Assessing the Sustainability Reporting Transparency and Engagement of European Energy Companies

Abstract

Energy companies are facing increasing pressure from institutional and industry stakeholders to prioritize their responsibility to the environment and society, including providing accurate, reliable, and comprehensive reports on their sustainability practices. Three metrics were developed in this study: the average sustainability reporting score and two sustainability performance reporting indices based on two different performance measurement methodologies. These were designed to assess the effect of mandatory non-financial disclosure on sustainability reporting and the level of transparency and engagement of energy companies. The study also examined the relationship between the level of sustainability reporting and sustainability performance in the period of 2016–2019 by correlating these metrics. The analysis sheds light on the effectiveness of non-financial disclosure regulations in promoting sustainability practices in the energy industry. The results revealed no difference in metric scores prior to, or even following, the adoption of Directive 2014/95/EU. Energy companies performed better in terms of sustainability when more indicators were reported. Their primary focus was on the economic aspect of sustainability, particularly corruption. They gave less importance to the environmental aspect, mainly reporting on emissions. The social aspect received the least attention, although indicators for employee education and training were mentioned most frequently. The analysis showed that the metrics are statistically significantly correlated and complement each other, highlighting the need to consider a variety of metrics when assessing sustainability performance in the energy industry.

1. Introduction

Sustainability reporting is a multifaceted concept that has been widely studied in the literature, yet a consistent definition remains elusive. Adams and Frost [1] defined it as the integration of ethical, social, environmental, and economic considerations in corporate reports, while Ashby [2] viewed it as a tool for managing change towards sustainable business practices due to the integration of long-term profitability with ethical behavior, social justice, and environmental stewardship. The motivations for companies to engage in sustainability reporting can vary, but can include enhancing corporate legitimacy [3], improving their corporate image and credibility with shareholders [4], and increasing transparency in accounting for the company’s activities and results [5]. Mandatory disclosure of comprehensive non-financial information, particularly environmental data, in corporate sustainability reports is expected to play a significant role in fostering investment in environmentally friendly and sustainable initiatives that are in line with the objectives of the European Green Deal [6].

The Global Reporting Initiative (GRI) Standards are widely recognized as the preeminent international framework for sustainability reporting, as various studies have shown [7,8,9].

Numerous authors have noted the widespread acceptance and implementation of the GRI Standards globally, as highlighted in the studies [10,11], among others. The comprehensiveness, prestige, and visibility of the GRI Standards are identified as key factors for their widespread adoption [12]. The European Union (EU) has recognized the benefits of using the GRI Standards for sustainability reporting. Nearly half of the world’s sustainability reports were published in the EU in 2012 based on them [13]. In 2017, this figure rose to 77%, according to KPMG [9]. The implementation of Directive 2014/95/EU on sustainability reporting in the national laws of EU member states has also elevated sustainability reporting to a higher level. This directive obliges large EU companies with more than 500 employees to report on non-financial information, including sustainable business dimensions, which started in 2017 and was included in their reports for the 2017 fiscal year.

Measuring the impact and performance of a company’s sustainability efforts through sustainability reporting is a complex and challenging task, mainly due to the lack of standardization. Companies report on selected GRI Standards and indicators related to their operations in a variety of ways, including mentioning, describing, or quantifying indicators. This varying application of and adherence to the GRI Standards results in inconsistencies in the presentation of the sustainable management dimensions, posing a challenge for the sustainability reporting assessment. To address these challenges, various composite sustainability indices have been created, considering factors such as geographic location, industry affiliation [14,15,16], stock market listing [17,18,19], and presence in specific databases [20].

As far as we are aware, there is currently neither a standardized system for measuring sustainability in companies nor a comprehensive study that assesses both the transparency and engagement of energy companies in reporting on sustainability performance. Furthermore, previous studies have not found a correlation between the level of sustainability reporting and the performance of energy companies. This results in three research questions that are addressed in this paper: (i) What is the level of sustainability reporting and the sustainability performance reporting of EU energy companies; (ii) Is there a clear relationship between sustainability reporting and sustainability performance of energy companies; and (iii) Have regulations, in particular Directive 2014/95/EU, impacted sustainability reporting in the EU energy industry? The lack of answers to these questions complicates determining their commitment to sustainability issues globally and within the EU. In light of institutional and stakeholder theories, this study therefore evaluated the transparency and engagement demonstrated by EU energy companies in their reporting on sustainability performance. According to these theories, stakeholder pressures in the corporate environment can lead to a consistent, sustainable system [21].

To achieve the study objectives, three sustainability reporting metrics were developed, and the relationship between them was analyzed for the period of 2016–2019. The metrics employed in this study were the Average Sustainability Reporting (ASR) score and, depending on the method applied (the Analytic Hierarchy Process (AHP) or Data Envelopment Analysis (DEA)), the Sustainability Reporting AHP Index (SRI_AHP) and the Sustainability Reporting DEA Index (SRI_DEA). Taken together, they provide a more comprehensive and accurate assessment of both the disclosure and the actual performance of energy companies in terms of sustainability. The energy industry is facing increasing pressures from institutions and stakeholders to take more responsibility in the area of sustainability [22] as concerns about environmental and social impacts, political influences, and sustainable development continue to grow. Despite these pressures, research in the field of sustainability reporting has so far been limited [23].

As the purpose of the analysis was also to evaluate the effect of mandatory reporting on the level of sustainability reporting and sustainability performance, the period of 2016–2019 was selected as the study period. This is because the European Commission [24] announced in December 2019 its intention to review Directive 2014/95/EU and conduct a public consultation on the weaknesses identified, which took place in the first half of 2020. The decision to limit the study to this timeframe was made to minimize the potential impact of the announcement on the non-financial reporting practices of energy companies.

This study adds to the literature by analyzing how regulatory laws, specifically Directive 2014/95/EU, affect sustainability reporting in the energy industry. By developing three metrics (ASR, SRI-DEA, and SRI_AHP scores) that simultaneously evaluate the level of sustainability reporting and performance, it deepens the understanding of the extent and manner in which energy companies report on and perform their sustainability initiatives. Furthermore, this study shows that companies that devote more resources to sustainability reporting also have higher sustainability performance levels by revealing the correlation between sustainability reporting and performance levels. The findings can serve as a valuable tool for regulators and policymakers when it comes to formulating guidelines and standardizing corporate sustainability reporting. The study aids in monitoring sustainable business practices at energy companies and communicating them to the public. Furthermore, the findings are of great importance to company management and investors, who can use them as a direct and reliable source of information about a company’s sustainability initiatives.

