The influence of sectoral carbon emission intensity on the quality of carbon emission disclosure

(1)

The influence of sectoral carbon emission intensity on the

quality of carbon emission disclosure

Master’s thesis Erwin van der Veen

S2403390 MSc Controlling

Faculty of Economics and Business University of Groningen Supervisor: Dr. D.A. de Waard

June 22, 2020 Word count: 6752

Abstract

The paper investigates the influence of sectoral carbon emission intensity on the quality of carbon emission disclosure. The sample is derived from the Dutch Transparency benchmark 2019 (Ministerie van Economische Zaken en Klimaat, 2019) and consists of the 130 highest ranked companies. It is furthermore expected that both the entry into force of the EU NFI Directive and the degree of competitiveness have an influence on the hypothesized relationship between sectoral carbon emission intensity and the quality of carbon emission disclosure. A positive and significant relationship between the influence of sectoral carbon emission intensity and the quality of carbon emission disclosure has been found, and thus the main hypothesis has been confirmed. Adding the entry into force of the EU NFI Directive into the regression did not yield significant results, the same goes for the degree of competitiveness. Therefore, it can be concluded that, based on this study, the entry into force of the EU NFI Directive, as well as the degree of competitiveness, do not have a significant influence on the relationship between sectoral carbon emission intensity and the quality of carbon emission disclosure. For further research, other countries and sectors could be investigated. It might be interesting to see if the same results still hold a few years after the entry into force of the EU NFI Directive.

Keywords

Sectoral carbon emission intensity, carbon emission disclosure, competitiveness, EU NFI Directive

(2)

2

1. Introduction

1.1 Introduction to the research topic

In December 2019, the Dutch Supreme Court has rejected the state’s cassation appeal, following a decision of a lower court, regarding the reduction of carbon emissions. This means that, in comparison to 1990, the Dutch state must reduce carbon emissions by 25% in 2020. Based on current calculations, the expected reduction in relation to 1990 would be 23% at most (Seijlhouwer, 2019). Furthermore, the Dutch Supreme Court states that all countries must bear a share of the responsibility, regardless of their role on a global scale (Seijlhouwer, 2019).

The news follows, and is in line with, the findings in UNEP’s annually published Emissions Gap Report, predicting that, even with all the commitments from the Paris Agreement in place, temperatures will go up with about 3.2 °C (The United Nations Environment Programme , 2019). An annual decrease of carbon emissions by at least 7.6% between 2020 and 2030 is needed to realize the Paris Agreement temperature goal of 1.5 °C (The United Nations Environment Programme , 2019).

Despite the global agreements concerning carbon emissions and the growing concerns about the performance of large industrial emitters regarding climate by stakeholders like investors, politicians and the general population (Talbot and Boiral, 2018), not enough action is being taken. In the quest for a more sustainable future, over the years, companies included a lot more non-financial information in their annual reports, or even published separate sustainability reports (Prado-Lorenzo and Garcia-Sanchez, 2010). Next to the concerns regarding climate performance, the reliability of the disclosed information is another point of concern (Talbot and Boiral, 2018). Furthermore, Talbot and Boiral (2018) mention that although the Global Reporting Initiative (GRI) “contributed to the idea that rational, relevant and transparent information is available” (p. 368), climate disclosure studies describe the embellishing of their image by companies, where sustainability reports are often used as a tool to show their commitment to stakeholders, or worse, to misinform them (Domenec, 2012).

In Europe, the inclusion of non-financial information has been accelerated by the EU NFI Directive. From 2017 on, large public-interest companies with over 500 employees are required to add non-financial information to their annual reports, as imposed by the Directive 2014/95/EU (European Commission, 2019).

But, as mentioned before, in the past years, adding non-financial information by companies did not lead to being on course with the temperature goal of 1.5 °C. In this paper,

(4)

4

the influence of sectoral carbon emission intensity on the quality of carbon emission disclosure will be investigated. In 2011, Gamerschlag et al. have looked into the CSR disclosures of 130 listed German companies. Content analysis was applied to distinguish different levels in CSR disclosure. The results show that “companies from “polluting industries” tend to have a higher level of environmental disclosures” (Gamerschlag et al., 2011, p. 257). Based on the article of Gamerschlag et al. (2011), it is expected that the quality of carbon emission disclosure is higher in sectors where the carbon emission intensity is higher. Besides the influence of sectoral carbon emission intensity on the quality of carbon emission disclosure, the role of competitiveness within a sector and the implementation of the EU NFI directive will also be investigated. The research question is therefore formulated as follows:

RQ: Does the degree of sectoral carbon emission intensity affect the quality of carbon emission disclosure and do the sectoral competitiveness and the introduction of the EU NFI Directive have an influence?

1.2 Scientific contribution

With the increase of non-financial disclosure over the years and the growing concerns on climate change, sustainability reporting is not an unexplored research field. Dienes et al. (2016) even conducted a systematic review to find the main sustainability reporting drivers amongst companies, based on existing studies from the years 2000-2015. Dienes et al. (2016) conclude that sustainability activities are only being reported in case of “economic benefit derived from improved reputation, reduced capital costs or alleviated public pressure” (p. 174).

In their study, Tuwaijri et al. (2004) report similar findings, concluding that poor environmental performers disclose less environmental information in comparison with their better performing counterparts. This is in line with the concluding remarks in the Hamrouni et al. (2015) paper. They anticipated that, based on the signaling theory and empirical evidence, “corporate voluntary disclosure may be considered as a signaling tool revealing firm performance.” (Hamrouni et al., 2015, p. 610). They conclude that “the level of voluntary information disclosed in annual reports plays a significant signaling role of firm performance.” (Hamrouni et al., 2015, p. 609).

Since the introduction of the EU NFI Directive, the main sustainability reporting drivers for large public EU companies with over 500 employees, are subordinate, as these companies are required to add non-financial information to their management reports. The EU NFI

(5)

5

Directive also causes the voluntary disclosure environment, for the applicable companies, to disappear (European Commission, 2019). With the implementation of the EU NFI Directive, companies are given flexibility in what they consider most useful regarding the disclosure of information, based on international, national or European guidelines (European Commission, 2019). In their report; “The next steps in non-financial information reporting”, KPMG (2019) states that this, principle based, EU NFI Directive with room for interpretation and the availability of over 100 frameworks on NFI topics is the reason “why companies are still struggling with what to report.” (KPMG, 2019, p. 21). Baumüller & Schaffhauser (2018) even conclude that “the Directive 2014/95/EU shows many similarities to the concept of integrated reporting as set forth by the IIRC” (p. 110). They add to this that in case of proper identification and communication of the EU NFI Directive, “via the means of nonfinancial reports, relying on the application of the principle of materiality”, “might serve the aims of sustainability more” (Baumüller & Schaffhauser, 2018, p. 110).

