� + 𝛼2 xit2 + u2it (3) rand of common stock, investors, or the market perceive an
average firm value of R22.30. The carbon performance mini-
In this situation, i = 1, 2, 3…N and t = 1, 2, 3…T. xit1′
mum and maximum values range from –3.593785 to 268.4,
refers to the vector of endogenous independent factors. xit2′
pointing to large differences in the firms’ carbon emissions
refers to the vector of endogenous independent factors. xit3′
reduction initiatives. The mean carbon performance is 1.359.
refers to the vector of instrumental parameters. β1 together
On average, a company undertakes 1.359 activities in reduc-
with β2 structural parameter vectors. α1 and α2 are reduced-
ing carbon emissions. The kurtosis of the control variables,
form variables. The vector (u1it; u2it ) is considered to be mul-
namely, the debt-to-equity ratio (DE), interest cover (IC),
tivariate normal with variance–covariance matrix:
price to cash flow (PCF), and current ratio (CR), are all more
�
� ∑�
1 12 1
� than 3 (excess kurtosis > 0). This result illustrates that the dis-
Var (u1it ; u2it ) = = ∑ ∑� tribution of all these ratios is leptokurtic (the tails are longer
12 1 22 1
and fatter, and the centre is very sharp and high). Moreo-
The framework is computed by employing Newey’s effi- ver, in the case of control factors, only PCF is negatively
cient two-step estimator. skewed—the tail of the distribution points leftward. This
Finally, this article deploys the IV-GMM model due to the implies that, while initially a few investors are prepared to
presence of heteroscedasticity. This model is more effective pay for cash flow produced by the company; the substantial
than the usual IV method. Furthermore, if heteroscedasticity majority are long term. The remaining control parameters
is not found, the GMM approach is no worse asymptotically have positively skewed PCFs.
when compared to the IV method. For instance, we employ
a simple regression framework:

13
28170 Environ Sci Pollut Res (2022) 29:28166–28179

Table 2  Statistical summary of Variable Min Std. dev Max Mean Skewness Kurtosis
variables
ROA − 121.8945 12.27552 46.85 6.876665 − 2.451006 28.02151
FIV 15.39487 1.949516 31.00545 22.30046 0.9638672 7.9811
Tobin’s Q − 0.0000249 3.340955 44.9605 0.6536921 10.47815 115.288
CARPER − 3.593785 12.4063 268.4 1.359019 19.06794 403.9869
DE − 25.95 2.99269 15.18 1.78686 0.5641117 20.50958
IC − 33.09 236.1772 3076 36.47366 10.10143 116.6607
PCF − 211 17.3625 94 10.6161 − 6.64145 95.03137
CR 0 9.15024 159.31 2.174017 13.63601 206.121

Correlations Table 4 outlines the findings regarding the GMM short-

run contexts on the link between carbon performance and
Table 3 below indicates a one-to-one relationship between financial performance and/or value. It shows that lagged
the variables utilised in this article. It is evident that the ROA and lagged FIV have a significant positive relation-
direct association involving carbon performance (CARPER) ship with ROA and FIV in the short run, although this link
and the dependent factors (ROA, FIV, Tobin’s Q) is negative is significantly negative in cases where lagged Tobin’s Q is
in all cases. However, this outcome is not absolute as vital involved. In this case, a 1% increase in previous ROA and
econometric processes are still outstanding due to the limita- previous FIV leads to an increase in ROA and FIV by 0.747
tions associated with simple correlations. On a separate note, and 0.720%, respectively. This finding is consistent with
the correlation matrix is imperative in the selection of instru- Heikal et al.’s (2014) study on automotive firms listed on
ments to be deployed in the GMM procedure. In this respect, the Indonesian Stock Exchange, which highlights that ROA
variables with the lowest correlation to the dependent vari- results in significant growth in company income. However,
ables are employed as instruments in the GMM analysis. a percentage rise in the previous Tobin’s Q in this research
reduces Tobin’s Q by 0.496%.
GMM findings Carbon performance among South African listed compa-
nies shows a positive and significant association with ROA,
Previous studies (for instance, Lahouel et al. 2021 and FIV, and Tobin’s Q in the short term. More specifically, a
Moneva et al. 2020) state that addressing endogeneity mat- 1% surge in carbon performance stimulates a 0.02%, 0.008%,
ters and accounting for the presence of dynamic impacts and 0.0016% increase in ROA, FIV, and Tobin’s Q, respec-
are major aspects when analysing the influence of carbon tively. This is in line with Busch et al.’s (2020) exploration
performance on financial performance and/or firm value, of publicly listed stock companies from 2005 to 2014, which
and vice versa. For procedural purposes and robustness found that increased carbon emissions (employed as a proxy
checks, the findings from traditional static estimators (ran- for carbon performance) are related to robust, enhanced
dom effects and fixed effects models) are computed along short-run corporate financial performance.
with the instrumental variable approach. These findings are The DE shows a negative and significant link with ROA
presented in the Appendix (see Tables 8, 9 and 10). Since in the short run. Thus, the finding illustrates that South
this article is interested in accounting for endogeneity and African companies which employed debt finance to acquire
dynamism effects, the GMM outcomes are presented in a assets experience a negative association with and/or effect
short-run and long-run setting. on corporate financial performance – ROA. This concurs