The structure of this paper is as follows: The next section reviews the practices and challenges related to sustainability reporting. Section 3 describes the methodology used in this study, including a description of the sample data and methodological approach. Section 4 presents the results of the empirical analysis of the sustainability reporting indices of European energy companies. Finally, Section 5 makes concluding remarks with policy implications.

2. Conceptual Background: Corporate Sustainability Reporting

The Triple Bottom Line (TBL) concept by Elkington (1997) provides a comprehensive framework for integrating sustainability into business practices [25]. This approach evaluates corporate sustainability by embedding economic, social, and environmental dimensions into business processes. Kantabutra and Ketprapakom [26] defined corporate sustainability as the leadership and management strategy that a company adopts to achieve profitable growth, taking into account all three dimensions of sustainability. Incorporating the economic, environmental, and social dimensions into the business process requires the ability to care for present and future generations [27]. Due to its focus on the future, sustainability is seen as a long-term business goal. Costa et al. [28] pointed out that the future needs of the company depend on how corporate sustainability secures the needs of shareholders. Accordingly, corporate sustainability increases shareholder value, sustainable growth, and efficiency [29].

Although sustainability is increasingly recognized as an important long-term corporate goal that increases shareholder value and promotes sustainable growth, there are still challenges in addressing social concerns, such as the needs of different specific stakeholder groups. Stakeholder involvement in corporate activities ensures transparent sustainability reporting [30]. Spallini et al. [31] emphasized that despite common requirements, sustainability information varies in transparency. Băndoi et al. [32] pointed out that legal requirements for sustainability reporting have the potential to improve transparency and encourage companies to report sustainably, but they can also be costly and prevent companies from complying with regulations.

Despite the legal requirements for sustainability reporting in the EU and the comprehensive and precise nature of the GRI Standards, the manner of sustainability reporting is still left to the discretion of companies, making it difficult to compare and ensure consistency in sustainability reporting between organizations and preventing accurate measurement of overall sustainability levels. The composite sustainability index is considered the most widely accepted measure for sustainability reporting due to its numerous advantages [33]. These include its ability to cover the main sustainability dimensions (economic, environmental, and social) and to take into account both quantitative and qualitative indicators. However, there is empirical evidence that various frameworks, approaches, methods, and data sources are used to calculate sustainability reporting indices. Some of them include the composite sustainability index [34,35], while others include the composite sustainability performance index [17,33,36]. Nevertheless, to our knowledge, no new studies have been found that have developed indices based on corporate sustainability reports that assess the engagement and transparency of energy companies in sustainability reporting, as well as examine the relationship between the level of sustainability reporting and sustainability performance.

Most sustainability reporting measures are based on the GRI Standards [37,38,39,40] or on the GRI Standards in combination with other frameworks such as the United Nations Committee on Sustainable Development Indicators and the OECD [41]. In addition, some measures are based on literature reviews [42]; specific databases, such as Sustainanalytics [20]; the Bureau of Economic Analysis [15] or Eurostat [16]; or a combination of the GRI Standards and literature reviews [34].

Several authors have advocated for the construction of sustainability reports and performance measures based on subjective methods, of which, the AHP is the most commonly used [14,43]. Others rely on objective methods, such as DEA [36,44] or principal component analysis [19,42], regression analysis [15,41,45], or a combination of the latter two [20]. Accordingly, there is a measurement gap that this study attempted to fill by developing metrics that incorporate both subjective and objective methods, each of which has advantages and disadvantages. According to Greco et al. [46], the former can result in inconsistencies and criticisms due to their bias, while the latter can result in data that are considered unrealistic due to the lack of subjectivity.

The source of data used to create sustainability measures also differs. Most previous studies relied on data from corporate sustainability reports and questionnaires sent to either the companies or academic experts [38,40], while some studies only used sustainability reports [19,47] or questionnaires [14,33]. Furthermore, the lack of a standardized approach for selecting indicators and determining the number of indicators per category results in the dependence of the number of measures on the methodology for their selection.

Although there are several studies on measuring sustainability in the energy industry, only a few of them deal with sustainability reporting in the EU or in Europe. For example, Tutak et al. [48] focused on EU countries in their study to assess them in terms of their energy development. Ligus and Peternek [49] calculated the Sustainable Energy Development Aggregated Index to assess EU member states on the path to sustainable energy, and Agudelo et al. [50] identified the drivers of the sustainable performance of companies in the European energy sector. They based their sustainability performance on the Refinitiv Eikon Datastream environmental, social, and governance (ESG) score. To our knowledge, there is a distinct lack of research on comprehensive sustainability measurement in the energy industry based on information from sustainability reports, especially in the EU and in recent years.

Energy companies are facing increasing pressure to take on a broader range of sustainable responsibilities, including social obligations that go beyond legal and environmental obligations [50].

Over 75% of greenhouse gas emissions in the EU are caused by the production and consumption of energy in all sectors of the EU economy [6]. These gas emissions are the main cause of global warming and climate change and cannot be ignored. It is extremely important for energy and environmental policy authorities to know what the energy industry is doing to operate sustainably. With this in mind, adopting sustainable business practices, with a particular emphasis on decarbonization from the perspective of energy companies, and achieving climate neutrality by 2050, is crucial [51].

This paper fills the research gap by developing a sustainability reporting framework for energy companies. This framework consists of three metrics that reflect the level of transparency and engagement of energy companies in reporting on their economic, environmental, and social sustainability performance. Based on the developed metrics, the sustainability reporting in business operations of energy companies was assessed, and their sustainability reporting in business operations was compared during the research period.

3. Sample, Data, and Methodology

3.1. Sample and Data

The sample consists of EU energy companies, selected from sector D35.1: Electricity production, transmission, and distribution. The reason for this is that 90.23% of companies in the EU energy sector belonged to this sector in 2016 [52]. Therefore, less than 10% of energy companies are involved in other activities within the EU energy sector. According to the same study, the security of the electricity supply is crucial for the European Member States. The selection of companies was based on a list of energy companies from a Study on the quality of electricity market data of transmission system operators, electricity supply disruptions, and their impact on the European electricity markets [53]. This study contains the most important electricity producers in the EU. Exact figures for the total population of European energy companies are challenging to determine due to the dynamic and fragmented nature of the industry, but the selected companies represent a significant proportion of the market share in their respective countries. These companies have a significant impact on the market as most of them have a dominant or monopolistic position in their respective national markets (e.g., EDF in France accounted for 82.5% in 2016). Due to their market dominance, they are the main players in shaping market trends and are more likely to have comprehensive sustainability reports. This makes them highly representative for the purposes of this research and ensures the relevance of the results. Based on sustainability reports, we selected companies that published online sustainability reports based on the GRI Standards in English in the period of 2016–2019.

