The introduction of the obligatory CSR reporting regime in Europe and its influence on quality of information disclosed to stakeholders in the annual reports published by the airline industry

The introduction of the obligatory CSR reporting regime in Europe

and its influence on quality of information disclosed to stakeholders in

the annual reports published by the airline industry.

University of Amsterdam

Master’s Thesis

Graduate School of Communication

Master’s programme Communication Science

Name: Olga Wacławek ID Number: 11886854 Date: 28.06.2019 Supervisor: dr. Irina Lock

Abstract:

The following paper aims at establishing whether the implementation of the European Union’s directive 2014/95/EU increased the quality of the CSR reporting and whether it eliminated the discrepancies in terms of quality of information disclosed to different stakeholder groups. By doing so, it takes a closer look at sixteen annual reports published by the European airline sector and compares the quality of information disclosed on two levels: between the reports published in 2017 and 2018 as well as between respective stakeholder groups addressed in these reports. Furthermore, it also provides a cross-continental analysis of the changes in the overall quality of the reports published throughout the years 2017 and 2018 by European airlines and non-European carriers. The results reveal that the new reporting regime contributed to the improvement of quality between the reports published in 2017 and 2018 as well as between respective stakeholder groups addressed in these reports. Furthermore, the findings in the reports published by non-European carriers did not prove an increase in quality. Despite the improvement of quality of the reports in 2018, the analysis showed that airlines tend to focus their CSR communication more solely on their primary stakeholders and therefore, follow the assumptions of instrumental stakeholder theory.

Introduction

Corporate social responsibility goes by many definitions. Some authors refer to it as triple bottom line (Elkington, 1998), or shared value (Kramer and Porter, 2011). Despite, the interpretation, it is undisputable that CSR activities, or the communication of these activities are becoming an important concern for organisations in the search for sustainable competitiveness (Ogrean, 2017). In 2011, the European Union defined corporate social responsibility (CSR) as “the responsibility of enterprises for their impacts on society” and indicated the need for companies to implement a “close collaboration with their stakeholders” (Bellantuono, Pontrandolfo and Scozzi, 2016). By doing, so it communicated the direction in which the member states should proceed in order to create a competitive economic and social system in Europe. The assumption for creation of such systems were best presented by the Europe 2020 Strategy, which proposed three major pillars that can be related to the triple bottom line theory formulated by Elkington (1998): (1) smart growth for a competitive economy sustained by an innovative and knowledge-based system; (2) sustainable growth, focusing on the green economy principles and (3) inclusive growth, suggesting a direct approach to the social cohesion targets (European Commission, 2010). In 2014, the EU published a directive (2014/95/EU) under which, starting from the reporting year 2018, all large public- interest entities exceeding the average number of 500 employees shall include in their management report a non-financial statement. This regulation could be seen as a reaction for the growing number of accusations concerning the low quality of sustainability reports published voluntarily (Habek and Wolniak, 2015). However, even though it obliges most organisations to disclose their CSR activities in a more standardized way in terms of the topics addressed, the regulation does not particularly define any specifications in accordance to which these topics should be addressed. Therefore, the question of whether it can contribute to quality improvement of the reports remains open.

In accordance to the literature review, a complete communication of the organisation’s CSR is essential in assessing its impact on environment and society (Kim, Hur and Yeo, 2015). The organisations must be accountable to the different stakeholder groups and society in order to determine their long-term survival. The most common definition of stakeholders depicts them as “individuals or a group that affects, or is effected, by the achievement of organisation’s objectives” (Freeman, 1984). Not all of the stakeholders are of the same value to organisations. It has been proven that many companies tend to focus rather on communicating to their primary stakeholders, such as: consumers, investors and neglect the secondary stakeholders, such as NGOs or local communities (Hörisch, Freeman &

Schaltegger, 2014). Typically, companies communicate with shareholders and investors by publishing annual reports, which include financial disclosures (Chen and Gavious, 2015). To provide an information to stakeholders about social and environmental activities, organisations can choose among several communications channels: press releases, newsletters, and organisations’ websites. Such means of communication are characterized by a large latitude when it comes to the published content and are not regulated by any means. That is why, to improve stakeholders' understanding and perception of organisations, companies more increasingly chose the form of a CSR report as a main tool of their communication with stakeholders (Albu and Wehmeier, 2013). CSR report is an effective method of improving an organisation’s relationship with stakeholders as many investors, environmental groups, and governing bodies have primarily relied on the annual reports to obtain an organisation’s financial and non-financial information (Neu et al., 1998) From the organisation’s point of view, publishing sustainability reports allows it to show its concerns towards stakeholders. From the position of respective stakeholder groups, sustainability reports allow for judging the organisations’ accountability and performance, as well as help to asses their actions that go beyond the fulfilment of legal obligations (Baviera-Puig, Gómez-Navarro, García-Melón and García-Martínez, 2015). Moreover, such reports give the stakeholders an opportunity to generate feedback and in some cases for questioning organisation’s legitimate claim. In this perspective, the relation between the organisation and its stakeholders are based on the principles of reciprocity, interdependence, and power (Andriof and Waddock, 2002, p.19)