Table 3  Showing correlation Variable ROA FIV TOB CARPER DE IC PCF CR

matrix
ROA 1
FIV 0.1034 1
Tobin’s Q 0.0810 0.1081 1
CARPER − 0.0081 − 0.0757 − 0.0107 1
DE − 0.1425 0.0492 − 0.0110 − 0.0180 1
IC 0.2823 − 0.0183 − 0.0207 − 0.0156 − 0.0694 1
PCF 0.1199 0.0168 − 0.0073 − 0.0052 0.0013 0.0544 1
CR − 0.1882 − 0.1375 − 0.0142 − 0.0047 − 0.0778 0.0739 0.0350 1

13
Environ Sci Pollut Res (2022) 29:28166–28179 28171

Table 4  Two-step system-GMM short-run findings with (a) ROA; (b) FIV; (c) Tobin’s Q
Regression—(ROA) Regression—(FIV) as Regression—(Tobin’s
as dep. variable dep. variable Q) as dep. variable

Coefficient Standard error Coefficient Standard error Coefficient Standard Error

ROAit−1 0.7472047 (0.000) 0.0099655
***
FIV it−1 0.7204744 (0.000) 0.0249979
***
�
Tobin sQit−1 − 0.496234 (0.000) 0.0583158
***
CARPER 0.0206518 (0.095) * 0.0123839 0.0080706 (0.014) ** 0.0032898 0.001668 (0.000) *** 0.0003351
DE − 1.400456 (0.000) 0.1313002 0.0733769 (0.000) 0.0074897 0.023231 (0.092) * 0.0137743
*** ***
IC 0.0033631 (0.000) 0.0003273 0.0001763 (0.000) 0.0000282 − 0.0002345 (0.000) 0.0000231
*** *** ***
PCF 0.0573583 (0.000) 0.005974 0.0002937 (0.792) 0.0011159 − 0.0004129 (0.468) 0.0005693
***
CR − 0.1274749 (0.000) 0.0033158 − 0.0079984 (0.000) 0.0004392 − 0.0023642 (0.149) 0.0016386
*** ***
Constant 3.44786 (0.000) *** 0.2313527 6.130266 (0.000) *** 0.5447417 0.7053676 (0.000) 0.0378188
***
Wald (χ2) 44,519.90 (0.000) *** 3378.33 (0.000) *** 228.99 (0.000) ***
Arellano-Bond test z =  − 1.77 z = -2.60 z = 0.67
for AR (1) in first Pr > z = 0.077* Pr > z = 0.009*** Pr > z = 0.0505*
differences
Arellano-Bond test z =  − 0.74 z = 0.23 z =  − 1.53
for AR (2) in first Pr > z = 0.459 Pr > z = 0.818 Pr > z = 0.126
differences
Hansen test of over- chi2(37) = 42.23; chi2(12) = 16.05; chi2(12) = 10.98;
identifying. Restric- Prob > chi2 = 0.255 Prob > chi2 = 0.189 Prob > chi2 = 0.531
tions
Observations 427 427 427

[1] ***, **, and * indicate that the coefficients are significant at the 1%, 5%, and 10% level of significance, respectively. Numbers in brackets
are p-values. [2] The null hypothesis of diagnostic statistical analysis shown in the table above is a The Arellano-Bond test for autocorrelation:
­H0 = no autocorrelation; b The Hansen: ­H0 = the set of instruments is valid

with Ullah et al.’s (2020) research on the textile sector in On the one hand, the PCF ratio shows a negative and
Pakistan during the period 2008 to 2017. However, other DE significant connection with Tobin’s Q in the short run. On
regression outcomes indicate that this variable has a positive the other, the PCF ratio generates a positive and significant
and significant relationship with FIV and Tobin’s Q in the relationship with both ROA and FIV in the short run. In
short term. In this case, a 1% increase in DE results in FIV this respect, a 1% climb in PCF produces a 0.057% and
and Tobin’s Q increasing by 0.07% and 0.02%, respectively. 0.00029% increase in ROA and FIV, respectively. This is in
The IC ratio of the listed companies develops a posi- line with Chakravarty and Hegde’s (2019) investigation of
tive and significant relationship with both ROA and FIV 55 distinct industries on the Indian Boards Database. Their
in the short term. As such, 1% growth in IC results in a study postulates that cash flows generated a positive and sig-
0.0033631% and 0.0001763% improvement in ROA and nificant association with ROA and the market-to-book ratio.
FIV, respectively. A higher interest coverage ratio means The CR generates a negative and significant link with all
that the firm is able to repay its debts. This concurs with proxies for financial performance and financial value in the
Leepsa and Mishra’s (2012) study on Indian manufacturing short run. For instance, a 1% rise in CR leads to a 0.127%,
companies from 2003/04 to 2006/07, where interest cover as 0.00799%, and 0.0023% reduction in ROA, FIV, and Tobin’s
a proxy for leverage showed a positive relationship with firm Q. Purwanto and Agustin (2017) evaluated 65 companies
financial performance. Nevertheless, IC generates a nega- listed on the Indonesian Stock Exchange for the period 2009
tive and significant link with Tobin’s Q. More particularly, to 2014. The study found that an increase in the current ratio
a 1% escalation in IC reduces Tobin’s Q by 0.0002345%. decreases the firm financial value. Theoretically, a higher CR