As the study covers the period up to 2016, it took into account the following changes that have occurred since then: (1) Electrabel became a subsidiary of Engie; (2) Esset became part of Innogy, which was acquired by E.ON in September 2019; (3) Lietuvos Energijos Gamyba became Ignitis Gamyba in September 2019, and is part of the Ignitis Group; and (4) Ørsted was Dong Energy until November 2017. This is important as, according to Directive 2014/95/EU, only parent companies are required to report on sustainability in their consolidated report if they are part of a group of companies.

The sample excludes the following companies: Electrocentrale Bucharest (no website available in English), EAD–Natsionalna Elektricheska Kompania, EAC–Electricity Authority of Cyprus, and Enemalta PLC (no sustainability reports published online); Ørsted, Electrabel subsidiary of Engie, Eesti Energia AS, EDF, Ignitis Group, Slovenské elektrárne, HSE–Holding Slovenske Elektrane (sustainability reports not published according to the GRI Standards); and Enovos, which is part of Encevo (inconsistent reporting according to the GRI Standards). Therefore, the final sample was reduced to 16 European companies (

Table A1. EU energy companies included in the sample.

No.	Company	Country
1.	Verbund	Austria
2.	ČEZ	Czech Republic
3.	Fortum	Finland
4.	PPC SA—The Public Power Corporation S.A.	Greece
5.	HEP—Hrvatska elektroprivreda	Croatia
6.	ESB—Electricity Supply Board	Ireland
7.	ENEL	Italy
8.	Latvenergo AS	Latvia
9.	MVM Hungarian Electricity	Hungary
10.	Innogy	The Netherlands
11.	E.ON	Germany
12.	PGE—Polska Grupa Energetyczna S.A.	Poland
13.	EDP—Energias de Portugal	Portugal
14.	Endesa	Spain
15.	Vattenfall	Sweden
16.	SSE—Scottish and Southern Energy	United Kingdom

in Appendix A).

The data sources used to calculate the sustainability reporting indices consist of 64 sustainability reports from 16 energy companies (4 per company) and a questionnaire sent in January 2022 to experts employed by the sample companies. The reports, which are based on the GRI Standards, were subjected to a content analysis. This is a common method for studying corporate sustainability and sustainability reporting [54,55]. The reports are based on a total of 92 individual GRI Standards, which are divided into three sustainability dimensions: economic, environmental, and social. The economic standards are divided into six categories (GRI 201 to GRI 207) and comprise 13 specific standards (indicators) that monitor the organization’s main economic impacts and the flow of capital among various stakeholders. All of these economic standards have been effective since 1 July 2018, except for GRI 207 Tax, which came into effect on 1 January 2021. Companies were able to apply these standards before the mandatory dates mentioned, as the standards were already available earlier. The environmental standards are divided into eight categories (GRI 301 to GRI 308), which include 35 individual standards. According to the GRI standards [56], the environmental dimension of sustainability encompasses all of the organization’s effects on the environment, encompassing both living and non-living natural systems (e.g., water, air, and land). The social standards are divided into 19 categories (GRI 401 to GRI 419), which comprise a total of 44 individual standards. Compared to the economic and environmental standards, the social standards represent the most numerous categorization and list of individual standards. These standards relate to the company’s stakeholders and human relations.

The questionnaire examined the views of experts from the EU energy companies surveyed on the importance of the GRI sustainability indicators. The experts were individuals holding senior positions in the company, making decisions related to sustainability, and responsible for sustainability reporting in the company. In previous studies, experts from the fields of activity of the companies studied were also interviewed (e.g., [38]—energy and petrochemistry; [33]—steel industry; [47]—banking sector; [16,41]—processing industry), as well as academic experts [14].

The GRI Standards for large companies have been in force since 1 January, 2017, replacing the previously used GRI G4 Guidelines. In order to enable an analysis of the sustainability reports from 2016 to 2019, the GRI G4 Guidelines were converted into the GRI Standards, allowing for comparisons of the sustainability indicators in 2016 with those in 2017, 2018, and 2019. The conversion of the GRI G4 Guidelines into the GRI Standards was carried out by the GRI, which performed the conversion in the G4 to the GRI Standards Mapping [57].

3.2. Methodological Approach

The development of sustainability reporting metrics usually involves several phases, the number of which can vary depending on the type of index and the nature of the data to be analyzed. As shown in Figure 1, the process begins with the selection of a conceptual framework and sustainability indicators. For this study, the triple bottom line (TBL) approach was chosen, and the GRI framework was used to measure sustainability in all three dimensions: economic, environmental, and social.

A content analysis of the sustainability reports of energy companies enables the determination of three different measures. One measure is utilized to assess the level of sustainability reporting (ASR score, based on economic, environmental, and social sub-measures), while the other two measures are employed to measure sustainability performance reporting (SRI_AHP and SRI_DEA), both of which are based on numerical indicators. As a subjective measure, SRI_AHP goes beyond numerical data and incorporates expert opinions from the field of sustainability and sustainability reporting. The measures derived from this—ASR, SRI_AHP, and SRI_DEA scores—are then used to assess the transparency and engagement of energy companies in reporting on sustainability.

This study employed content analysis to assess how the selected energy companies present the individual GRI Standard/indicator in their sustainability reports. A total of 64 sustainability reports from the period between 2016 and 2019 were analyzed. The reports were reviewed by the authors and manually coded. The content analysis was conducted manually due to the complexity of the textual data, which also includes numerical information. Additionally, the reports contain illustrations such as tables, graphs, and photographs, making them more challenging for software processing. Moreover, content analysis was applied to evaluate the substance and quality of each company’s sustainability report. This cannot be achieved with a computer program. The manual approach allowed for the identification of similarities and differences in the sustainability reporting practices adopted by the companies. The analysis assessed how frequently individual GRI indicators were represented in the sustainability reports of the companies and the manner in which these indicators were disclosed in each observed energy company’s sustainability reports. Based on the method of disclosure, the indicators were categorized into five groups depending on how they were presented in the sustainability reports: not published, published without description, qualitatively described, quantitatively described, and both qualitatively and quantitatively described. The overall summary of the published sustainability indicators, the specific expressions, and the observations led to recommendations for the indicators from which composite indices were constructed. As there are no mandatory or prescribed sustainability indicators, companies have the flexibility to decide which indicators to include in their reports and how to present them. The analysis focused on quantitative indicators.