This paper will give readers information about the quality of carbon emission disclosure and sectoral carbon emission intensity, the role of competitiveness within a sector and the influence of the EU NFI Directive on carbon emission disclosure. The remainder of this paper is divided into the following four sections: the theoretical framework in chapter two, chapter three discusses the research methodology, the results will be displayed in chapter four and it will end with the discussion and conclusion in chapter five.

(6)

6

2. Theoretical framework

Chapter two will give a look into the applicable theories regarding carbon emission disclosure. Next to the applicable theories, the variables as mentioned in the introduction, will also be discussed. Based on the applicable theories and the variables, hypotheses will be developed. In the section below the legitimacy theory, the stakeholder theory and institutional isomorphism will be discussed. Thereafter the variables will be addressed.

2.1 Applicable theories

2.1.1 Legitimacy theory

The first applicable theory is the legitimacy theory. Deegan (2002) mentions the following about the legitimacy theory: “Organisations exist to the extent that the particular society considers that they are legitimate, and if this is the case, the society “confers” upon the organisation “state” of legitimacy” (p. 292). Deegan (2002) found this to be consistent with Mathews (1993) and refers to Mathews (1993) statement regarding legitimacy theory: “The social contract would exist between corporations (usually limited companies) and individual members of society. Society (as a collection of individuals) provides corporations with their legal standing and attributes and the authority to own and use natural resources and to hire employees. Organisations draw on community resources and output both goods and services and waste products to the general environment. The organisation has no inherent rights to these benefits, and in order to allow their existence, society would expect the benefits to exceed the costs to society.” (Mathews, 1993, p. 26)

As mentioned in the introduction, Talbot and Boiral (2018) describe the growing concerns of the general population, and thus the society, regarding carbon emissions. The legitimacy theory plays an important role here because of the existing contract between the organizations and the society. With the Paris Agreement in place, countries bear responsibility to reduce their carbon emissions and the society is getting more aware of the impact organizations have and the need for them to conduct a more sustainable way of doing business.

In the light of the legitimacy theory, Talbot and Boiral (2018) furthermore write that “voluntary disclosure is in fact a response to external pressures” (p. 368) and that these external pressures, rather than “genuine commitment to sustainable development” (p. 368), cause the voluntary disclosure to increase. With the EU NFI Directive, applicable companies are required to include certain non-financial information, possibly leading the amount of voluntary

(7)

7

disclosure to decrease. Too enhance their legitimacy, the inclusion of more and more detailed information is a way for companies to protrude above their competitors.

2.1.2 Stakeholder theory

“Any group or individual who can affect or is affected by the achievement of the firm’s objectives” (p. 25) is defined as a stakeholder by Freeman (1984). For a firm, attaining the ability to balance the conflicting demands of various stakeholders is a major objective (Roberts, 1992). Roberts (1992) also states: “The more critical stakeholder resources are to the continued viability and success of the corporation, the greater the expectation that stakeholder demands will be addressed” (p. 599). This contributes to the idea that the quality of carbon emission disclosure will increase when stakeholders expect better and more explicit information regarding carbon emissions. The misinforming of stakeholders, as referred to in the introduction, may decrease due to the better defined requirements for non-financial information by the EU NFI Directive.

Although there is common ground between legitimacy theory and stakeholder theory, the legitimacy theory looks at the overall society in relationship to the organizational legitimacy where the stakeholder theory explains the forming of management strategies, taking the role of particular stakeholders in consideration (Cotter & Najah, 2012).

2.1.3 Institutional Isomorphism

In their paper on Institutional Isomorphism, Frumkin & Galaskiewicz (2004) conclude that under the influence of external pressures, companies are moving “toward the mean of all organizations regardless of the sector” (p. 303). DiMaggio and Powell (1983) identify three forms of isomorphism: “1) coercive isomorphism that stems from political influence and the problem of legitimacy; 2) mimetic isomorphism resulting from standard responses to uncertainty; and 3) normative isomorphism, associated with professionalization” (p. 150).

Organizations adapting to the EU NFI Directive are dealing with coercive pressures, whereas organizations mimicking successful competitors, are undergoing a form of mimetic isomorphism. Adding to this, Dawkins and Fraas (2011) describe a process wherein pacesetters increase disclosure to gain momentum and “then recoil to a more standard, safe level of disclosure” (p. 316). With the comparison of data between 2016 and 2018, the deviation from the mean of all organizations can be investigated, considering the sectoral carbon emission

(8)

8

intensity, the degree of sectoral competitiveness and the entry into force of the EU NFI Directive.

2.1.4 Reporting materiality

Lai et al. (2017) indicate that materiality made its first appearance “in the context of financial reporting” (p. 3) and mention the general consensus on materiality as follows: “information is material if it influences the decisions of (reasonable) users” (p. 3).

In their research article, Gerwanski et al. (2019) state that without “strong reliance on materiality and “integrated thinking”, the risk of greenwashing and information overload would not be mitigated, and IR might be abused as a “marketing tool” without distinct improvements regarding transparency and decision usefulness” (p. 750). Just as the Integrated Reporting Framework they refer to, the EU NFI Directive is principle based. On their web page on non-financial reporting the European Commission indicates this by stating: “Directive 2014/95/EU gives companies significant flexibility to disclose relevant information in the way they consider most useful. Companies may use international, European or national guidelines to produce their statements” (European Commission, 2019).

Coming back to the Gerwanski et al. (2019) research article on the determinants of materiality disclosure quality in integrated reporting, concluding remarks show that, with this principle based framework, in practice, detailed time horizons information are missing, as well as material risks. Also, learning effects need to be considered by standard setters (Gerwanski et al., 2019). Gerwanski et al. (2019) recommend “the issuance of a “best practice guide” for materiality disclosure, specifically for first‐year appliers” (p. 764).

Through the absence of a concrete framework with strict rules, the question is whether the entry into force of the EU NFI Directive is contributing to the reporting materiality of non-financial information and leads to less greenwashing and an abundancy of information.