13
28172 Environ Sci Pollut Res (2022) 29:28166–28179

results in increased firm value and vice versa since investors in company strategy and hence tend to be overlooked. More
prefer highly liquid companies. specifically, the CDP (2018:3) asserts that “Despite South
The diagnostic tests presented in Table 4 above also African companies having been global leaders in supporting
validate the results in this section. It is apparent that for the low carbon transition for the first decade of CDP, in 2016
all regression models, the first-order serial correlations are and 2017, the data indicated that South African companies
rejected, but the second-order serial correlations are accepted may be faltering in some respects. In 2018, the data is mixed
at a 5% significance level, thereby validating the instruments but the integration of climate change into governance is not
employed in this research. Furthermore, the Hansen tests producing the actions we need.”
are not weakened by employing the instruments as the null Secondly, the GMM results in Table 6 indicate that the
hypothesis in all the regression models is not rejected. financial performance proxies (ROA, Tobin’s Q) and the
Table 5 present the results on the long-run relationship FIV show a positive and significant relationship with car-
between carbon performance and corporate financial per- bon performance in the short term. As such, a 1% rise in
formance and/or firm value. It shows that a 1% increase in ROA, Tobin’s Q, and FIV increase carbon performance by
carbon performance for the South African listed firms sig- 0.0136%, 0.0026%, and 0.199%, respectively, in the short
nificantly reduces ROA and FIV by 0.727% and 0.712%, run. Alexopoulos et al.’s (2018) analysis of 931 Greek manu-
respectively, in the long run. These results conflict with facturing firms during the period 2001 to 2007 concurs and
Trinks et al.’s (2020) study of 1572 international compa- adds that companies with superior financial performance tend
nies over the years 2009 to 2017, which concludes that to achieve improved environmental performance (inverted
high carbon-efficient firms have superior financial perfor- score of environmental pollution per production unit).
mance. However, the current study’s finding that a percent- The GMM outcomes also show that the DE ratio is sig-
age upswing in carbon performance heightens Tobin’s Q by nificantly positively related to carbon performance in the
0.497% concurs with Trinks et al. (2020). The debt-to-equity case of the ROA and Tobin’s Q regression models, but
ratio, interest cover ratio, price to cash flow ratio and current the link is significantly negative in the context of the FIV
ratio also have significantly negative connections with both regression framework. Since the DE ratio ascertains the
ROA and FIV in the long run. However, all these variables’ degree to which a firm is financing its operations by means
link with Tobin’s Q is significantly positive in the long run. of debt versus wholly-owned funds, South African com-
Table 6 illustrates the GMM short-run outcomes when panies could be attracting investors due to their ability to
carbon performance is employed as a dependent variable for demonstrate company greening initiatives, thereby enhanc-
both the financial performance and financial value regres- ing firm financial performance. However, in the case of
sion frameworks. Firstly, lagged carbon performance in a the FIV model, increased DE significantly reduces car-
regression model (ROA, FIV, and Tobin’s Q) demonstrates a bon performance. Thus, it can be determined that carbon
significantly negative association with carbon performance. performance is not a transmission channel whereby the
Thus, a 1% upswing in carbon performance reduces current DE ratio promotes corporate financial value among these
ROA, FIV, and Tobin’s Q by 0.27%, 0.46%, and 0.325%, listed firms. In theory, a high DE ratio implies high risk
respectively, in the short term. This finding demonstrates (firm financing its growth with debt), negatively influenc-
that South African companies’ previous carbon performance ing firm value.
initiatives are not able to influence existing carbon perfor- The IC ratio is positive and significantly associated with
mance measures. This could be due to weak regulatory prac- carbon performance in Tobin’s Q model in the short run,
tices, a lack of greening skills initiatives by companies and although that connection is negative and significant in the con-
the fact that carbon management practices are not a priority text of the ROA and FIV regression frameworks. Interest cover

Table 5  Two-step system-GMM long-run findings with (a) ROA; (b) FIV; (c) Tobin’s Q
Regression -(ROA) as dep. Regression—(FIV) as dep. Regression—(Tobin’s
variable variable Q) as dep. variable

Coefficient Standard error Coefficient Standard error Coefficient Standard error

CARPER − 0.7265529 (0.000) *** 0.0168725 − 0.7124038 (0.000) *** 0.0251853 0.497902 (0.000) *** 0.0584578
DE − 2.14766 (0.000) *** 0.1310445 − 0.6470975 (0.000) *** 0.0235936 0.5194649 (0.000) *** 0.0659242
IC − 0.7438416 (0.000) *** 0.0101811 − 0.720298 (0.000) *** 0.0250093 0.4959995 (0.000) *** 0.0583175
PCF − 0.6898464 (0.000) *** 0.0121235 − 0.7201807 (0.000) *** 0.0251933 0.495821 (0.000) *** 0.0581792
CR − 0.8746796 (0.000) *** 0.0076558 − 0.7284727 (0.000) *** 0.0247444 0.4938698 (0.000) *** 0.0591767

***, **, and * mean significant at 1%, 5%, and 10% significance level, respectively. Numbers in brackets are p-values

13
Environ Sci Pollut Res (2022) 29:28166–28179 28173

Table 6  Two-step system-GMM short-run findings with carbon performance as a transmission channel for firm financial performance and value
Regression Regression— Regression—
-(CARPER) as dep. (CARPER) as dep. (CARPER) as dep.
variable variable variable