Constructing the ASR Score. Content analysis was used in the study to determine the frequency of reporting on each GRI Standard by company, year, and sustainability dimension. Based on the content analysis data regarding the presence of indicators (published and unpublished), the sustainability sub-indices (economic, environmental, and social) were calculated. Calculating the ASR score consists of three steps: (1) coding the values; (2) calculating the sub-indices; and (3) calculating the ASR index. The sub-indices are calculated based on the coded values. A value of 0 was assigned to unpublished indicators, and a value of 1 to published indicators. Each sub-index includes GRI indicators for its respective category (e.g., the economic sub-index includes all GRI economic indicators). The sub-indices are then calculated by summing the assigned values and calculating the arithmetic mean by dividing that sum by the total number of indicators within each category (economic: 13; environmental: 35; social: 44). The ASR score is the sum of the sub-measures of sustainability reporting for each company observed during the study period. The sub-measures range from 0 to 1, and the ASR score ranges from 0 to 3, with higher values indicating companies that report on more sustainability indicators and have stronger communication with the public regarding sustainability.

In contrast to the approach used for the ASR score, where all 92 GRI indicators were analyzed, the sustainability performance reporting indices are developed based on a subset of indicators deemed relevant by companies or sustainability experts. This approach provides a more targeted and precise measurement of the sustainability performance of energy companies.

Constructing the SRI_AHP. The SRI_AHP was calculated on the basis of quantitative indicators that occur in over 50% of sustainability reports. This criterion was met by 20 GRI indicators, consisting of 4 economic, 10 environmental, and 6 social indicators. The complete list of these indicators can be found in

Table A3. Selected GRI Standard indicators.

Code	Sustainability Dimension	Label
201-1	Economic	Direct economic value generated and distributed
201-4		Financial assistance received from government
203-1		Infrastructure investments and services supported
205-3		Confirmed incidents of corruption and actions taken
301-1	Environmental	Materials used by weight or volume
302-1		Energy consumption within the organization
303-1		(2016) Water withdrawal by source
303-3		(2018) Water withdrawal *
305-1		Direct (Scope 1) GHG emissions
305-2		Energy indirect (Scope 2) GHG emissions
305-3		Other indirect (Scope 3) GHG emissions
305-7		Nitrogen oxides (NOX), sulfur oxides (SOX), and other significant air emissions
306-1		(2016) Water discharge by quality and destination
303-4		(2018) Water discharge *
306-2		Waste by type and disposal method
307-1		Non-compliance with environmental laws and regulations
401-1	Social	New employee hires and employee turnover
401-3		Parental leave
403-2		(2016) Types of injury and rates of injury, occupational diseases, lost days, and absenteeism, and number of work-related fatalities
403-9		(2018) Work-related injuries *
404-1		Average hours of training per year per employee
404-3		Percentage of employees receiving regular performance and career development reviews
405-1		Diversity of governance bodies and employees *

* Indicators were taken into account together because they represent the same indicator.

in Appendix A. The indicators were classified for each company and each year, as well as for the entire period under review, based on the mean value for each sustainability dimension.

Normalization is required to determine appropriate weights for the selected indicators. For this purpose, the units were standardized, missing values were replaced by average values, and the data were transformed by logarithmization to ensure comparability. The distance from a reference point method, as outlined by the OECD [58] guidelines, was then used to normalize the individual indicators. The reference point was established based on the indicator’s impact on business sustainability, with the minimum undesirable and maximum desirable performance serving as reference points for the entire study period. The normalized values were obtained by dividing the indicator value for a given company and period by the reference indicator value, which was assigned a normalized value of 1.

The subjective weighting was determined using the AHP method, in which experts assessed the importance of each indicator category relative to the others within each sustainability dimension. Using Saaty’s 9-point scale [59], the experts awarded points for each indicator category. Based on the expert opinions, pairwise comparison matrices were constructed for each sustainability dimension, which were then combined into a matrix with relative weights. These weights were normalized by dividing each weight in the matrix by the sum of the weights in the corresponding column. The normalized AHP weights were calculated for each category of indicators, and each indicator within each category was assigned a normalized AHP weight based on the category’s weight.

To obtain the weighted value of each indicator, the normalized indicator values were multiplied by the corresponding AHP weights. These weighted values were then used to calculate the economic, environmental, and social sub-indices, which were summed to derive the SRI_AHP. The SRI_AHP’s value ranges from 0.50 to 2.25, with the highest value indicating that the reference values for all indicators (highest values for positive indicators and lowest values for negative indicators) were achieved during the entire observation period.

Constructing the SRI_DEA. The SRI_DEA was constructed using an input-oriented, nonparametric mathematical programming-based Banker–Charnes–Cooper (BCC) DEA method [60]. The BCC DEA is a model with variable returns to scale (VRS) in which the level of output can vary depending on changes in input (it can remain the same or increase or decrease). The BCC DEA is divided into input-oriented and output-oriented models. An input-oriented model means that an inefficient unit becomes efficient by reducing its inputs while its outputs remain at least at the same level.

This method was utilized to assess both the quality of sustainability reporting and the efficiency of energy companies in producing sustainable outputs from inputs. The DEA method has been used for the same purpose in various empirical studies, either as a stand-alone approach [36,44,61,62] or in combination with other methods such as Principal Component Analysis (PCA) [20]. For this study, input and output indicators were only selected if they were quantitatively published in at least 75% of the sustainability reports. The sustainable efficiency of each decision-making unit (in this case, energy companies) was calculated for each company using the minimization function under a given set of constraints. The DEA assigns efficient companies (in terms of sustainability) an efficiency score of 1, while the scores of inefficient companies are in the range [0,1]. The improvement in efficiency is achieved by minimizing inputs for a given level of output.

Microsoft Excel 2016 was used to calculate the sustainability measures and the AHP method, while DEAOS software was used for the DEA method. All other analyses were carried out using the IBM Statistical Package for the Social Sciences (SPSS) 25.0.