2.2 Variables

2.2.1 Independent variable: degree of sectoral carbon emission intensity

As mentioned before, Gamerschlag (2011) found that “companies from “polluting industries” tend to have a higher level of environmental disclosures” (p. 240). Dawkins & Fraas (2011) endorse this finding by stating that companies “with less favorable environmental records use disclosure as a safety net against threats to legitimacy” (p. 315). In general, organizations from polluting industries have less favourable environmental records when compared to

(9)

9

organizations from industries that have a lower carbon emission intensity. This study focuses on a comparison between different sectors.

2.2.2 Dependent variable: quality of carbon emission disclosure

Talbot and Boiral (2018) summarize in their paper; “the literature on climate disclosure highlights the lack of transparency and problems concerning the quality and validity of the information disclosed” (p. 371). In their study on the development of reporting mechanisms for greenhouse gases, Kolk et al. (2008) indicate the importance of the need for stricter carbon emissions disclosure based on clear guidelines “that do not vary per year and require external verification” (p. 742). This improves the comparability significantly. According to Kolk et al. (2008), the “frequent lack of disclosure of types and meaning of emissions data, and of reliability checks” (p. 741) made it difficult to fathom the reported emissions.

Therefore, in this study, the quality of carbon emission disclosure will be reviewed. Other CSR aspects are not taken into account. In this study, the quality of carbon emission disclosure refers only to factual information concerning emissions. The transparency, meaning of emissions of data, the corresponding reliability checks and the overall comparability will, amongst others, be taken into account. Based on the variables mentioned above, the following hypothesis is composed:

Hypothesis 1: The degree of sectoral carbon emission intensity has a positive correlation with the quality of carbon emission disclosure.

2.2.3 Moderator variables: degree of sectoral competitiveness and the EU NFI Directive

As mentioned above, under the influence of external pressures, companies are moving “toward the mean of all organizations regardless of the sector” (Frumkin & Galaskiewicz, 2004, p. 303). It is expected that when the mean quality of sustainability reporting goes up, companies follow this trend. It is interesting to see what difference the competitiveness within the sector makes. When an organization has a monopoly, and thus no competitors, the organization is subjected to fewer external pressures. In the extreme scenario of a monopoly, the quality of carbon emission disclosure of one organization, is the mean quality of the concerned sector. In this case there could be coercive and normative isomorphism, but there will almost be no mimetic isomorphism. The other way around, having a lot of competitors will lead to an increase of external pressures and possibly to mimetic isomorphism, thus the quality of carbon emission

(10)

10

disclosure is expected to go up. A more plausible explanation is that in order to compete properly with competitors, companies pull out all the stops. This, in turn, can lead to increased emissions, in other words, to poorer environmental performance. In the context of legitimacy, a company can then decide to provide better and more detailed information, which may lead to higher quality of carbon emission disclosure. Based on the arguments above, the degree of sectoral competitiveness is expected to have a strengthening effect on the quality of carbon emission disclosure.

With the entry into force of the EU NFI Directive, as mentioned before, large public EU companies with over 500 employees are required to add non-financial information to their management reports. This causes the external pressures to increase, possibly leading to improved quality of sustainability reporting, partly because of the pacesetters that want to gain momentum (Dawkins and Fraas, 2011), and partly because organizations want to enhance their legitimacy. Based on empirical evidence, the mean quality of carbon emission disclosure in sectors where the carbon emission intensity is high, is expected to be higher, compared to sectors where the carbon emission intensity is low. It is expected that the mean quality of carbon emission disclosure in low carbon emissions intensity sectors takes bigger steps than its counterparts in high carbon emissions intensity sectors. Furthermore, it is interesting to see if companies within the sample, that are not subject to the EU NFI Directive, are undergoing a form of isomorphism, and enhance the quality of their carbon emission disclosure significantly, due to external pressures. However, it is premature to say that a shift in quality is due to isomorphism, that cannot be stated on the basis of absolute differences, arising from the research conducted in this study. Based on the moderator variables, the following hypotheses have been composed:

Hypothesis 2: The degree of sectoral competitiveness has a strengthening effect on the quality of carbon emission disclosure.

Hypothesis 3a: The entry into force of the EU NFI Directive has a strengthening effect on the quality of carbon emission disclosure of companies subject to the EU NFI Directive.

Hypothesis 3b: The entry into force of the EU NFI Directive has a strengthening effect on the quality of carbon emission disclosure of companies not subject to the EU NFI Directive.

(11)

11

(12)

12

3. Methodology

For this study, the sample consists of the 130 highest ranked companies out of the Dutch Transparency benchmark 2019 (Ministerie van Economische Zaken en Klimaat, 2019). This benchmark ranks the 500 biggest organizations in the Netherlands, based on their transparency regarding sustainability reporting (Ministerie van Economische Zaken en Klimaat, 2019). Out of the 130 companies in the sample, 15 companies are subject to the EU NFI Directive. For the 115 companies that don’t have to comply to the EU NFI Directive, it is interesting to see whether the entry into force of the EU NFI Directive affects their quality of carbon emission disclosure.

The main reason to focus on 130 out of the 500 ranked companies, is the relative score achieved by these companies. From position 130 on, the relative score drops below 30%, contributing to the idea that little can be said about the level of transparency regarding sustainability reporting of these companies. The Transparency benchmark uses a framework specially developed to cover a broad spectrum of sustainability. Because of the many aspects covered, it is difficult to go into depth with this framework.

For this study, certain aspects from the Tauringana & Chithambo (2015) disclosure index will be used, to create a more detailed benchmark to measure the quality of carbon emission disclosure. The benchmark will be included as Appendix 1.

The data will come from annual reports or sustainability reports from the years 2016 and 2018, to measure any differences in quality of the non-financial information of before and after the implementation of the EU NFI Directive. It is furthermore worth mentioning that the Transparency benchmark, as mentioned above, is only conducted once every two years. Therefore, no benchmark has been performed on annual reports from 2017.

3.1 Variables

3.1.1 Independent variable: degree of sectoral carbon emission intensity

In the Dutch Transparency benchmark 2019, the companies are classified per sector. Participating companies had to assign themselves in 1 of the 15 designated sectors (Ministerie van Economische Zaken en Klimaat, 2019). For each of the 130 companies in the sample, the corresponding sector based on the classification of Statistics Netherlands will be used. Statistics Netherlands provides national emission figures on a yearly basis (Statistics Netherlands, 2020).