Coefficient Standard error Coefficient Standard error Coefficient Standard error

CARPERit−1 − 0.2741592 (0.000) 0.003752 − 0.4604968 (0.000) 0.0490563 − 0.3250628 (0.000) 0.0022732
*** *** ***
ROA 0.0135549 (0.000) 0.0003534
***
FIV 0.1999875 (0.028) ** 0.0907402
Tobin’s Q 0.0026182 (0.000) 0.0004617
***
DE 0.0041483 (0.047) ** 0.0020848 − 0.0277709 (0.502) 0.0414074 0.0134917 (0.000) 0.0011308
***
IC − 0.0000179 (0.000) 2.91e-06 − 0.0002836 (0.038) 0.0001368 0.0001307 (0.000) 0.0000141
*** ** ***
PCF 0.0018419 (0.000) 0.0002067 − 0.0034768 (0.005) 0.0012481 0.0022901 (0.000) 0.0000739
*** *** ***
CR − 0.0037854 (0.000) 0.0001015 0.0010524 (0.815) 0.0044987 − 0.0057417 (0.000) 0.0004791
*** ***
Constant 0.5615374 (0.000) 0.0982053 − 3.806417 (0.037) 1.826601 0.9340579 (0.000) 0.0743434
*** ** ***
Wald (χ2) 10,751.76 90.85 47,313.40
(0.000) *** (0.000) *** (0.000) ***
Arellano-Bond test z =  − 1.37 z =  − 1.40 z =  − 1.38
for AR (1) in first Pr > z = 0.171 Pr > z = 0.160 Pr > z = 0.166
differences
Arellano-Bond test z = 1.30 z = 1.10 z = 1.29
for AR (2) in first Pr > z = 0.193 Pr > z = 0.272 Pr > z = 0.198
differences
Sagan test of over- chi2(40) = 12.16; chi2(12) = 0.71; chi2(41) = 6.90;
identifying. Restric- Prob > chi2 = 1.000 Prob > chi2 = 1.000 Prob > chi2 = 1.000
tions
Observations 427 427 427

[1] ***, **, and * indicate that the coefficients are significant at the 1%, 5%, and 10% level of significance, respectively. Numbers in brackets
are p-values. [2] The null hypothesis of diagnostic statistical analysis shown in the table above is a The Arellano-Bond test for autocorrelation:
­H0 = no autocorrelation; b The Sagan: ­H0 = the set of instruments is valid

determines the firm’s ability to honour its debts and a high ratio company is trading at a high price but is unable to obtain
indicates better financial health. As such, when market-based adequate cash flow to promote firm operations. The CR sig-
proxies such as Tobin’s Q are employed, carbon performance nificantly demotivates carbon performance in the ROA and
is a transmission channel through which the IC ratio motivates Tobin’s Q frameworks in the short term, but significantly
firm financial performance rather than through accounting- promotes carbon performance in the FIV framework.
based proxies such as ROA and firm financial value. The diagnostic test in Table 6 demonstrates there is no
The PCF ratio is positive and significantly related to car- first-order autocorrelation in the disturbances of all the
bon performance in the ROA and Tobin’s Q models for the regression models, thereby concurring with Habimana
listed South African companies, but the link is significantly (2017) and De Andres and Vallelado (2008). Most impor-
negative in the case of the FIV framework. For both account- tantly, for this study, the absence of second-order serial cor-
ing-based measures (ROA) and market-based indicators relation in the disturbances is accepted for all regression
(Tobin’s Q), carbon performance is an identifiable transmis- equations. The Sargan test also demonstrates that the prob-
sion channel whereby PCF promotes corporate financial per- lem of over-identification is not present in the equations.
formance. However, in the case of FIV, PCF decreases carbon Table 7 shows very interesting results in relation to car-
performance and hence ceases to be a transmission channel. bon performance as a transmission channel for firm financial
The theory supports these FIV findings as a high PCF ratio performance and value on a long-run basis. It appears that in
is less preferable than low PCF ratios since it shows that the a long-term setting, higher return on assets, firm value, and

13
28174 Environ Sci Pollut Res (2022) 29:28166–28179

Tobin’s Q increases corporate carbon performance. Thus, the way forward. There is also a need to establish sustain-
high financial performance and firm value are likely to spur ability accounting departments in South African companies,
high carbon performance among South African companies. and sustainability issues should form part of their strategic
Other findings also illustrate that a heightened DE ratio, IC policy. In the long run, this will enable companies to reap the
ratio, PCF ratio, and CR significantly increase carbon per- benefits of incorporating zero-carbon practices.
formance in the long run. It also implies that, in the long This research also illustrated that when carbon per-
term, carbon performance is a vital transmission medium formance was adopted as the dependent variable, critical
through which these variables promote improved financial results were generated. Firstly, lagged carbon performance
performance and firm value. in all the regression models (ROA, FIV, and Tobin’s Q)
has a significantly negative association with carbon per-
formance in the short term. This points to how the com-
Implications of the study: examining panies under study practice business. The fact that previ-
the context of carbon performance ous carbon performance has a significantly negative link
with current carbon performance suggests that these South
This article produced critical findings in the context of carbon African companies may be abandoning carbon emission
performance that has implications for policy development, mitigation initiatives undertaken in the past. Adopting new
corporate practice, theory, and future research. It showed or alternative strategies in subsequent years will produce
that carbon performance develops a positive and significant negative results. The previous year’s zero-carbon activities
association with ROA, FIV, and Tobin’s Q in the short term. in these companies should inform the carbon performance
Thus, it is apparent that green issues are well received by policy of the following years. There is also a need for com-
South African companies in the short run. There is a ten- panies to critically analyse the pros and cons and lessons
dency for these firms to integrate practices that are impera- learnt, and use such findings to inform current and future
tive to achieve low-carbon emissions. However, in the long mitigation plans. In addition, efficient records should be
run, the relationship between carbon performance and either maintained on sustainability issues in the context of car-
ROA or FIV tends to be significantly negative although its bon emissions as these will inform future green practices.
relationship with Tobin’s Q remains positively significant. It was found that ROA, FIV, and Tobin’s Q had a posi-
This points to gaps in the integration of sustainability issues tive and significant connection with carbon performance for
in company policy, particularly for the long term. It would the analysed companies in the short and long run. Thus,
seem that the measures adopted to promote emissions reduc- when the companies experienced improved financial perfor-
tion are not deeply embedded in company strategic policy. It mance and value in the short run, their carbon performance
is for this reason that a negative association is demonstrated was enhanced. This concurs with the earlier findings where
in the long term. These results can be explained by a num- carbon performance was the explanatory parameter in the
ber of factors. Firstly, government policy may not have been short term. Hence, in the short run, these variables and their
strong enough to motivate companies to engage in initiatives relationships significantly complement one another. Ideally,
to mitigate carbon emissions and to sustain such practices. A there is a need for the companies under study to extend their
lack of green human capital could be another factor. Insuf- short-term high level of commitment to carbon reduction
ficient expertise may prevent companies from understanding initiatives to the long term. For example, durable green