4. Empirical Results on Sustainability Reporting in Energy Companies

Table 1. Descriptive statistics of key characteristics for selected energy companies.

	Mean
Characteristic	2016	2017	2018	2019
Number of employees	20,038	21,109	21,249	23,395
Total revenue (in millions of euros)	15,901,529	16,284,255	15,270,364	15,232,375
Total asset (in millions of euros)	30,175,658	29,721,367	30,657,070	34,185,726
	Minimum	Maximum	Mean of a period	Std. Dev.
Number of employees	2742	78,948	21,448	18,741
Total revenue (in millions of euros)	13,746	80,327,000	15,672,131	20,024,969
Total asset (in millions of euros)	41,439	171,426,000	31,184,955	38,720,828

shows the descriptive statistics of the key characteristics for the sample energy companies and illustrates the diversity that exists between them. It depicts the average number of employees in the sample for the entire period, indicating an overall average of 21,448 employees, with the highest average recorded in 2019. The lowest number of employees was documented at Verbund in 2018 with 2742, while the highest level was observed at E.ON in 2019 with 78,948. On an annual basis, the highest average revenue among the monitored companies was achieved in 2017, with the lowest total revenue in 2017 at Vattenfall with EUR 13,746 million. Conversely, the highest total revenue was recorded at Enel in 2019, at EUR 80,327 million. The average total assets of all companies observed amounted to EUR 31,184,955 million. Notably, the average values for the total number of employees, total revenue, and total assets exhibited a declining trend over the observed period.

Three metrics were developed to assess the level of transparency and engagement displayed by energy companies in their sustainability reporting.

4.1. Sustainability Reporting Level in Energy Companies

Based on the methodology presented above, the ASR score was calculated for each company and the energy sector as a whole, for each year, considering each sustainability dimension (economic, environmental, and social). The corresponding ASR values are presented in

Table 2. Average sustainability reporting scores.

Company	2016	2017	2018	2019	Change (%)	Overall Mean	Reporting Level	Sustainability Dimension
								Economic	Environmental	Social
EDP	2.63	2.63	2.63	2.63	0.00	2.630	High (1)	1.00	0.86	0.77
Endesa	2.16	2.53	2.53	2.69	7.59	2.478		0.96	0.8	0.72
Enel	2.11	2.19	2.33	1.76	−5.87	2.098		0.71	0.67	0.72
HEP	2.13	2.18	2.12	1.71	−7.06	2.035		0.88	0.63	0.52
MVM	2.09	1.77	1.84	1.71	−6.47	1.853	Medium (2)	0.83	0.68	0.35
Fortum	1.54	1.58	1.78	1.78	4.95	1.670		0.62	0.54	0.51
ČEZ	1.36	1.72	1.72	1.78	9.39	1.645		0.5	0.59	0.55
Verbund	1.87	1.61	1.52	1.46	−7.92	1.615		0.71	0.49	0.41
PPC	1.00	1.17	1.23	1.15	4.77	1.138		0.37	0.41	0.36
ESB	0.76	1.11	1.05	1.35	21.11	1.068	Low (3)	0.38	0.41	0.27
Innogy	1.31	1.00	1.14	0.76	−16.60	1.053		0.37	0.36	0.33
Latvenergo	0.85	0.87	0.80	0.98	4.86	0.875		0.46	0.21	0.2
SSE	0.71	0.73	0.73	0.75	1.84	0.730	Very Low (4)	0.31	0.29	0.14
PGE	0.68	0.71	0.61	0.61	−3.56	0.653		0.27	0.19	0.19
Vatenfall	0.64	0.66	0.66	0.62	−1.05	0.645		0.23	0.34	0.08
E.ON	0.57	0.52	0.48	0.64	3.94	0.553		0.08	0.29	0.18
Mean	1.40	1.44	1.45	1.40		1.42		0.54	0.49	0.39
Std. dev.	0.68	0.69	0.71	0.67		0.67		0.28	0.21	0.22
Min.	0.57	0.52	0.48	0.61	−16.6	0.553		0.08	0.19	0.08
Max.	2.63	2.63	2.63	2.69	21.1	2.63		1.00	0.86	0.77

.

The reliability of the ASR score was found to be good, with a Cronbach’s alpha value of 0.923, indicating that the sustainability reporting sub-indices consistently measured this construct. The highest ASR score of 2.69 was achieved by Endesa in 2019, while the lowest value of 0.48 was obtained by E.ON in 2018. The industry average for the ASR scores was narrow, ranging from 1.40 to 1.45, which indicated not only a low level of reporting (with a maximum score of 3), but also stagnation in sustainability reporting. This observation was also supported by the results of the dependent t-tests, where the p-value was greater than 0.1, indicating that there was no difference in sustainability reporting before and after the introduction of the legal obligations, as well as during their implementation. However, there are contradictory findings in the literature regarding the impact of such regulations on sustainability reporting. While some studies reported favorable effects [18], others reported unfavorable effects [19,55], suggesting the contextual dependence of the regulation’s impact.

The analysis of the sustainability reporting levels enabled the classification of energy companies into three trend categories: growth/decline (>±1.5%: 8 and 6 companies, respectively) and stagnation (<±1.5%; 2 companies), as presented in Table 2. During the observed period, ESB achieved the highest increase in reporting level, while Innogy experienced the largest decrease. The significant changes in these companies are due to shifts in the number of indicators reported in the companies’ sustainability reports. ESB has notably expanded its reporting, with the number of environmental indicators rising from 9 (2016) to 20 (2019) and social indicators from 5 (2016) to 14 (2019). Conversely, Innogy significantly reduced its reported indicators, particularly the environmental indicators from 15 (2016) to 9 (2019) of 35 possible indicators and the economic indicators from 7 (2016) to 3 (2019) of a total of 13 possible indicators. EDP had the highest reporting level (average 2.63), and interestingly, the reporting level remained consistent throughout the period. Notably, some companies consistently reported above or below the industry average, indicating consistent reporting practices over time. However, more companies with above-average sustainability reporting scores experienced a decline in reporting levels over the observed period compared to those that experienced growth. This could be a sign of a movement towards a consensus on sustainability reporting practices (see [55] for discussion), which may contribute to consistency [39] and credibility [45] of reporting. Similar observations were made for individual sustainability dimensions. However, achieving consistency in sustainability reporting requires standardization at a higher level of reporting.