(13)

13

3.1.2 Dependent variable: quality of carbon emission disclosure

To measure the quality of carbon emission disclosure, an adjusted benchmark, based on the Tauringana & Chithambo (2015) disclosure index, will be used. The benchmark will be included as Appendix 1.

The benchmark consists of 30 items, for each item included in the annual- or sustainability reports 1 point will be assigned to the organizations. In total, 30 points can be achieved. Items that are missing out, score 0 points. The disclosure items, as used by Tauringana & Chithambo (2015), were used if they were directly related to carbon emissions. In addition some adjustments have been made.

3.1.3 Moderator variables: degree of sectoral competitiveness and the EU NFI Directive

To determine the degree of competitiveness, the HHI index of each of the companies in the sample has been calculated. The information on competitors is derived from Company.info (Company.info, 2020). Company.info uses the Dutch Chamber of Commerce as a source. This provides more accurate information about the market in which a company operates and its competitors than the information provided by the Dutch Transparency benchmark (Ministerie van Economische Zaken en Klimaat, 2019), which distinguishes between 15 sectors only. The revenues of relevant competitors in the corresponding sector have been added up to get the total revenue of the corresponding market in the Netherlands. Dividing the revenue per company by the total revenue of the market results in the market shares for each individual company within the market. Squaring and then summing up the individual market shares will give the HHI index of the corresponding market. Ertl & McCarrell (2002) use the following ranges to classify market structures based on the corresponding HHI index: <0.2 = perfect competition, between 0.2 and 0.4 = monopolistic competition, between 0.4 and 0.7 = an oligopoly and from 0.7 to 1 is described as a monopoly.

To see the effect of the EU NFI Directive, carbon emission disclosure scores, based on the benchmark included (see Appendix 1), of before and after the entry into force will be compared.

(14)

14

3.1.4 Control variables

The next control variables will be included: firm size, firm age and profitability. To control for firm size, the logarithm of total assets will be used. In their paper, Karim et al. (2013) conclude that “disclosure is positively linked to firm size” (p. 866). The companies in the sample are all relatively large. However, there are still considerable differences in size, based on the amount of total assets. Based on the conclusion mentioned above, it is likely for bigger firms to have better quality of carbon emission disclosure. Therefore, firm size is implemented as a control variable.

The second control variable included is firm age. Shrivastava & Tamvada (2019) conclude “that while both external and internal greening strategies have an impact on firm performance for young firms and small firms, internal greening strategies are more important for middle-aged firms and large firms” (p. 962). The youngest firm in the sample is 3 years old, while the oldest firm is 406 years old. For the younger firms it could be beneficial to focus on the external part, such as reporting. Therefore the control variable firm age is included.

The third and last control variable is profitability. Tuwaijri et al. (2004) found that “good environmental performance and economic profitability go hand-in-hand” (p. 467) and that they are “both related to the quality of management” (p. 467). Firms that are more profitable are likely to have better management, which can also lead to higher disclosure quality. For profitability, the profit after tax will be divided by the total amount of assets.

(15)

15

3.2 Analysis and regression model

CO2 SC = β0 + β1 CO2Sector + β2 Competitiveness + β3 EUNFI + β4 LOGTA + β5 LOGFA

+ β6 Profitability + ε

CO2 SC = quality of carbon emission disclosure in 2018

β1 CO2Sector = sectoral carbon emissions in 2018

β2 Competitiveness = degree of competitiveness (HHI index)

β3 EUNFI = difference in quality of carbon emission disclosure, before and after the entry into force of the EU NFI Directive

β4 LOGTA = logarithm of total assets

β5 LOGFA = logarithm of firm age

β6 Profitability = profitability (profit after tax divided by total assets)

ε = residual error

(16)

16

4. Results

4.1 Descriptive statistics

To get an overview of the sample, derived from the Dutch Transparency benchmark 2019 (Ministerie van Economische Zaken en Klimaat, 2019), table 1 is included in the appendix. This table gives an overview of the companies within the sample and the corresponding sector classification. The sample consists of 130 companies, divided into 26 sectors (classification Statistics Netherlands). The most polluting sector present in the sample is the energy supply sector, with an emission of 46,942 million kg CO2. Seven companies in the sample are part of this sector. The most represented sector is financial services, with 31 representatives. The emission of this sector amounts to 421 million kg CO2.

In Table 2, the mean benchmark scores for 2016 and 2018 are displayed. A distinction is made between companies that do and do not have to comply with the EU NFI Directive. Only 15 out of the 130 companies within the sample have to comply with the EU NFI Directive. Although the EU NFI Directive has only been in force since 2017, this distinction was also made for the year 2016. This was partly due to the fact that companies may have been preparing for the EU NFI Directive.

In both years, the mean score for companies that have to comply to the EU NFI Directive is higher. The highest score in both years was achieved by 1 organization and amounted 26. In both years some organizations scored 0 points.

Table 2 Mean scores 2016 vs. 2018

EU NFI 2016 2018 No Mean 7.55 9.18 N 115 115 Yes Mean 8.93 11.80 N 15 15 Total Mean 7.71 9.48 N 130 130

(17)

17 Table 3 HHI index

Mean 0.337107 Minimum 0.035961 Maximum 1.000000

To measure competitiveness, the HHI index has been used. Table 3 shows the minimum and the maximum HHI index. The market structures vary from perfect competition to a monopoly.

4.2 Regression analysis

4.2.1 Correlation

To see if there is any multicollinearity between variables, a correlation analysis has been conducted. The table below (table 4) shows the degree of correlation between the variables.

Table 4 Correlations

Mean

Std.

Deviation 1 2 3 4 5 6 7

CO2 SC (1) 9,485 6,246 -

CO2sector (2) 4,91E+09 1,08E+10 ,206* - Competitiveness (3) 0,337 0,261 0,054 ,253 ** _- EUNFI (4) 0,115 0,321 0,134 0,128 -0,062 - LOGTA (5) 9,495 0,958 ,377** 0,035 0,149 ,189* - LOGFA (6) 1,661 0,425 0,046 -0,081 0,101 0,061 0,019 - Profitability (7) 0,033 0,081 0,155 -,254** -0,080 -0,015 0,040 0,069 -

*. Correlation is significant at the 0.05 level (2-tailed). **. Correlation is significant at the 0.01 level (2-tailed).