Nevertheless, the practice of publishing CSR reports is often confronted with criticism. Fassin (2008) claims that sustainability reports are more of a publicity tool than an actual evidence for organisations’ CSR activities. Compared to others, the assessment of the CSR reports in the airline industry seem to be especially low. The study conducted by Cowper-Smith and de Grosbois (2011) indicates that the airline industry’s adoption of CSR is relatively slow, as the number of airlines consistently disclosing the CSR information for more than a decade is small in relation to the size of the sector. Furthermore, it has been reported that when it comes to The 2014 Dow Jones Sustainability Index (DJSI) the airline industry achieved the lowest scores from all of the participating industries (Kuo, Kremer, Phuong & Hsu, 2016). Taking into consideration the report published by The World Tourism Organization, which revealed that tourism constitutes 5% of global carbon dioxide (CO2) emissions, and the level of emissions is expected to grow 150% by 2035, these statistics call

Using a method of quantitative content analysis, this master thesis takes a closer look at European airline industry in order to compare the voluntary reports of 2017 with the reports of 2018 published in the new obligatory reporting regime. The purpose of this research is to discover whether this mandatory reporting regime contributed to the reduction of existing discrepancies found in the quality of information disclosed to respective stakeholder groups, as well as whether it improved the overall quality of the reports. Doing so it aims to answer the guiding research question, namely: To what extent the 2018’s obligatory CSR reporting regime improved annual reports in terms of the quality of information disclosed towards different stakeholder groups?

Theoretical background

Communication challenges of the airline industry

The legitimacy theory states that in order to survive in the competitive market, organisations need a “licence to operate” (Deegan, 2002). This licence can be obtained by regular and accurate responding to changing norms, values and social expectations. Approval of society is crucial for legitimizing one’s activities (Deegan, 2002). Airline industry is indisputably leaving a significant environmental footprint and society seems to be recognizing this issue very well (Kuo, Kremer, Phuong & Hsu, 2016). Regardless of the efforts and technological progress, prevention or even reduction of the emissions caused by the rapidly growing air traffic is challenging (European Commission, 2011). For that reason, airline industry may be facing problems with its legitimacy (Kuo, Kremer, Phuong & Hsu, 2016). In order to prevent a legitimacy gap from evolving, the perpetuate reclamation of the social contract is crucial. This reclamation happens on multiple channels, involving sustainability, Corporate Social Responsibility (CSR) or Creating Shared Value (CSV) reports. As the airline industry is strongly connected to gas and noise emissions, the active practice of CSR seems to be essential for maintaining a good public opinion. The term “sustainable aviation” becomes commonplace (Kuo, Kremer, Phuong & Hsu, 2016). Nevertheless, despite the little amount of research on CSR in airline industry, some academics are concerned for the industry’s sustainability programs. The study conducted by Cowper-Smithand de Grosbois (2011) identified two most commonly communicated dimensions significant to the airlines in setting their goals, namely: the environment and the social and economic dimension. However, the same research revealed that most of the goals are never being reached.

Several environmental initiatives have emerged in order to help the airline industry become more sustainable. For instance, The International Air Transport Association (2009) has developed a four pillar strategy aimed at reducing aviation emissions. The main focus of this strategy lies in: (1) technology and developing new alternative energy sources; (2) operations, maximizing the efficiency and minimizing weight; (3) infrastructure when it comes to optimizing air routes and airport procedures and (4) economic instruments aiming at introducing certain incentives for a lower environmental footprint. Many airlines admit compliance to these standards and try to gain recognition for their own programmes aiming to improve sustainability performance. Furthermore, an increased number of airlines communicate implementations of sustainable practice that take place not only in the air, but also on the ground. Such practice includes the reduction of consumption of water and paper, the reduction of the amount of waste produced, the recycling of waste and the transition to use more biodegradable materials on all levels of airlines’ operations. Because of the amount of the environmental challenges the airline industry faces, the more social aspects of the corporate responsibility seem to be neglected in reporting on the CSR activities. Nonetheless, as it will be explained in the following paragraph, the aviation industry holds responsibility towards multiple stakeholder groups. Such responsibility includes: the assurance of health and safety in the workplace, the assurance of the good labour conditions when it comes to gender equality, labour management relations etc., the assurance of compliance with human rights within one’s company as well as the supply chain, the prevention of human trafficking and unlawful sex tourism (GRI, 2013).