Table 2 presents the classification of energy companies into four groups based on their average reporting level (high, medium, low, and very low), with each level corresponding to a quantile range of 25%. The companies with a high reporting level were EDP, Endesa, Enel, and HEP, while SSE, PGE, Vattenfall, and E.ON had a very low reporting level. It is noteworthy that EDP had both the highest reporting level and the highest level in each sub-index of sustainability, achieving the maximum score in the economic dimension. The table also shows that companies generally reported the most on the economic dimension of sustainability, while the social dimension was the least represented, despite the number of GRI Standards available for reporting. This discrepancy has been observed in previous studies regardless of country or industry [14,34,39], and could be due to companies’ profit orientation (they are profit-driven and therefore prioritize the economic dimension, which is directly related to financial performance) and their perception of stakeholder expectations (they may perceive that their stakeholders, such as investors, are more interested in their financial performance). The results of the paired sample t-test confirmed statistically significant differences between the mean values of the fractions for the sub-measures of sustainability reporting level.

The ASR values, ranging from 0.48 to 2.69, indicate that while all companies are complying with their reporting obligations, there is still considerable room for improvement in their reporting practices to the public.

In the field of energy reporting, there are both desirable and undesirable indicators. This study revealed that companies generally prioritize reporting on desirable indicators, such as employee training and board and workforce diversity, while often avoiding detailed reporting on undesirable indicators such as workplace injuries, corruption, and harmful gas emissions. However, energy companies are expected to increase their efforts to strengthen social responsibility and reduce undesirable environmental indicators, especially given their role as major polluters with ever-increasing emissions.

We anticipated an unsatisfactory environmental performance, offset by other dimensions, especially the economic one. These expectations align with the fact that energy companies contribute to three-quarters of all CO₂ emissions [63]. It seems that emphasizing compensation in the economic dimension serves as a strategic move to avoid penalties, preserve reputation, and maintain positive relationships with stakeholders, especially investors. Thus, our findings suggest that it may not be advisable for companies to voluntarily choose reporting standards due to the potential for manipulation and the need for standardization given the specificities of the sector. Directive 2014/95/EU underlines the importance of standardization to ensure access to all essential sustainability information [64].

4.2. Sustainability Performance Reporting in Energy Companies

To enable a more comprehensive and accurate assessment of sustainability practices in the energy industry, two sustainability performance reporting indices were constructed. The first index is based on the AHP method, while the second one is based on the DEA method. The combination of these two methods provides a more robust evaluation of sustainability performance, taking into account both subjective and objective factors.

SRI_AHP. The content analysis identified categories of quantitative indicators that were present in more than 50% of the sustainability reports, resulting in a total of 13 indicator categories: 3 economic, 6 environmental, and 4 social (see

Table A2. Selected GRI Standard categories.

Category	Sustainability Dimension	Label
GRI 201	Economic	Economic Performance
GRI 203		Indirect Economic Impacts
GRI 205		Anti-corruption
GRI 301	Environmental	Materials
GRI 302		Energy
GRI 303		Water and Effluents
GRI 305		Emissions
GRI 306		Waste
GRI 307		Environmental Compliance
GRI 401	Social	Employment
GRI 403		Occupational Health and Safety
GRI 404		Training and Education
GRI 405		Diversity and Equal Opportunity

in the Appendix A). These indicators were then reviewed by experts to determine their importance or weight. Based on the expert opinions, a consolidated pairwise comparison matrix was created for the selected indicator categories and sustainability dimensions. It is worth noting that the expert opinions were consistent, as supported by the consistency ratio. Specifically, the consistency ratio was 3.9%, 2.9%, and 0.5% for the economic, environmental, and social dimensions, respectively, which is well below the acceptable upper limit of 0.1 or 10%. The indicator categories that received the highest AHP importance weighting were anti-corruption (205; 47.88%) for the economic dimension, emissions (305; 23.36%) for the environmental dimension, and health and safety at work (403; 36.81%) for the social dimension. Specifically, the most frequently reported economic indicator was confirmed cases of corruption and actions taken (205–3) at 70%, while the most frequently reported environmental indicator was direct (Scope 1) GHG emissions (305–1) at 100%. For the social dimension, the most frequently reported indicator was work-related injuries ((403–2 (2016), 403–9 (2018)) at 100%. Most of the energy companies in the sample are either wholly or partly owned by the state (see [65]), which could be a factor in their efforts to present themselves as socially responsible and environmentally conscious to their stakeholders. According to stakeholder theory, government ownership can have a positive impact on sustainability reporting [17], which has been demonstrated in the energy industry by [37]. Notably, emissions were the most commonly reported environmental GRI Standard, which was likely due to the continued emphasis on sustainability in the EU. Additionally, Avram et al. [39] found that energy companies are generally more concerned about the environment than companies in less environmentally sensitive industries. Other studies showed that the importance of image and credibility in sustainability reporting can vary from industry to industry [55] and that environmentally sensitive industries are more inclined to disclose their sustainability practices and information in sustainability reports to improve their credibility and transparency [66].

Using the methodology described in the previous section, the SRI_AHP scores were calculated for each company, each year, and for each sustainability dimension. The industry average was also calculated (

Table 3. Sustainability reporting index scores based on AHP method.

Company	2016	2017	2018	2019	Change (%)	Overall Mean	Performance Level	Sustainability Dimension
								Economic	Environmental	Social
Enel	0.79	0.79	0.78	0.76	−1.28	0.780	High (1)	0.24	0.294	0.246
EDP	0.74	0.76	0.78	0.74	0.00	0.755		0.239	0.281	0.235
Endesa	0.7	0.65	0.65	0.71	0.47	0.678		0.189	0.287	0.199
SSE	0.66	0.66	0.65	0.65	−0.51	0.655		0.199	0.276	0.18
MVM	0.65	0.61	0.66	0.64	−0.52	0.640	Medium (2)	0.21	0.241	0.187
ČEZ	0.64	0.66	0.63	0.62	−1.05	0.638		0.14	0.248	0.249
Fortum	0.64	0.64	0.64	0.62	−1.05	0.635		0.192	0.283	0.158
PGE	0.62	0.6	0.59	0.58	−2.20	0.598	Low (3)	0.129	0.25	0.221
PPC	0.57	0.59	0.62	0.61	2.29	0.598		0.169	0.232	0.196
Innogy	0.58	0.62	0.62	0.56	−1.16	0.595		0.142	0.239	0.215
Verbund	0.59	0.56	0.59	0.56	−1.72	0.575		0.159	0.213	0.202
Vatenfall	0.57	0.57	0.56	0.57	0.00	0.568		0.181	0.209	0.176
E.ON	0.55	0.55	0.56	0.56	0.60	0.555	Very low (4)	0.125	0.24	0.193
ESB	0.55	0.55	0.57	0.55	0.00	0.555		0.159	0.202	0.197
HEP	0.51	0.5	0.56	0.6	5.57	0.543		0.171	0.233	0.138
Latvenergo	0.5	0.5	0.5	0.51	0.66	0.503		0.142	0.201	0.16
Mean	0.62	0.61	0.62	0.62		0.62		0.25	0.20	0.19
Std. dev.	0.08	0.08	0.08	0.07		0.07		0.03	0.03	0.138
Min.	0.5	0.5	0.5	0.51	−2.20	0.5025		0.201	0.138	0.249
Max.	0.79	0.79	0.78	0.76	5.57	0.78		0.240	0.294	0.249