Table 4 shows weak relationships between a number of variables used. The highest degree of correlation in this model exists between the dependent variable CO2 SC (the quality of carbon emission disclosure in 2018) and the control variable LOGTA (logarithm of total assets). The correlation between these variables is 0,377 and is significant at the 0.01 level (2-tailed). Other weak relationships exist between CO2Sector (2) and Competitiveness (3) (significant at the 0.01 level (2-tailed)), between CO2sector (2) and Profitability (7), also significant at the 0.01

(18)

18

level (2-tailed). Significant at the 0.05 level (2-tailed) is the weak relationship between EUNFI (4) and LOGTA (5).

4.2.2. Regression analysis

The following table (table 5) shows the results from the regression analysis of the independent variable, sectoral carbon emissions in 2018 (CO2sector), on the quality of carbon emission disclosure in 2018 (dependent variable).

Table 5 Regression independent variable

Model B Std. Error t Sig. R 2 Adjusted R2 Sig. F Change 1 (Constant) -14,580 5,415 -2,692 0,008 0,163 0,143 0,000 LOGTA 2,422 0,532 4,551 0,000 LOGFA 0,434 1,200 0,361 0,718 Profitability 10,637 6,304 1,687 0,094 2 (Constant) -15,120 5,250 -2,880 0,005 0,220 0,195 0,003 LOGTA 2,347 0,516 4,547 0,000 LOGFA 0,667 1,166 0,572 0,568 Profitability 15,452 6,312 2,448 0,016 CO2sector 1,430E-10 0,000 3,031 0,003

As can be seen in table 5, the coefficient is positive and significant (0.01, 2-tailed), this means there is a positive relationship between sectoral carbon emission intensity and the quality of carbon emission disclosure. The first hypothesis can be accepted.

To see if there is a strengthening effect of competitiveness on the relationship between sectoral carbon emission intensity and the quality of carbon emission disclosure, a regression of the moderator variable Competitiveness has been conducted. The outcomes are presented in table 6 on the next page. When adding the interaction term (model 3), the adjusted R2 decreases. The interaction term is not significant at all, meaning that there is no interaction effect of competitiveness on the relationship between sectoral carbon emission intensity and the quality of carbon emission disclosure. The lack of this enhancing effect means that hypothesis 2 can be rejected.

(19)

19

Table 6 Regression moderator variable Competitiveness

Model B

Std.

Error Beta t Sig. R

2 Adjusted R2 Sig. F Change 1 (Constant) -14,580 5,415 -2,692 0,008 0,163 0,143 0,000 LOGTA 2,422 0,532 0,371 4,551 0,000 LOGFA 0,434 1,200 0,030 0,361 0,718 Profitability 10,637 6,304 0,138 1,687 0,094 2 (Constant) -15,354 5,273 -2,912 0,004 0,223 0,192 0,010 LOGTA 2,399 0,523 0,368 4,588 0,000 LOGFA 0,768 1,177 0,052 0,652 0,516 Profitability 15,317 6,328 0,199 2,420 0,017 CO2sector 1,512E-10 0,000 0,263 3,100 0,002 HHI index -1,365 1,997 -0,057 -0,683 0,496 3 (Constant) -15,353 5,295 -2,900 0,004 0,223 0,185 0,992 LOGTA 2,398 0,525 0,368 4,568 0,000 LOGFA 0,766 1,190 0,052 0,644 0,521 Profitability 15,310 6,397 0,199 2,393 0,018 CO2sector 1,520E-10 0,000 0,264 1,568 0,119 HHI index -1,355 2,183 -0,057 -0,621 0,536 CO2sector x HHI index -1,716E-12 0,000 -0,002 -0,011 0,992

The regression outcomes of the moderator variable EU NFI have been displayed in table 7. The interaction term is not significant, but in model 2, as well as in model 3, the EU NFI variable in itself is significant. This means that the entry into force of the EU NFI Directive has a direct and positive effect on the quality of carbon emission disclosure. Nevertheless, there is no enhancing effect of the entry into force of the EU NFI Directive on the relationship between sectoral carbon emission intensity and the quality of carbon emission disclosure. In the regression associated with table 7, no distinction has yet been made between companies that are subject to the EU NFI Directive and companies that are not subject to the EU NFI Directive. Table 8 and 9 are included in the appendix. Table 8 shows the regression outputs of companies

(20)

20

that are subject to the EU NFI Directive, the regression associated with table 9 only takes the companies that to not subject to the EU NFI Directive in consideration.

Table 7 Regression moderator variable EU NFI

Model B

Std.

Error Beta t Sig. R

2 Adjusted R2 Sig. F Change 1 (Constant) -14,580 5,415 -2,692 0,008 0,163 0,143 0,000 LOGTA 2,422 0,532 0,371 4,551 0,000 LOGFA 0,434 1,200 0,030 0,361 0,718 Profitability 10,637 6,304 0,138 1,687 0,094 2 (Constant) -12,663 5,042 -2,511 0,013 0,299 0,270 0,000 LOGTA 2,066 0,497 0,317 4,157 0,000 LOGFA 0,458 1,111 0,031 0,412 0,681 Profitability 12,831 6,051 0,166 2,121 0,036 CO2sector 1,099E-10 0,000 0,191 2,401 0,018 EUNFI 0,451 0,121 0,290 3,727 0,000 3 (Constant) -12,853 5,069 -2,536 0,012 0,300 0,266 0,589 LOGTA 2,078 0,499 0,319 4,165 0,000 LOGFA 0,538 1,124 0,037 0,479 0,633 Profitability 13,355 6,145 0,173 2,173 0,032 CO2sector 8,683E-11 0,000 0,151 1,387 0,168 EUNFI 0,418 0,136 0,269 3,073 0,003 CO2sector x EUNFI 6,119E-12 0,000 0,065 0,542 0,589

It is interesting to see that the EU NFI variable is significant for the regression conducted on the sample containing only companies that do not subject to the EU NFI Directive (table 9). There is still no moderating effect of the EU NFI Directive on the relationship between sectoral carbon emission intensity and the quality of carbon emission disclosure. For the sample containing only companies that do subject to the EU NFI Directive (table 8), none of the variables is statistically significant.