Table 7  Two-step system-GMM long-run findings with carbon performance as a transmission channel for firm financial performance and value
Regression -(CARPER) Regression—(CARPER) Regression—(CARPER)
as dep. variable as dep. variable as dep. variable

Coefficient Standard error Coefficient Standard error Coefficient Standard error

ROA 0.2877141 (0.000) *** 0.0038428
FIV 0.6604843 (0.000) *** 0.1109357
Tobin’s Q 0.327681 0.0023725
(0.000) ***
DE 0.2783075 (0.000) *** 0.0035449 0.4327259 (0.000) *** 0.0639482 0.3385545 (0.000) *** 0.0022479
IC 0.2741413 (0.000) *** 0.0037529 0.4602132 (0.000) *** 0.0490218 0.3251934 (0.000) *** 0.0022833
PCF 0.2760011 (0.000) *** 0.003709 0.4570199 (0.000) *** 0.0487102 0.3273529 (0.000) *** 0.0023018
CR 0.2703738 (0.000) *** 0.0037421 0.4615492 (0.000) *** 0.0498491 0.319321 (0.000) *** 0.0019557

***, **, and * mean significant at 1%, 5%, and 10% significance level, respectively. Numbers in brackets are p-values

13
Environ Sci Pollut Res (2022) 29:28166–28179 28175

technologies, sustainable green investment practices and company could consider would be to lower the among bor-
conduct in company projects, and permitting top manage- rowed for non-climate friendly projects or operations or
ment to participate in matters relating to firm green policy pay current debts while reducing operating costs for non-
are vital to ensure sustenance. environmentally compatible activities.
The results also showed that the DE ratio is significantly Furthermore, the results confirm that the price to cash
positively related to carbon performance in the ROA and flow ratio is positive and significantly related to carbon per-
Tobin’s Q regression frameworks, but it is significantly formance for indicators of firm financial performance, but
negative in the context of the FIV regression framework in this link is significantly negative in the case of firm financial
the short term (in the long-term, significantly positive rela- value in the short run. In the long run, the price to cash flow
tionships are valid in all equations). While carbon perfor- ratio develops a significantly positive connection with all
mance is determined to be a transmission channel whereby the financial performance measures as well as firm value.
the DE ratio promotes corporate financial performance in Therefore, in most cases, carbon performance is a trans-
accounting-based and market-based contexts, this is not the mission channel whereby the presentation of the price to
case with financial value. Interestingly, the DE ratio has a cash flow ratio benefits both financial performance and the
positive and significant relationship with FIV and Tobin’s Q value of the company. In this situation, it is imperative that
(although this relationship is negative and significant in the companies integrate green stocks since this valuation ratio
case of ROA). This suggests the need for corporate managers computes the company stock value with non-cash costs (for
to integrate realistic carbon performance practices that build instance, depreciation). The price to cash flow ratio is vital
shareholder confidence that the company’s financial value for analysts and investors to gain a less distorted perspective
can be sustained, even in cases where it obtains increased of the firm’s financial standing since its cash flows are not
financing by borrowing funds in the short run. This is sup- as controlled and/or exploited as its earnings. As such, it is
ported by the finding that in the long run, the DE ratio of vital for companies to introduce carbon emissions reduc-
the studied JSE listed firms shows a positive and significant tion practices and policies that spur green perceptions and
association with carbon performance for all the regression address risk tolerance among investors.
equations—ROA, FIV, and Tobin’s Q. Moreover, the com- Finally, the findings indicate that a rise in the current ratio
panies should acquire loans with green features, pay them significantly lessens carbon performance in all the financial
timeously, and introduce an efficient green inventory man- performance measures but supports corporate financial value
agement system as this ensures that funds are not wasted. in the short run. In the long run, an upsurge in the current
They should also restructure their debt. Given that market ratio has a significantly positive influence on carbon perfor-
rates for green loans are normally low, such loans would mance for all the corporate financial performance indicators
reduce the overall DE ratio. and financial value. This finding demonstrates the positive
The results of this research demonstrate that the IC ratio impact that the current ratio can have on carbon performance
is positive and significantly associated with carbon per- and financial performance and/or firm value. It is thus vital
formance in Tobin’s Q equation in the short run, although for these companies to sell nongreen assets since such
that connection is negative and significant in the context resources are no longer producing financial gains to the firm.
of the return on assets and firm value regression frame- They should also investigate whether term green loans are
works. This implies that by improving carbon performance suitable for reamortisation and should not acquire non-green
in the context of market-based firm financial performance, capital purchases using cash (for green capital purchases, the
interest cover improves corporate financial performance. company could consider delaying the acquisition).
However, this is not the case for company financial value
and accounting-based proxies. In the long run, however,
by increasing carbon performance, interest cover is able Conclusions
to significantly raise all financial performance indicators
along with the firm financial value. This implies that these This article investigated the impact of carbon performance
companies should continue to consider green risk in their on corporate financial performance and company finan-
projects along with operations. Adopting green instru- cial value among South African CDP listed forms from
ments implies that the company is promoting sustainability 2014 to 2018. The corporate performance proxies were
as well as encouraging the comprehensive deployment of the accounting-based indicator, ROA and market-based
climate-related and/or environmentally friendly projects. measure, and Tobin’s Q. The two-step generalised method
Given the existing green investment climate, if the inter- of moments was used to ascertain this relationship from
est cover ratio captures greening aspects, creditors and a short-run to a long-run horizon. The short-run results
prospective lenders will not consider that high risks are show that carbon performance has a significantly posi-
involved in lending to the company. Other measures the tive link with return on assets, firm value, and Tobin’s Q.