).

The SRI_AHP scores ranged from 0.50 to 0.77. However, energy company sustainability performance reporting stagnated at the industry level during the observed period, as confirmed by the paired samples t-test (p > 0.1), suggesting that these energy companies followed the existing sustainability reporting practices, similar to many European companies (see [18]. This is consistent with the perspective of institutional theory, which emphasizes the importance of the institutional framework for the quality of sustainability reporting (see [55]). In addition, previous research has found that energy companies are reporting more information in their sustainability reports every year [67], choosing between reactive and proactive sustainability strategies to manage various sustainability responsibilities that go beyond legal and environmental compliance [68], producing higher-quality sustainability reports, and are less responsive to regulatory changes [55].

To gain insight into the progress made in reporting sustainability performance at the company level, the companies were categorized into three groups: companies with growth (HEP, PPC, Latvenergo, E.ON, and Endesa), decline (PGE, Verbund, Enel, Innogy, ČEZ, Fortum, MVM, and SSE), and stagnation (EDP, ESB, and Vattenfall). Enel and EDP had the highest level of sustainability performance reporting, while Latvenergo had the lowest SRI_AHP score in the observed period. Similar to the ASR scores, the analysis revealed a decline in sustainability performance reporting for companies that had previously had above-average reported, and this trend was consistent across the individual sustainability dimensions. Notably, reporting on the economic dimension was the most comprehensive, while reporting on the environmental and social dimensions showed no statistically significant difference (p < 0.001). Sartori et al. [44] reported similar results in the Brazilian electricity industry, where economic indicators dominated the GRI reports, and only two companies disclosed more than 70% of the indicators related to sustainability issues. These results are consistent with the other studies mentioned in Section 4.1. Therefore, it is crucial to place more emphasis on environmental and social indicators [43] to improve sustainable performance [14,19].

The categorization of sustainability performance was based on the same percentile approach as for the ASR scores. Table 3 indicates that companies with a medium level of sustainability performance reporting showed a decreasing trend in reporting, while companies with the lowest level generally showed an increasing trend. It is important to note that all companies can still improve sustainability performance reporting, as the maximum value of the index is 1.0.

SRI_DEA. The SRI_DEA was calculated using the BCC DEA method with VRS, using input and output data published in 75% or more of the sustainability reports (codes 302-1, 305-1, 305-2, 305-3, 305-7, 403-2, 403-9, and 405-1; see Table A3). This was performed for two reasons: first, to construct a measure based on indicators that the energy companies themselves recognize as the most important, and second, to minimize the number of variables in the DEA model, as a large number of variables makes it difficult to differentiate sustainability performance [35]. Two emission indicators (tCO2) were selected as output variables, which represent the total sum of all emissions produced (305-1, 305-2, 305-3, and 305-7) and the number of work-related injuries, which is also the sum of two individual indicators (403-2 and 403-9).

Total revenue (EUR), energy consumption within the organization (GJ), and the diversity of governance bodies and employees (405-1) were selected as input variables.

Table 4. Sustainability reporting index scores based on DEA method.

Company	2016	2017	2018	2019	Change (%)	Efficiency Score	Efficiency Level
HEP	1	1	1	1	0.00	1.000	Efficient (1)
Innogy	1	1	1	1	0.00	1.000
Latvenergo	1	1	1	1	0.00	1.000
PGE	1	1	1	1	0.00	1.000
Vatenfall	1	1	1	1	0.00	1.000
ČEZ	1	0.79	1	1	0.00	0.948	Medium (2)
ESB	0.94	0.89	0.88	0.84	−3.68	0.888
E.ON	0.82	0.7	1	0.72	−4.24	0.810
Fortum	1	0.78	1	0.36	−28.86	0.785	Low (3)
MVM	1	1	0.45	0.5	−20.63	0.738
Enel	0.43	1	0.41	1	32.49	0.710
Verbund	0.48	0.6	0.58	1	27.72	0.665	Very low (4)
PPC	0.5	0.58	0.38	0.99	25.57	0.613
SSE	0.15	0.49	0.1	1	88.21	0.435
Endesa	0.4	0.48	0.25	0.6	14.47	0.433
EDP	0.38	0.23	0.23	0.43	4.21	0.318
Mean	0.76	0.78	0.71	0.84		0.77
Std. dev.	0.31	0.25	0.35	0.24		0.23
Min	0.15	0.23	0.1	0.36	−28.86	0.318
Max	1	1	1	1	88.21	1

shows these results.

Five companies (HEP, Innogy, Latvenergo, PGE, and Vatenfall) operated efficiently with an SRI_DEA score of 1. However, the remaining companies were classified as inefficient and were categorized as medium, low, and very low. EDP was identified as the least efficient of all the companies throughout the observation period. It is noteworthy that all inefficient companies (68.75%) had the potential to operate at the efficiency frontier. Other studies also pointed out the lower sustainable efficiency of energy companies [35] and emphasized the need to invest in emission mitigation measures [61].

Looking at the values year to year, they ranged between 0.71 and 0.84, showing an increasing trend. However, the paired-samples t-test revealed that the difference was not statistically significant (p > 0.1), indicating that there was no significant change in the sustainability levels during the observation period, despite institutional pressure.