(21)

21

5. Discussion and conclusion

5.1 Findings and conclusions

5.1.1 Findings and conclusions

In this chapter the findings presented in the previous chapter will be reflected to ultimately give an answer to the proposed research question. Starting with hypothesis 1. As can be seen in table 5, adding the independent variable CO2sector to the three control variables (logarithm of total assets, logarithm of firm age and profitability) increases both R2 and adjusted R2 significantly. There is a positive and significant correlation between the independent variable and the dependent variable (0.01, 2-tailed). Therefore, hypothesis 1 can be accepted. The degree of sectoral carbon emission intensity has a positive correlation with the quality of carbon emission disclosure.

Moving on to the first hypothesis regarding moderator variables, hypothesis 2. Testing for a moderating effect of competitiveness on the relationship between the described positive relation between the independent variable and the dependent variable above did not give significant results. Adding the interaction term to the equation does not increase R2 and even decreases the adjusted R2. Hypothesis 2 can be rejected. The degree of sectoral competitiveness does not have a strengthening effect on the quality of carbon emission disclosure.

Hypothesis is split into hypothesis 3a and 3b. For hypothesis 3a, only the companies subject to the EU NFI Directive are included in the regression. As can be seen in table 8, no significant results were obtained. Hypothesis 3a can be rejected. For hypothesis 3b, the companies not subject to the EU NFI Directive are included in the regression. Adding the interaction term does not yield significant results. It is interesting to see that the EU NFI variable has a positive and significant effect on the quality of carbon emission disclosure. Nevertheless, there is no interaction effect, the EU NFI Directive does not have a strengthening effect on the quality of carbon emission disclosure of companies not subject to the EU NFI Directive. Hypothesis 3b can also be rejected.

Based on the reflection of the hypotheses, the research question - Does the degree of

sectoral carbon emission intensity affect the quality of carbon emission disclosure and do the sectoral competitiveness and the introduction of the EU NFI Directive have an influence?- can

be answered. There is a positive and significant correlation between the degree of sectoral carbon emission intensity and the quality of carbon emission disclosure. The degree of sectoral carbon emission intensity does affect the quality of carbon emission disclosure. Regarding the

(22)

22

second part of the research question, sectoral competitiveness and the introduction of the EU NFI Directive do not have an influence and the quality of carbon emission disclosure.

The conclusions drawn are partly in line with the theories mentioned earlier in chapter 2. According to Gamerschlag (2011) “companies from “polluting industries” tend to have a higher level of environmental disclosures” (p. 257). This is supported by the findings in this study where there is a positive and significant correlation between the degree of sectoral emission intensity and the quality of carbon emission disclosure. The link can be made with the legitimacy theory and the stakeholder theory, because of the existing contract between the organizations and society. The external pressures are higher in sectors where the carbon emission intensity is high, companies could anticipate by increasing their quality of non-financial information. Based on this study, competitiveness does not have an influence on the relationship between sectoral carbon emission intensity and the quality of carbon emission disclosure. A reason for the lack of influence could be that the coercive pressures outweigh the pressures to mimic successful competitors. Political influence and legitimacy problems in itself could result in companies improving the quality of their carbon emission disclosure. The EU NFI Directive has no effect on the relationship between sectoral carbon emission intensity and the quality of carbon emission disclosure, but has a positive and significant direct effect on the quality of carbon emission disclosure of companies not subject to the EU NFI Directive. In this case an explanation could be that companies subject to the EU NFI Directive had already prepared their carbon emission disclosure before the entry into force of the EU NFI Directive. Mimetic isomorphism could then be a possible explanation for the influence of the EU NFI Directive on the quality of carbon emission disclosure of companies not subject to the EU NFI Directive, but as mentioned before, this cannot be concluded based on the data that is used to conduct this study.

5.2 Limitations and further research

5.2.1 Limitations

As with most studies, there are some limitations. First of all, the relatively small sample is limited to 130 companies based in the Netherlands. Secondly, the selected companies form the top 130 of the Dutch Transparency benchmark 2019. To some extent, the transparency of these companies has already been tested.

(23)

23

Next to the aforementioned limitations, this study used the HHI index to measure the degree of competitiveness. The fact that companies are in the same sector does not necessarily mean that they are actually competitors of each other. The markets they serve may differ.

To see whether the EU NFI Directive affected the relationship between the independent variable and the dependent variable, the quality of carbon emission disclosure in 2016 and 2018 has been measured and compared. However, the EU NFI Directive has no strict requirements. Measuring the actual quality therefore remains relatively subjective. Hence it is difficult to measure the actual influence of the EU NFI Directive. Another point of concern is that companies may have already prepared for the entry into force of the EU NFI Directive.

5.2.2 Further research

For future research, it can be interesting to see if the same results still hold when conducting research on carbon emission disclosure a few years after the implementation of the EU NFI Directive. There may be more consensus about the determinants of the quality of carbon emission disclosure. Other opportunities lie in the investigation of other sectors and countries.

(24)

24

6. References

Baumüller, J., & Schaffhauser-Linzatti, M. (2018). In search of materiality for nonfinancial information—reporting requirements of the Directive 2014/95/EU.

Nachhaltigkeitsmanagementforum | Sustainability Management Forum, 26(1-4),

101-111.

Company.info: Marktleider bedrijfsinformatie en bedrijfsnieuws. (2020). Retrieved May 11, 2020, from https://companyinfo.nl/

Cotter, J., & Najah, M. M. (2012). Institutional investor influence on global climate change disclosure practices. Australian Journal of Management, 37, 169-187.

Dawkins, C., & Fraas, J. (2011). Coming clean: The impact of environmental performance and visibility on corporate climate change disclosure. Journal of Business Ethics, 100(2), 303-322.

Deegan, C. (2002). Introduction: the legitimising effect of social and environmental

disclosures–a theoretical foundation. Accounting, Auditing & Accountability Journal,

15(3), 282-311.

Dienes D, Sassen R, Fischer J (2016) What are the drivers of sustainability reporting? A systematic review. Sustain Account Manag Policy J 7(2):154–189.

DiMaggio, P., & Powell, W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48(2), 147-160.

Domenec, F. (2012). The ‘‘greening’’ of the annual letters published by Exxon, Chevron and BP between 2003 and 2009. Journal of Communication Management, 16(3), 296–311. Ertl, H., & McCarrell, H. (2002). The state of telecommunications services

European Commission. (2019). Non-financial reporting. Retrieved from European Commission.

Freeman, IL, Strategic Management.. a Stakeholder Approach (Marshall, MA Pitman, 1984) Frumkin, P., & Galaskiewicz, J. (2004). Institutional isomorphism and public sector

organizations. Journal of Public Administration Research and Theory: J-Part, 14(3), 283-307.