13
28176 Environ Sci Pollut Res (2022) 29:28166–28179

Nevertheless, in the long run, the association between car- performance, but the association was significantly negative
bon performance and both return on assets and firm value in the case of firm financial value in the short run. In the
is negative and significant although the relationship with long run, the price to cash flow ratio had a significantly
Tobin’s Q remains positive and significant. Other impor- positive relationship with all the financial performance
tant findings emerged where carbon performance was used measures as well as firm value. The findings thus provide
as the dependent variable. Return on assets, firm value strong evidence that an increase in the current ratio sig-
and Tobin’s Q showed a significantly positive relationship nificantly reduces carbon performance in all the financial
with carbon performance in both the short- and long-run performance measure frameworks but supports corporate
horizons. The results highlight that the DE ratio is posi- financial value in the short run. In the long run, an upsurge
tive and significantly linked with carbon performance for in the current ratio had a significantly positive influence
the return on assets and Tobin’s Q equations, while that on carbon performance for all the corporate financial per-
association was found to be significantly negative for the formance indicators and financial value.
firm value equation in the short run. In the long-term, It is important to note that this research only considered
the article demonstrated that the DE ratio is significantly large companies listed on the JSE; hence, the findings do
positively related to carbon performance for the return on not reflect small and medium company settings. Future stud-
assets, Tobin’s Q and firm value regression equations. ies could expand the sample by taking into account diverse
This research also found that the interest cover ratio company sizes and comparing results. Future research could
was significantly positive linked with carbon performance also investigate how other facets of sustainability, such as
in the context of Tobin’s Q regression in the short term, social and economic sustainability impact the studied com-
while the link was ascertained to be negative and signifi- panies’ financial performance and firm value.
cant in the return on assets and firm value regression equa-
tions. In the long term, interest cover was positive and
significantly connected with carbon performance for the
ROA, firm value, and Tobin’s Q regressions. Furthermore, Appendix
the PCF ratio was significantly positively associated with
carbon performance for all the proxies for firm financial Tables 8, 9 and 10

Table 8  Findings of panel data for regression 1: ROA

Random effect model Fixed effect model Instrumental variable (IV) model
Coefficient Standard error Coefficient Standard error Coefficient Standard error

CARPER 0.000182 (0.996) 0.0359167 0.0033307 (0.929) 0.0374406 − 0.0133191 (0.754) 0.0424372
DE − 0.6096503 0.1917924 − 0.66335 (0.005)*** 0.2322801 − 2.049495 0.6286435
(0.001)*** (0.001)***
IC 0.0098202 (0.000)*** 0.00274 0.0023813 (0.531) 0.0037983 0.013511 0.0022922
(0.000)***
PCF 0.0635652 (0.011)** 0.0251083 0.0540582 (0.039)** 0.0260547 0.0814705 0.0303428
(0.007)***
CR − 0.0291869 (0.586) 0.0535738 0.1226658 (0.036)** 0.0583669 − 0.335223 0.059512
(0.000)***
Constant 7.001639 (0.000)*** 0.9476775 7.140721 (0.000)*** 0.6157663 9.903208 1.335937
(0.000)***
R2 0.1130 0.0166 0.0275
Wald (χ2) 29.88*** 85.02
F statistic 3.56
Breusch–Pagan test 212.51 (0.000)***
(χ2)
Durbin (score) chi2(1) 6.89518
Wu–Hausman (0.0086) ***
F(1523) = 6.8938
(0.0089)***
No. of observations 534 534 534 534 530 530
***
, **, and * indicate that the coefficients are significant at the 1%, 5%, and 10% level of significance, respectively. Numbers in brackets are
p-values

13
Environ Sci Pollut Res (2022) 29:28166–28179 28177

Table 9  Findings of panel data for regression 2: FIV

Random effect model Fixed effect model Instrumental variable (IV) model
Coefficient Standard error Coefficient Standard error Coefficient Standard error