Stakeholder theory

Stakeholder theory is arguably one of the most prominent and well-know theories of business management (Stieb, 2009). Since its publication in 1984 by Freeman, it got a lot of recognition and became the subject of many debates in academia. In short, the theory assumes that companies should shift their decision-making power and the benefits of labour from stockholders, or shareholders to stakeholders, defined by Freeman as “any group or individual who can affect or is affected by the achievement of the activities of an organization.” (Freeman, 1984, p.46). The stakeholder theory recognises the legitimate claim of shareholders but opposes the idea that shareholders should be the privileged group over the interests of other legitimate stakeholders (Stieb, 2009). It assumes that organisations should strive to achieve their own objectives, such as profitability, while equally satisfying the legitimate

its interpretations and explanations. Following Jones and Wicks (1999) there are four main statements at the core of stakeholder theory “(1) the firm has relationships with constituent (stakeholder) groups, (2) the processes and outcomes associated with these relationships are of interest, (3) the interests of all legitimate stakeholders have value, (4) the focus of stakeholder theory is on managerial decision-making” (Hillman and Keim, 2001). Hill and Jones (1992) summarise two similar principles of the theory, namely: “that to perform well, managers need to pay attention to a wide array of stakeholders; and that managers have obligations to stakeholders which include, but extend beyond, shareholders” (p.20). All of these theorists notice the tendency for organisations to think about the business and its stakeholder through the prism of a “Value Allocation Question: How should one distribute the burdens and benefits of corporate activities among stakeholders?” (Freeman, 1984). The stakeholder theory assumes that instead of focusing on the Value Allocated Question, an organisation should rather pay attention to the “Value Creation Question: How can one create as much value as possible for all stakeholders?” (Freeman, 1984). Value as such can be created by the use of multiple channels, but the primary channel is communication.

The importance of communication is often acknowledged as an essential element of successful CSR strategy along with dialogue, which is seen as a key element in stakeholder management (Szwajkowski, 2000). “Content is everything”, therefore it is crucial for the organisations to conform its norms and behaviours to the concerns and capabilities of each respective stakeholder group (Szwajkowski, 2000). Constructive and open dialogue is essential for securing a short-term performance, as well as long-term organisational wealth benefiting the whole society, as defined by the economic, social and political environment (Kakabadse, Rozuel & Lee-Davies, 2005).

Subsequently, the stakeholder theory developed into two opposing schools, originating from the work of Donaldson and Preston (1995). The first one – instrumental stakeholder theory perceives stakeholder management as an instrument to achieve assumed outcomes, principally profit. The second one – normative stakeholder theory gives the biggest importance to the ethical legitimacy of the stakeholders’ claims on the organisational purpose (Hill and Jones, 1992). The entities of instrumental stakeholder theory are often perceived as primary as it tends to be assumed that organisations which do not include (in their strategy) their primary stakeholders’ concerns, challenge their long-term survival. However, the reality proved that it is often the normative dimension of the stakeholder theory which turns out to be critical (Hillman and Keim, 2001).

value for an organisation. There are several ways of categorizing the stakeholder groups. The most common framework is the internal/external stakeholder classification (Kakabadse, Rozuel & Lee-Davies, 2005). It is often seen that stakeholders are referred to as primary/secondary, in other words as directly/indirectly affected or benefitting from the organisations’ activities (Kakabadse, Rozuel & Lee-Davies, 2005). Another, suggested classification is one, dividing the stakeholder into voluntary and involuntary ones. Furthermore, an interesting categorization is one developed by Mitchell et al. (1997), in which stakeholders are divided in accordance to the attributes they hold (power, urgency, legitimacy). The more attribute respective stakeholder group possess, the more salient it is to an organisation. Despite, the used classification it is often that not all the stakeholder groups receive the same recognition. In their CSR communication, organisations tend to focus rather on primary/voluntary stakeholders or the stakeholders that possess at least two out of three mentioned above attributes (Mitchell et al., 1997; Hörisch, Freeman & Schaltegger, 2014). The new regulation regime

The European Union regulation, by its attempt of standardizing and pointing out certain subjects that ought to be addressed, somehow seems to level discrepancies in addressing different stakeholder groups by organisations. The Directive 2014/95/EU obligates Large undertakings which are public-interest entities and have an average number of 500 employees during the financial year to include in the management report a non-financial statement containing information about its activity in relation to environmental, social and employee matters, respect for human rights, anti-corruption and bribery matters. Such a report must include an information on: policies pursued by the organisation in relation to these matters, the outcome of such policies, the principal risks related to those matters and the efforts undertaken to manage those risks, organisation’s business model, and non-financial key performance indicators relevant to the particular business (EC, 2014). Therefore, the directive puts a strong emphasis on the reports to include relevant information and appropriate reviews when it comes to both: coverage and data quality (Akisik and Gal, 2014, p. 266). Furthermore, in some measure, it makes sure that the report also addresses issues such as: environment, human rights, anti-corruption and bribery matters which are of importance to secondary stakeholders. On these grounds, the first hypothesis is being proposed:

H1: Reports from 2018 will be more complete when it comes to the number of stakeholder groups addressed.