Gamerschlag, R., Möller, K., & Verbeeten, F. (2011). Determinants of voluntary CSR disclosure: Empirical evidence from Germany. Review of Managerial Science, 5(2), 233-262.

(25)

25

Gerwanski, J., Kordsachia, O., & Velte, P. (2019). Determinants of materiality disclosure quality in integrated reporting: Empirical evidence from an international setting.

Business Strategy and the Environment, 28(5), 750-770.

Hamrouni, A., Miloudi, A., & Benkraiem, R. (2015). Signaling firm performance through corporate voluntary disclosure. Journal of Applied Business Research, 31(2), 609-620. Karim, K. E., Pinsker, R., & Robin, A. (2013). Firm size and the voluntary disclosure of

nonfinancial information by private versus public firm managers. Managerial Auditing

Journal, 28(9), 866–892

Kolk, A., Levy, D., & Pinkse, J. (2008). Corporate responses in an emerging climate regime: The institutionalization and commensuration of carbon disclosure. European

Accounting Review, 17(4), 719-745.

KPMG. (2019). The next steps in non-financial information reporting. KPMG.

Lai, A., Melloni, G., & Stacchezzini, R. (2017). What does materiality mean to integrated reporting preparers? An empirical exploration. Meditari Accountancy Research. Mathews, M. R. (Martin Reginald) (1993). Socially responsible accounting (1st ed).

Chapman & Hall, London ; New York.

Ministerie van Economische Zaken en Klimaat. (2019). Transparantiebenchmark 2019. 's-Gravenhage.

Prado-Lorenzo, J. M., & Garcia-Sanchez, I. M. (2010). The role of the board of directors in disseminating relevant information on greenhouse gases. Journal of Business Ethics,

97(3), 391–424.

Roberts, R. (1992). Determinants of corporate social responsibility disclosure: An application of stakeholder theory. Accounting, Organizations and Society, 17(6), 595-612.

Seijlhouwer, M. (2019, December 20). Hoge Raad: Staat moet uitstoot terugdringen voor het

einde van 2020. Retrieved from DuurzaamBedrijfsleven:

https://www.duurzaambedrijfsleven.nl/energietransitie-business/33017/klimaatuitspraak-urgenda-uitstoot

Shrivastava, M., & Tamvada, J. P. (2019). Which green matters for whom? Greening and firm performance across age and size distribution of firms. Small Business Economics, 52(4), 951-968.

Statistics Netherlands. (2020). Emissies naar lucht door de Nederlandse economie; nationale

rekeningen. Retrieved from StatLine:

(26)

26

Talbot, D., & Boiral, O. (2018). GHG Reporting and Impression Management: An

Assessment of Sustainability Reports from the Energy Sector. Journal of Business

Ethics, 147(2), 367-383.

Tauringana, V., & Chithambo, L. (2015). The effect of DEFRA guidance on greenhouse gas disclosure. The British Accounting Review, 47(4), 425-444.

Al-Tuwaijri, S., Christensen, T., & Hughes, K. (2004). The relations among environmental disclosure, environmental performance, and economic performance: A simultaneous equations approach. Accounting, Organizations and Society, 29(5), 447-471

The United Nations Environment Programme . (2019, November 26). Cut global emissions by

7.6 percent every year for next decade to meet 1.5°C Paris target - UN report. Retrieved

from UN Environment: https://www.unenvironment.org/news-and-stories/press-release/cut-global-emissions-76-percent-every-year-next-decade-meet-15degc

(27)

27

Appendix

Appendix 1

Benchmark Carbon Emission Disclosure Score

1. Base year. 1/0

2. Explanation for a change in base year. 1/0

3. GHGs covered, including those not required by the Kyoto Protocol. 1/0 4. Conversion factors used/methodology used to measure or calculate emissions. 1/0 5. Explanation for any changes to methodology or conversion factors previously used. 1/0 6. Explanation of any country excluded, if global total is reported. 1/0 7. Explanations for changes in performance of scope 1 emission. 1/0 8. Details of any specific exclusion of emissions from scope 1. 1/0 9. Explanation for the reason for any exclusion from scope 1. 1/0

10. Scope 1 emissions. 1/0

11. Comparative data on scope 1 emissions. 1/0

12. Future estimates of scope 1 emissions. 1/0

13. Explanations for changes in performance of scope 2 emissions. 1/0 14. Details of any specific exclusion of emissions from scope 2. 1/0 15. Explanation for the reason for any exclusion from scope 2. 1/0

19. Electricity consumption. 1/0

20. Explanations for changes in performance of scope 3 emissions. 1/0

24. Explanations for changes in performance of total GHG emissions in CO2 metric tonnes. 1/0

25. Total GHG emissions in CO2 metric tonnes. 1/0

26. Comparative data on total GHG emissions in CO2 metric tonnes. 1/0

27. GHG emission targets set and achieved. 1/0

28. Comparative information on targets set and achieved. 1/0

29. CO2 emissions per location/business unit. 1/0

(28)

28 Table 1 Sector Carbon emission Company Electrical engineering industry 37000000 Koninklijke Philips N.V. KENDRION TKH GROUP VEON Ltd. STMicroelectronics N.V. ASML Cisco Systems BE Semiconductor

Neways Electronics International N.V.

Mail and couriers 142000000 PostNL

Electrical appliances industry 171000000 Signify NV Information and communication 186000000 KPN ORDINA Wolters Kluwer N.V. ALTICE Transport equipment industry 243000000

Fiat Chrysler Automobiles N.V. Damen Shipyards Group N.V.

Accell Group

Machinery industry 270000000

SBM Offshore ASM INTERNATIONAL

Batenburg Techniek N.V.

Rental and trade of real

estate 310000000 Unibail Rodamco WDP VASTNED NSI N.V. WERELDHAVE EUROCOMMERCIAL Financial services 421000000 ABN AMRO

Ernst & Young Nederland LLP Van Lanschot Kempen Nederlandse Waterschapsbank N.V.

FMO: Ned. Financierings-Mij. voor Ontwikkelingslanden N.V.

Coöperatieve Rabobank U.A. ASR Nederland N.V.

Onderlinge Waarborgmij. Centrale Zorgverzekeraars groep, Coöperatief Deloitte U.A.