CARPER 0.0026953 (0.169) 0.0019597 0.002949 (0.135) 0.0019669 − 0.0055463 (0.710) 0.0149363
DE 0.0515344 0.0120311 0.0523853 0.0122026 1.311278 (0.000)*** 0.2212584
(0.000)*** (0.000)***
IC 0.0001012 (0.603) 0.0001943 0.0001083 (0.588) 0.0001995 0.0010096 (0.211) 0.0008068
PCF − 0.0015507 (0.256) 0.001364 − 0.0016444 (0.230) 0.0013688 0.0006606 (0.951) 0.0106795
CR − 0.0040722 (0.181) 0.003046 − 0.0035343 (0.250) 0.0030662 0.0022162 (0.916) 0.0209459
Constant 22.22617 (0.000) 0.1872622 22.2261 (0.000) *** 0.0323486 19.8785 (0.000)*** 0.4701987
***
R2 0.0031 0.0025 4.2707
Wald (χ2) 23.02 (0.0003)*** 37.88
F statistic 4.61
Breusch–Pagan test 941.55
(χ2) (0.0000)***
Durbin (score) 188.365
chi2(1) (0.000) ***
Wu–Hausman F(1523) = 288.365
(0.000) ***
No. of observations 534 534 534 534 530 530
***
; **, and * indicate that the coefficients are significant at the 1%, 5%, and 10% level of significance, respectively. Numbers in brackets are
p-values

Table 10  Findings of panel data for regression 3: Tobin’s Q

Random effect model Fixed effect model Instrumental variable (IV) model
Coefficient Standard error Coefficient Standard error Coefficient Standard error

CARPER − 2.36e-06 (0.999) 0.0030211 0.0000359 (0.991) 0.0030443 − 0.002507 (0.815) 0.0106934
DE − 0.0104002 (0.576) 0.0185916 -0.0103425 (0.584) 0.018887 − 0.0223459 (0.888) 0.1584072
IC − 0.0000177 (0.953) 0.0003011 − 4.31e-06 (0.989) 0.0003088 − 0.0002604 (0.652) 0.0005776
PCF 0.0004226 (0.841) 0.0021027 0.0004386 (0.836) 0.0021185 − 0.0006453 (0.933) 0.0076459
CR 0.0002878 (0.951) 0.004699 0.0003549 (0.940) 0.0047459 − 0.0042418 (0.777) 0.014996
Constant 0.6679097 (0.042) 0.3280199 0.6679123 (0.000) 0.0500687 0.6565196 (0.051) * 0.336633
** ***
R2 0.0001 0.0001 0.0007
Wald (χ2) 0.35*** 0.35
F statistic
Breusch–Pagan test 991.89 (0.000) ***
(χ2)
Durbin (score) 0.004081
chi2(1) (0.9491)
Wu–Hausman F(1523) = 0.004027
(0.9494)
No. of observations 534 534 534 534 530 530
***
, **, and * indicate that the coefficients are significant at the 1%, 5%, and 10% level of significance, respectively. Numbers in brackets are
p-values