The ambiguous nature of quality

The concept of quality is of an ambiguous, complex and subjective nature (Leitoniene and Sapkauskiene, 2015). For that reason, assessment of quality requires concise explanation. The existing literature on CSR reporting provides many definitions of the concept of quality. In accordance to Leitone and Sapkauskiene (2015) quality is often understood as good properties of the object, its compliance with certain standards or exceedance. Moreover, some studies derive its definition of quality from the IFRS’s specifications for the quality of financial information disclosure. These specifications include two fundamental characteristics: relevance and faithful representation. Furthermore, IFRS is distinguishing some additional characteristics, such as: comparability, verifiability, timeline and understandability (Leitoniene and Sapkauskiene, 2015). However, existing literature reveals an ongoing discussion whether these characteristic of quality can be also applicable to the disclosure of social information (Boesse and Kumar, 2007; Whittington and Ekara 2013). Beretta and Bozzolan (2008) argue that it is difficult to distinguish between the quantity and quality of information as measures of quantity are often used in assessing the quality. On the other hand, few academics attempted to create their own definition of the quality of information in non-financial reports. For instance, Habek and Wolniak (2015) propose two characteristics for measuring the quality of CSR information, namely credibility and relevance. Whittington and Ekara (2013) developed a similar scale that includes relevance, reliability and comparability. Furthermore, Baviera-Puig at al. (2015) proposed a scale for measuring the quality of CSR reports that consist of 4 clusters: data comprehension, presentation, data quality and reliability; as well as distinguishes twelve sub-clusters, including: communication technique, comparability, clarity and simplicity, coherence, design, explanatory data presentation, ease of finding specific data, completeness/relevance, accuracy, data reliability, stakeholder participation and external audit. Having a closer look at the definitions behind mentioned above concepts, it is noticeable that many authors give different names, or creates different scales, however understand the concept in a similar, if not to say almost identical manner. For instance, Baviera–Puig at al. (2015) presents logic and consistency of the reports’ structure as a definition for coherence, whereas Leitone and Sapkauskiene (2015) see its as an attribute of relevance.

In order to create the definition of quality which will be used in the following chapters of this thesis, all of these definitions have been examined, and the most common, overlapping’ concepts have been selected and included into the 8-points scale. Furthermore,

the special attention has been paid to avoid the inclusion of concepts that could focus more on the quantity aspects than quality. Subsequently, the scale used for the purpose of this thesis consist of two main clusters: readability/representation and data quality, containing four sub-clusters each. The first cluster focuses on measuring the ease of understanding communicated message and takes a closer look not only at the text itself, but also at graphical in-text features that help to navigate and understand the report. It consists of the following sub-clusters: clarity and simplicity, language quality, data visualisations and timeline. The second cluster measures the quality of presented data. It checks its coherence, accuracy, and relevance for a respective stakeholder group measured on two levels: (I) outcomes of stakeholder dialogue as well as (II) goals and plans for future development. Further explanation of the individual concepts is to be find in the methodology section.

The Quality of information in current reporting practices

As Habek and Wolniak (2015) point out nowadays, a growing number of companies is willing to publish a CSR report. However, the quality of the information disclosed varies. Due to the technological development, publishing of CSR reports became simpler, as well as the public interest of social responsibility rapidly increased. As a result of that, the problem of the quality of information provided in such reports have risen. It is often alleged that organisations focus rather on volume than quality. Up until recently, for most of the European companies being sustainable and communicating on their CSR activities was rather a welcome gesture than an obligation. Concerning the the adoption of compulsory reporting, the need for a regulation was widely signalized in the existing literature (Habek and Wolniak, 2015). The most frequent allegations made against voluntary reports were: lack of completeness, accuracy, neutrality, objectivity and comparability (Habek and Wolniak, 2015; Sapkauskiene and Leitoniene, 2015; Beretta and Bozzolan, 2008). The question arises, ‘to what extent will the implementation of the regulation improve current reporting practices?’ Providing the answer may be difficult as there is still a vast information gap in reporting on CSR in Europe. Therefore, it may be that the regulation will not noticeably improve the quality of the reports but rather force the organisations reluctant to disclose non-financial information to publish such a report. In order to test this issue, two following hypotheses are being proposed:

respective stakeholder groups, will prove to be higher in contrast to that of 2017. H3: The difference between quality scores for information disclosed to respective stakeholder groups will be smaller in the reports of 2018 than it was in the reports of 2017.