4.3. Relationship between Sustainability Reporting and Sustainability Performance in Business

The analysis of the effect of the established level of sustainability reporting and reporting on the sustainability performance of energy companies for each year observed provides insights into their transparency and engagement in sustainability reporting. Notably, the study found statistically significant differences in the scores of all constructed measures of corporate sustainability reporting, including ASR and SRI_AHP (t = 10.229, df = 63, p < 0.05), ASR and SRI_DEA (t = 6.272, df = 63, p < 0.05), and SRI_AHP and SRI_DEA (t = −3.755, df = 63, p < 0.05). The results are shown visually in Figure 2 and are based on Table 2, Table 3 and Table 4.

The results of the study indicate that the sustainability reporting and sustainability performance of all energy companies differed significantly, with the exception of ČEZ, which consistently reported at a medium level in all three measures. For the majority of companies (9), the level of sustainability reporting was the same when comparing the ASR and SRI_AHP scores. However, there were two scenarios in which ASR > SRI_AHP (PGE, SSE, and Vattenfall) and in which ASR < SRI_AHP (ESB, HEP, Latvenergo, and Verbund). Additionally, for all companies except ČEZ, there was a difference in the level of sustainability performance reporting when comparing the SRI_AHP and SRI_DEA scores. The companies that were efficient according to the DEA method primarily had a lower level of sustainability performance reporting, and vice versa. The relationship between the level of sustainability reporting and performance reporting was further supported by the correlation coefficients shown in

Table 5. Pearson correlation coefficients of sustainability reporting measures in business.

	ASR	SRI_AHP	SRI_DEA
ASR	1.000
SRI_AHP	0.602 **	1.000
SRI_DEA	−0.403 **	−0.551 **	1.000

** Correlation is significant under a 0.01 threshold.

. These coefficients not only measure the strength of the relationship but also indicate the degree of substitutability and compatibility.

The results show a significant positive correlation between the ASR and the SRI_AHP scores. This indicates that companies reporting a higher number of sustainability indicators (ASR score) have a higher level of sustainability performance (SRI_AHP). These findings are consistent with previous studies [69,70], which found that higher levels of sustainability reporting are associated with higher levels of corporate sustainability. In contrast, a statistically significant negative correlation was found between the ASR and the SRI_DEA scores as well as between the SRI_AHP and the SRI_DEA. This discrepancy resulted from differences in the methods used and the data selected. In particular, the SRI_DEA focuses exclusively on undesirable outputs, while the SRI_AHP considers both desirable and undesirable outputs.

Despite these differences, the ASR, SRI_AHP and SRI_DEA measures complemented each other and together improved the quality of sustainability reporting. As each measure reflects different aspects of sustainability performance, researchers can use them individually or in combination—depending on research objectives and data availability—to effectively assess sustainability performance.

5. Conclusions and Policy Implications

The aim of this study was to evaluate the transparency and engagement of energy companies in their sustainability reporting by developing three metrics. These metrics were then used to examine the relationship between the level of sustainability reporting and sustainability performance from 2016 to 2019. The calculations were based on the reports of these companies, which adhere to the GRI Standards.

Despite the legal obligation for large energy companies to report on sustainability at the EU level, this practice is still in the development stage. Although companies base their sustainability reports on GRI Standards, it is at their discretion to select the sustainability indicators. This discretion can lead to almost 70% of indicators being reported at below average levels, while only just under 30% of the most frequently reported indicators were reported at above average rates in the period analyzed. This suggests a discrepancy between reporting according to the GRI Standards and institutional pressure. The level of sustainability reporting was also related to the method of reporting, i.e., the lack of quantitative and qualitative data, with 50% of energy companies achieved a low or very low level of sustainability reporting during the observed period. An important challenge in analyzing and comparing reports is the inconsistency in the use of measurement units.

To bridge the methodological gap between objective and subjective metric construction, two metrics for sustainability performance were developed, SRI_AHP and SRI_DEA, which both use data from the sustainability reports of energy companies. SRI_AHP incorporates the opinions of sustainability experts and focuses on frequently reported GRI Standards—a limitation that emphasizes widely used standards and consistently reporting companies. These metrics are specific to energy companies and individual sustainability dimensions: anticorruption (economic), emissions (environmental), and health and safety at work (social). This specificity reflects experts’ interest in negative aspects of the energy sector such as corruption and emissions, which feature prominently in media reports and climate change protests. In contrast, SRI_DEA relies exclusively on consistent quantitative data from the GRI Standards over time, allowing the DEA model to assess how efficiently energy companies manage inputs to influence undesirable outputs.

The results of the study suggest that energy companies have low sustainability performance, highlighting the significant potential for improvement in sustainability reporting and performance. However, it is important to acknowledge that the measurement of sustainability was based solely on the information available in the sustainability reports, which may not be comprehensive. The sustainability performance of companies that do not report on their activities cannot be measured, potentially underestimating their actual achievement in this area. This study found a significant correlation between the level of sustainability reporting and sustainability performance, implying that companies investing more in sustainability reporting also tend to have higher sustainability performance. This finding is plausible, as companies that prioritize sustainability tend to showcase their sustainable activities and commitment to the economic, environmental, and social aspects of their business to investors and the public.

Based on the findings set out in this paper, it is advisable to standardize and normalize sustainability reporting at the EU level, inducing annual external independent audits in a comprehensive format. The specification of binding units of measurement for reporting on sustainability performance in all EU sustainability reports is crucial. To enforce audit recommendations and standardized units, legal regulations on sustainability reporting with defined standards are deemed necessary. While the EU has announced this, it is recommended that issuers of guidelines make these standards mandatory for comprehensive indicator information. Companies not applying a particular indicator should explicitly state its inapplicability and provide reasons for this.

The current performance levels are inadequate; energy companies should prioritize the quantification of sustainable activities to mitigate negative impacts and enhance positive ones for improved sustainability. In this context, it is recommended to include as many quantitative indicators with standardized reporting units as possible in the EU standards. Establishing a supervisory authority at the individual member state level is also advisable to enforce the legal obligation for sustainability reporting, along with clearly defined sanctions for non-compliance.

The main limitation of this study is the lack of standardization of reporting by energy companies, which highlights the need for decision-makers to prioritize the standardization process in the industry. Future research should focus on recommending ways to accelerate this harmonization process and assessing its effects, as well as examining other factors that may affect the industry’s sustainability performance. A better understanding of these factors can help policy authorities to develop effective strategies to promote sustainable development in the energy sector.