13
28178 Environ Sci Pollut Res (2022) 29:28166–28179

Supplementary Information The online version contains supplemen- De Andres P, Vallelado E (2008) Corporate governance in banking: the
tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 11356-0​ 21-1​ 8467-2. role of the board of directors. J Bank Financ 32(12):2570–2580
ESI-Africa (2020) Carbon tax: accelerating the transition to a low-
Author contribution Ganda Fortune: conceptualised the paper; investi- carbon economy. https://​www.​esi-​africa.​com/​indus​try-​secto​rs/​
gated and provided all resources; performed the research methodology; finan​ce-​and-​policy/​carbon-​tax-​accel​erati​ngthe-​trans​ition-​to-a-​
implemented formal data analysis; wrote the full paper; reviewed and low-c​ arbon​ econo​ my/#:~
​ :t​ ext=A ​ n%2​ 0incr​ ease%2​ 0of%2​ 05.6%2​ 5%​
undertook to edit. 20for​,tCO2e%​20when%​20exc​eeding %20the%20limit Accessed
16 October 2021
Fernández-Cuesta C, Castro P, Tascón MT, Castaño FJ (2019) The
Data availability The datasets used and/or analysed during the cur-
effect of environmental performance on financial debt European
rent study are available from the corresponding author on reasonable
evidence. J Clean Prod 207:379390
request.
Ganda F (2021a) The impact of health expenditure on environmental
quality: the case of BRICS. Dev Stud Res 8(1):199–217
Declarations Ganda F (2021b) The influence of carbon performance on the financial
debt of listed companies in an emerging economy: does company
Ethics approval Not applicable. size matter? https://o​ nline​ libra​ ry.w
​ iley.c​ om/d​ oi/a​ bs/1​ 0.1​ 002/b​ sd2.​
182 Accessed 27 December 2021
Consent for publication Not applicable. Ganda F (2019) Carbon emissions, diverse energy usage and economic
growth in South Africa: investigating existence of the environ-
Competing interests The author declares no competing interests. mental Kuznets curve (EKC). Environ Prog Sustain Energy
38(1):30–46
Ganda F, Ngwakwe CC, Ambe C (2015) The role of corporate green
investment practices on sustainable development. Environ Econ
6(1):33–44
Heikal M, Khaddafi M, Ummah A (2014) Influence analysis of
References return on assets (ROA), return on equity (ROE), net profit mar-
gin (NPM), debt to equity ratio (DER), and current ratio (CR),
Abban AR, Hasan MZ (2021) The causality direction between envi- against corporate profit growth in automotive in Indonesia Stock
ronmental performance and financial performance in Australian Exchange. Int J Acad Res Bus Soc Sci 4(12):101
mining companies-a panel data analysis. Res Policy 70:101894 Habimana O (2017) Do flexible exchange rates facilitate external
Agyabeng-Mensah Y, Afum E, Ahenkorah E (2020) Exploring finan- adjustment? A dynamic approach with time-varying and asym-
cial performance and green logistics management practices: metric volatility. Int Econ Policy 14(4):625–642
examining the mediating influences of market, environmental and JSE (2021) Welcome to the FTSE/JSE responsible investment index
social performances. J Clean Prod 258:120613 series. https://​www.​jse.​co.​za/​servic​ es/​indic​es/​ftsej​se-​respo​nsible-​
Alexopoulos I, Kounetas K, Tzelepis D (2018) Environmental and inves​tment-​index-​series Accessed 19 December 2021.
financial performance. Is there a win-win or a win-loss situa- Lahouel BB, Zaied YB, Song Y, Yang GL (2021) Corporate social
tion? Evidence from the Greek manufacturing. J Clean Prod performance and financial performance relationship: a data envel-
197:1275–1283 opment analysis approach without explicit input. Financ Res Lett
Ashraf N, Comyns B, Tariq S, Chaudhry HR (2020) Carbon perfor- 39:101656
mance of firms in developing countries: the role of financial slack, Leepsa NM, Mishra CS (2012) Post-merger financial performance: a
carbon prices and dense network. J Clean Prod 253:119846 study with reference to select manufacturing companies in India.
Boakye DJ, Tingbani I, Ahinful GS, Nsor-Ambala R (2021) The rela- Int Res J Financ Econ 83(83):6–17
tionship between environmental management performance and Lin WL, Cheah JH, Azali M, Ho JA, Yip N (2019) Does firm size mat-
financial performance of firms listed in the alternative investment ter? Evidence on the impact of the green innovation strategy on
market (AIM) in the UK. J Clean Prod 278:24034 corporate financial performance in the automotive sector. J Clean
Busch T, Bassen A, Lewandowski S, Sump F (2020) Corporate carbon Prod 229:974–988
and financial performance revisited. Organ Environ. https://​doi.​ Luo L, Tang Q (2014) Does voluntary carbon disclosure reflect
org/​10.​1177/​10860​26620​935638 underlying carbon performance? J Contemp Account Econ
Carbon Brief (2018) The carbon brief profile: South Africa. https://​ 10(3):191–205
www. ​ c arbo ​ n brief. ​ o rg/ ​ t he- ​ c arbon- ​ b rief- ​ p rofi ​ l e- ​ s outh- ​ a frica Matsumura EM, Prakash R, Vera-Munoz SC (2014) Firm-value
Accessed 18 September 2021 effects of carbon emissions and carbon disclosures. Account Rev
CAT (2020) Country summary–South Africa. https://​c lima​t eact​ 89(2):695–724
iontr​acker.​org/​count​r ies/​south-​africa/ Accessed 18 September Moneva JM, Bonilla-Priego MJ, Ortas E (2020) Corporate social
2021 responsibility and organisational performance in the tourism sec-
CDP (2018) CDP South Africa climate change 2018 executive sum- tor. J Sustain Tour 28(6):853872
mary. https://​www.​nbi.​org.​za/​wp-​conte​nt/​uploa​ds/​2019/​06/​NBI_​ Okafor A, Adusei M, Adeleye BN (2021) Corporate Social Responsi-
CDP-​South​Africa_​Clima​te_​change_​Report_​2018.​pdf Accessed bility and financial performance: evidence from US Tech Firms.
13 November 2021. J Clean Prod 292:126078
Chakravarty S, Hegde P (2019) The joint entrance exam, overconfident Purwanto P, Agustin J (2017) Financial performance towards value
directors and firm performance. J Corp Financ 56:298–318 of firms in basic and chemicals industry. Eur Res Stud J XX
Chung KH, Pruitt SW (1994) A simple approximation of Tobin’s q. 2A:443–460
Financ Manag 23(3):70–74 Sutantoputra AW, Lindorff M, Johnson EP (2012) The relationship
Clarkson PM, Li Y, Richardson GD, Vasvari FP (2008) Revisiting between environmental performance and environmental disclo-
the relation between environmental performance and environ- sure. Australas J Environ Manag 19(1):51–65
mental disclosure: an empirical analysis. Account Organ Soc The South African Department of Environmental Affairs (2014) GHG
33(4–5):303–327 Inventory for South Africa: 2000–2010. https://w ​ ww.e​ nviro​ nment.​

13
Environ Sci Pollut Res (2022) 29:28166–28179 28179

gov.​za/​sites/​defau​lt/​files/​docs/​g reen​house​gas_​invet​oryso​uthaf​ Velte P, Stawinoga M, Lueg R (2020) Carbon performance and disclo-
rica.p df Accessed 01 November 2021 sure: a systematic review of governance-related determinants and
Trinks A, Mulder M, Scholtens B (2020) An efficiency perspective financial consequences. J Clean Prod 254:120063
on carbon emissions and financial performance. Ecol Econ Wang S, Wang H, Wang J, Yang F (2020) Does environmental informa-
175:106632 tion disclosure contribute to improve firm financial performance?
Tzouvanas P, Kizys R, Chatziantoniou I, Sagitova R (2020) Environ- An examination of the underlying mechanism. Sci Tot Environ
mental and financial performance in the European manufacturing 714:136855
sector: an analysis of extreme tail dependency. Br Account Rev Zhang D, Vigne SA (2021) The causal effect on firm performance of
52(6):100863 China’s financing– pollution emission reduction policy: firm-level
Ullah A, Pinglu C, Ullah S, Zaman M, Hashmi SH (2020) The nexus evidence. J Environ Manag 279:111609
between capital structure, firm-specific factors, macroeconomic
factors and financial performance in the textile sector of Pakistan. Publisher’s note Springer Nature remains neutral with regard to
Heliyon 6(8):e04741 jurisdictional claims in published maps and institutional affiliations.

13