Up until recently, many countries gave organisations freedom of deciding whether or not to report on non-financial activities. However, this situation seems to be changing and the increased number of countries have begun to implement some kinds of soft regulations, such as ‘comply or explain’. Following the suggestion of Lock and Seele (2016) this research decided to implement four non-European airlines, deriving from two countries in which CSR reporting is not regulated by any means, and two countries –which follows ‘comply or explain’ approach. As a cross-country study has proven, in countries which introduced regulation in regards to communication of CSR (egz. France) the information disclosed is of a higher quality than in countries without the regulation. Following this line of thought, the last hypothesis is being proposed:

H4: The overall quality of information disclosed to stakeholders will improve throughout the years 2017-2018 in the reports published by European airlines, whereas when it comes to the reports of non-European airlines, it will remain the same.

Methodology:

In order to test the presented hypotheses, the following study takes the form of a quantitative content analysis, often recommended in academia as an adequate tool for researching different forms of corporate annual reports (Gray et al., 1996). The subjects of the analysis were all of the European airlines which published either a stand-alone sustainability report or an integrated report respectively in 2017 and 2018, before the 1st of June, 2019. Therefore, the used sample mirrors the population and consists of 16 reports, drawn from eight European carriers (see appendix A). Following the same requirements, four additional airlines, representing respectively: America, Asia, Middle-East and Oceania were randomly selected and included into the dataset. The implementation of non-European airlines allowed for conducting a cross-continental analysis, which contributed to the formulation of a complete

answer to the guiding research question. It provided the grounds for comparing the reporting practices on a bigger scale as well as determining whether the alleged improvement in quality of the disclosed information could be seen as a direct effect of the regulation, or rather a result of world-wide trends in the CSR communication. The total number of 24 reports (12 published in 2017 and 12 published in 2018) were subject to coding.

Codebook and coding procedure:

The used codebook was created especially for the purpose of this research and the grounds for its development are strongly rooted in the existing literature. It contained of two parts (see appendix B). The first part of the codebook consisted of six variables. The first four variables focused on the general information, such as: the name of the organisation, reporting year, country of origin and the number of employees. The remaining two variables aimed at revealing the industry’s tendencies when it comes to a) which stakeholder groups are being addressed (V5) and b) which industry- specific subjects are being included in the reports (V6). The item V5 was coded binary by 0-not present, 1 –present. The item V6 was coded by the 3-point scale, where 0 indicated ‘subject not addressed’, 1 indicated ‘subject partially addressed’ and 2 marked ‘subject fully addressed’. The suggested list of the industry-specific subjects was adapted based on the GRI recommendation for Sustainability Topics for Sectors (see appendix B). Based on the pre-coding and the results of the item V6 the total number of eight stakeholders/stakeholder-related topics was identified:

1 employees / working conditions / work environment

2 government / political connections / policy makers / public administration / corporate citizenship / local development / local economy / anti-bribery and anti-corruption matters 3 customers / service provided / product responsibility

4 investors / financial disclosures / performance disclosures 5 neighbours / noise

6 NGOS / human rights / charity / social help 7 suppliers / contractual partners / airports

8 environment / emissions / climate change / waste / environmental footprint

as topic areas of their interest and aimed at revealing the quality of the communication of these topics. It consisted of eight items formulating the quality scale. The quality scale derives from the existing academic literature and was developed collectively, by two of the Corporate Communication Master Track students. The eight items, were grouped in two categories. The first category took a closer look at the general representation of communicated message. The second category examined the quality of the disclosed data. Almost all of the items were binary-coded, by the 0-1 scale, where 0 indicates ‘not present’, and 1 indicates ‘present’. Only item 1.3 (Language quality) was treated as a continuous variable which original values ranging from 0-100, were transformed into 0 –1 continuous scale.

The process of coding was conducted manually by a single coder familiar with the field.

However, in order to reassure the reliability of the codebook, a second coder was assigned to code 19% (3) of the articles. Then, the Krippendorff's reliability test has been conducted on the overlapping reports. The Krippendorff's Alpha values for each individual quality-scale item are to be found in the appendix (see appendix C).

Subsequently, the results of the coding were subjected to analyses. IBN SPSS Statistics 24 was the software used for conducting all of the analyses. In order to test the hypotheses 1 and 2, paired sample T-test was conducted on the European airlines sample. To give the answer to hypotheses 3, two analyses of variance ANOVA has been run, respectively – on the reports published in 2017 and on the reports published in 2018. Last, but not least, to test hypotheses 4, the paired sample T-Test was conducted which aimed at comparing the quality of the information disclosed to the respective stakeholder groups between the reports delivered by European airlines, and the reports published by non-European carriers. The results of these tests are further explained in the following section.