(29)

29

Coöperatie VGZ U.A. De Nederlandsche Bank (DNB)

PricewaterhouseCoopers

Noordelijke Ontwikkelings Maatschappij (NOM) Triodos Bank N.V.

Brabantse Ontwikkelings Maatschappij (BOM) Achmea B.V.

Coöperatie KPMG U.A. BNG Bank ING Groep

Industriebank Limburgs Instituut voor Ontwik. en Finan. (LIOF)

Ontwikkelingsmaatschappij OOST Nederland (OOST) NN GROUP

Robeco Institutional Asset Management B.V. Aegon N.V.

PGGM Atradius N.V. NIBC HOLDING

Menzis N.V.

Regionale Ontwikkelingsmaatschappij InnovationQuarter B.V.

de Volksbank N.V.

Waste landfills 431000000 Covra: Centrale Organisatie Voor Radioactief Afval Renewi Support B.V.

Other services 489000000

Energie Beheer Nederland (EBN) Stedin Holding N.V. Dela Coöperatie U.A.

Juva Culture, sports and

recreation 549000000

Holding Nationale Goede Doelen Loterijen N.V. Nederlandse Loterij B.V.

Education 658000000 Wageningen University & Research Rijksuniversiteit Groningen Vrije Universiteit Universiteit Twente Specialist business services 753000000 FUGRO ARCADIS HaskoningDHV Nederland B.V. VVAA Groep B.V. Storage, transport services 938000000

Royal Schiphol Group N.V. Havenbedrijf Rotterdam (HbR)

(30)

30

ProRail Health and welfare

services 1504000000 Universitair Medisch Centrum Utrecht Rental and other

business services 2436000000

Randstad Global

BRUNEL INTERNATIONAL

Construction industry 3323000000

Heijmans Royal BAM Group

VolkerWessels

ROYAL BOSKALIS WESTMINSTER N.V TBI Holdings

Van Oord N.V.

Trade 4027000000

Coop Nederland U.A. PLUS Holding B.V.

GRANDVISION

A.S. Watson Health & Beauty Benelux SLIGRO FOOD GROUP

HEMA B.V.

Coöperatie Royal FloraHolland U.A. Koninklijke Ahold Delhaize N.V.

Jumbo Groep Holding B.V. Stern Groep N.V. Zeeman Groep B.V. Ingka Holding B.V.

Louis Dreyfus Company B.V. (v.h. Louis Dreyfus Commodities B.V.)

Beter Bed Holding N.V.

Food, stimulants

industry 4438000000

Heineken N.V. Unilever N.V.

Swinkels Family Brewers Holding N.V. Koninklijke FrieslandCampina N.V.

WESSANEN VanDrie Group

Industrie- en Handelsonderneming Vreugdenhil B.V. Vion N.V.

Van Beek Group BV FORFARMERS Koninklijke Agrifirm Group Land transport 5394000000 Nederlandse Spoorwegen

Rotterdamse Electrische Tram N.V., afgekort R.E.T. N.V. Basic metal industry 6929000000 Tata Steel IJmuiden B.V.

Water companies and

(31)

31

Petroleum industry 10127000000 Royal Dutch Shell Transport by air 12698000000 AIR FRANCE -KLM

Chemical industry 19460000000 Koninklijke Vopak N.V. DSM N.V. CORBION AKZO Nobel N.V. OCI Energy supply 46942000000 Alliander N.V. Enexis Holding N.V. TenneT Holding B.V. N.V. Nederlandse Gasunie GasTerra Eneco Holding N.V. Vattenfall N.V.

(32)

32

Table 8

Table 8 Regression moderator variable (companies subject to) EU NFI

Model B

Std.

Error Beta t Sig. R

2 Adjusted R2 Sig. F Change 1 (Constant) -6,785 25,169 -0,270 0,792 0,051 -0,208 0,198 LOGTA 1,593 2,109 0,275 0,756 0,466 LOGFA 0,812 5,203 0,046 0,156 0,879 Profitability 42,382 76,073 0,204 0,557 0,589 2 (Constant) -0,290 24,927 -0,012 0,991 0,376 0,030 0,151 LOGTA 0,562 2,080 0,097 0,270 0,793 LOGFA 2,579 4,914 0,147 0,525 0,612 Profitability -39,084 92,477 -0,188 -0,423 0,682 CO2sector 1,958E-10 0,000 0,553 1,984 0,079 EUNFI 0,508 0,478 0,378 1,061 0,316 3 (Constant) 1,829 26,955 0,068 0,948 0,385 -0,075 0,739 LOGTA 0,239 2,381 0,041 0,100 0,922 LOGFA 3,116 5,403 0,177 0,577 0,580 Profitability -60,828 116,006 -0,292 -0,524 0,614 CO2sector 2,704E-10 0,000 0,764 1,127 0,293 EUNFI 0,710 0,774 0,528 0,918 0,386 CO2sector x EUNFI -1,688E-11 0,000 -0,243 -0,345 0,739

(33)

33

Table 9

Table 9 Regression moderator variable (companies NOT subject to) EU NFI

Model B

Std.

Error Beta t Sig. R

2 Adjusted R2 Sig. F Change 1 (Constant) -15,494 5,835 -2,655 0,009 0,170 0,147 0,000 LOGTA 2,518 0,579 0,377 4,345 0,000 LOGFA 0,362 1,245 0,025 0,291 0,772 Profitability 10,044 6,366 0,137 1,578 0,117 2 (Constant) -12,971 5,509 -2,355 0,020 0,289 0,256 0,000 LOGTA 2,134 0,549 0,319 3,884 0,000 LOGFA 0,301 1,165 0,021 0,259 0,796 Profitability 11,941 6,280 0,163 1,902 0,060 CO2sector 8,195E-11 0,000 0,130 1,490 0,139 EUNFI 0,474 0,133 0,301 3,576 0,001 3 (Constant) -13,150 5,537 -2,375 0,019 0,291 0,251 0,587 LOGTA 2,139 0,551 0,320 3,881 0,000 LOGFA 0,408 1,185 0,028 0,344 0,731 Profitability 12,477 6,376 0,170 1,957 0,053 CO2sector 5,643E-11 0,000 0,089 0,780 0,437 EUNFI 0,440 0,147 0,280 2,995 0,003 CO2sector x EUNFI 6,675E-12 0,000 0,069 0,545 0,587

The influence of sectoral carbon emission intensity on the quality of carbon emission disclosure