Results

Quality scale and Descriptive Statistics

To begin the process of hypotheses testing, calculating the quality value was the first step of the analysis. In order to do so, the new variable ‘quality’ was created for each respective stakeholder group. The variable was computed by adding up all of the values scored on 8

components of a quality scale: (1) clarity and simplicity, (2) language quality, (3) data visualisation, (4) timeline, (5) coherence, (6) relevance when it comes to the goals and future development, (7) relevance when it come to the stakeholder dialogue and (8) accuracy. Therefore, the newly created variable was a continues one with values ranging from 1 (low quality) to 8 (high quality). When it comes to the reporting year 2017, the average value of quality of information disclosed to respective stakeholder was 2.92 (SD=.86). The stakeholder groups of whom the quality of the information disclosed achieved the average score higher than the mean were: employees, investors and environment. The stakeholder groups, to which the quality of the disclosed information was below average were: government, customers, neighbours, NGOs and suppliers. When it comes to the reporting year 2018, the average value for quality was 3.67 (SD=1.21). The stakeholder groups, to which the quality of the disclosed information achieved the average score higher than the mean, were identical as in the reporting year 2017.

Following, the airline which scored the highest value for an overall quality of the report in 2017 was Luthansa (M=3.91), and the airline that achieved the lowest value for the overall quality of the report was Wizzair (M=1.38). The airline which scored the highest value for an overall quality of the report in 2018 was, once again, Luthansa (M=5.52), and the airline that achieved the lowest value for the overall quality of the report was Wizzair (M=2.14). The individual quality values for respective stakeholder groups are to be found in the table below.

Table 1: Descriptive Statistics

2017 2018 M SD N M SD N Employees 4.45 1.56 8 5.07 1.67 8 Government 1.38 1.26 8 3.11 2.74 8 Customers 3 2.25 8 3.27 2.5 8 Investors 3.93 2.62 8 4.89 2.33 8 Neighbours 2.43 2.16 8 2.48 2.18 8 NGOs 1.8 1.6 8 2.59 2.4 8 Suppliers .92 1.09 8 1.72 1.67 8 Environment 5.47 1.44 8 6.24 .63 8 Overall quality 2.92 .86 8 3.67 1.21 8

Hypothesis testing

To test the hypothesis 1, paired sample t-test was conducted. The assumptions of the normal distribution, as well as the equality of variance were met. The results of the analysis were statistically significant, therefore proved that the reports of 2018 are more complete when it comes to the number of stakeholder groups addressed p<0.01, t(7)=-1.43, 95%CI [-.99, .25]. However, the size of this effect is rather weak (d=0.29). Furthermore, the analysis of frequencies has shown that the reports improved in addressing wider range of stakeholders only when it comes to 2 stakeholder groups: customers end NGOs. In comparison to reporting year 2017, two of the airlines extended the content of their reports for the information disclosed to customers in 2018 and one of the airlines extended the content of its report for the information disclosed to NGOs in 2018.

Paired sample t-test was also used to test the assumptions of hypothesis 2, stating that: the quality of the reports from 2018, in terms of the information disclosure to respective stakeholder, groups will prove to be higher in contrast to that of 2017. All of the assumptions needed for conducting the analysis were met. The dependent variable was continuous, and the independent variable consisted of two categorical, matched pairs – reports from 2017 and reports from 2018 delivered by the same companies. Moreover, the normal distribution of values was confirmed. The paired-sample T-test has shown that, in general the quality of the reports published in 2018 was higher than the quality of the reports published in 2017. However, the results of the analysis were significant only when it comes to the increase of quality for two stakeholder groups: government p=0.03, t(7)=-2.69, 95%CI [-3.26, -0.21] and suppliers p=0.04, t(7)=-2.5, 95%CI [-1.55, -0.04]. When it comes to addressing government, the quality of the reports increased by 1.73 point on average (M=-1.73, SD=1.82). In case of the information devoted to suppliers, the quality of the reports increased by 0.8 point on average (M=-0.8, SD=0.9). The effect size for these stakeholder groups equalled respectively d= 0.81 and d=0.56. Therefore, it can be assumed that the size of the effect was moderate to large and the hypotheses 2 is confirmed in two cases: the increase of the quality when it comes to the information disclosed to government and the increase of the quality when it comes to the information disclosed to suppliers.

M SD 95% CI t df p Employees -.62 1.22 -1,63 .4 -1.43 7 .2 Government _{-1.73 1.82 -3.26 -.21 -2.7 7 .03} Customers -.26 1.09 -1.17 .64 -.69 7 .5 Investors _{-.95 2.07 -2.68 .78 -1.3 7 .24} Neighbours _{-.05 .11 -0.14 .04 -1.28 7 .24} NGOs _{-.79 1.13 -1.73 .16 -1.98 7 .09} Suppliers _{-.8 .9 -1.55 -.04 -2.5 7 .04} Environment -.78 1.04 -1.64 .09 -2.1 7 .07

In order determine whether the difference between quality scores for information disclosed to the respective stakeholder group got smaller in the new reporting regime (hypothesis 3) the two analyses of variance ANOVA has been conducted. The first analysis compared the quality scores for the information disclosed to respective stakeholder groups in in the year 2017. The second analysis did the same, but for the reporting year – 2018. In both cases, the assumptions of normality, homogeneity of variance, and independence of observations were met. The analysis on the reports published in 2017 proved statistically significant differences in quality scores for information disclosed to respective stakeholder groups F (7,56)=6.03, p<0.001. The size of this effect equalled η²=.43, therefore, it was very large. The Bonferroni’s Post Hoc Test revealed that there is a statistically significant difference when it comes to the quality scores for the information disclosed between the following stakeholder groups:

1. employees and government (Mdifference =3.07, p=.04)

2. employees and suppliers (Mdifference =3.53, p=.01)

3. government and environment (Mdifference =-4.09, p<.01)

4. neighbours and environment (Mdifference =-3.04, p=.05)

5. NGOs and environment (Mdifference =-3.67, p<.01)

SS df MS F p

Between groups 143.24 7 20.46 6.03 <.01

Within groups 189.92 56 3.39

Total 333.16 63

The analysis of the reports published in 2018 also proved the statistical significance of the differences in quality scores for information disclosed to respective stakeholder groups F(7,57)=4.41, p<.01. However, the size of this effect equalled η²=.35 and therefore, was smaller than in the year 2017. The Bonferroni’s Post Hoc Test revealed that there is a statistically significant difference when it comes to the quality scores for information disclosed between the following stakeholder groups:

1. neighbours and environment (Mdifference =-3.77, p=.02)

2. NGOs and environment (Mdifference =-3.65, p=.03)

3. suppliers and environment (Mdifference =-4.52, p<.01)

Table 4: ANOVA for the reports published in 2018

SS df MS F p

Between groups 135.5 7 19.36 4.41 <.01

Within groups 249.96 57 4.39

Figure 1: The difference in quality of information disclosed to the respective stakeholder

groups in 2017 and 2018.

The analysis of variance ANOVA has confirmed the hypothesis. The difference between quality scores for information disclosed to respective stakeholder groups was higher in the reports of 2017 than in the reports of 2018. In contrast to the reports of 2017, in which the statistically significant difference in quality score appeared in 6 cases, the reports of 2018 noted the statistically significant difference only in 3 cases. Moreover, in 2 out of 3 cases the difference in quality scores decreased.

When it comes to the comparison of the quality change on cross-national level (hypotheses 4). 2 paired-sample T-tests were conducted, in order to compare the changes in overall quality of the reports delivered by European airlines and in the reports published by non-European carriers. The results for the European airlines showed the increase of overall quality of the reports (M=-.75, SD=.5) between the year 2017 and 2018. The results were statistically significant p<.01, t(7)=-4.24, 95%CI [-1.17, -.33] and the size of the effect was very strong (d=.71). 0 1 2 3 4 5 6 7

Employees Government Customers Investors Neighbours NGOs Suppliers Environment

The difference in quality of information disclosed to the

respective stakeholder groups in 2017 and 2018

Table 5. Paired sample t-test – quality of the reports published by European airlines M SD 95% CI t df p Overall quality 2017- overall quality 2018 -.75 .5 -1.17 -.33 -4.24 7 <0.01

When it comes to the non-European carriers the results showed a small decrease in the quality of the reports published respectively in 2017 and 2018 (M=.01, SD=.56). However the results were not statistically significant p=0.97, t(3)=.043 95%CI [-.88, .9].

Table 5. Paired sample t-test – quality of the reports published by non-European airlines M SD 95% CI t df p Overall quality 2017- overall quality 2018 .01 .56 -.88 .9 .043 3 .97

Therefore, it can be concluded that hypotheses 4 is confirmed as the overall quality of information disclosed in the reports increased over the years 2017-2018 only in case of European airlines.

Discussion and Conclusions

The conducted analyses confirm the original predictions that the legal regulations of the CSR reporting practices improved the quality of information disclosed towards respective stakeholder groups. However, the definite answer, aiming to assess the extent of that improvement, is open for interpretation, and can be completely dependant on one’s initial expectations.

The analyses revealed that, even though the completeness of the reports when it comes to identifying respective stakeholder groups increased between 2017 and 2018, some tendencies seemed to remain. It has been noticed that organisations, unwittingly modify ones established outline of the reports, and seldom increase their focus on the new stakeholder groups or stakeholder-related subjects. That may be the reason for the relatively low average quality score of the reports published in both years. Despite this fact, the reports from 2018

seem to improve on their quality, when it comes to addressing such stakeholder groups as: government and suppliers. The effect of the improvement of the information disclosed to government was large, whereas the effect of the improvement of the information disclosed to suppliers was of a moderate strength. Taking into consideration that, in the reporting year 2017, the information disclosed to these stakeholder groups received two of the lowest average quality scores, such a result seems to be optimistic for future reporting practices. The reason for this improvements can be potentially tracked to the Directive 2014/95/EU which obligated the airlines to report on anti-corruption and bribery matters, as well as to disclose meticulous information about the supply chain and subcontracting chain. Continuing, in the reports from 2018, the type of a stakeholder group addressed, did not affect the quality of the information score to such extent as it did in 2017. The analysis of the reports published in 2017 showed many discrepancies between the quality of the information disclosed to respective stakeholder groups. The quality of information disclosed to suppliers, government, neighbours and NGOs was significantly lower than in comparison to employees (two cases) and environment (four cases). Nevertheless, the discrepancies in the quality of information disclosed to respective stakeholder groups decline in the reports published in 2018. The number of significant differences in the quality of information disclosed to respective stakeholder groups decreased by three. The types of stakeholder groups for which such discrepancy remained significant were: neighbours, NGOs and suppliers in relation to environment. Even though, some of the discrepancies seem to continue, in two, out of the three cases their value decrease. The strong position of environment, when it comes to both reporting periods, mirrors the initial assumption of airlines’ tendencies to strongly focus their information disclosure on the sustainability matters. Revealing the information on environmental issues, and communicating the actions taken to reduce the environmental footprint, appears to be the most popular way of securing airlines’ legitimate claim.

Such results fit into the narratives of the instrumental stakeholder theory, as they show that organisations tend to focus rather on the interest of their primary stakeholders (Clarkson, 1995). Drawing from the definition that primary stakeholders are the one directly affected or benefitting from the organisations’ activities – suppliers, government, neighbours and NGOs can definitely be considered as secondary stakeholders for the airline industry (Kakabadse, Rozuel & Lee-Davies, 2005). Following this line of argumentation, employees, investors, customers and environment appear to be a primary stakeholder as they, on average, received a fair amount of attention (environment together with employees were the only stakeholder

information disclosed to them is the closest to the mean. Furthermore, these stakeholder groups are the most impactful on achieving organisations’ primary objective – profit. On these grounds it can be concluded that despite the regulation, European airline companies still tend to follow the assumptions of instrumental stakeholder theory, and that the need for recognition of the importance of its normative dimensions remains.

Last but not least, the research has confirmed the increase of the overall quality of the reports between the years 2017 and 2018. This increase was only significant when it comes to the European airline industry. In case of the non-European carriers the analysis showed a slight decrease of an overall quality of the information disclosed. However, the results were not statistically significant. Therefore, future research considering a larger sample is suggested.

This study is not without limitations. Firstly 2018 was the first year in which organisations published information on CSR in the new, obligatory reporting regime. That is why the sample, even though it reflects the population is relatively small, and the presented analysis may provide different results when conducted in the future and on the bigger number of the reports. Secondly, this study investigated the reporting practices of only one sector. Therefore, its finding reflects the changes and tendencies of airline industry only and it cannot be assumed that they will prove to be true for other sectors.

References:

Akisik, O., & Gal, G. (2014). Financial performance and reviews of corporate social responsibility reports. Journal Of Management Control, 25(3-4), 259-288. doi: 10.1007/s00187-014-0198-2

Albu, O., & Wehmeier, S. (2013). Organizational Transparency and Sense-Making: The Case of Northern Rock. Journal Of Public Relations Research, 26(2), 117-133. doi: 10.1080/1062726x.2013.795869

Andriof, J., and S. Waddock (2002) ‘Unfolding Stakeholder Engagement’, in J. Andriof, S. Waddock, B. Husted and S. Sutherland Rahman (eds.), Unfolding Stakeholder Thinking: Theory, Responsibility and Engagement (Sheffield, UK: Greenleaf Publishing): 19-42.

Baviera-Puig, A., Gómez-Navarro, T., García-Melón, M., & García-Martínez, G. (2015). Assessing the Communication Quality of CSR Reports. A Case Study on Four Spanish Food Companies. Sustainability, 7(8), 11010-11031. doi: 10.3390/su70811010

Bellantuono, N., Pontrandolfo, P., & Scozzi, B. (2016). Capturing the Stakeholders’ View in Sustainability Reporting: A Novel Approach. Sustainability, 8(4), 379. doi: 10.3390/su8040379

Beretta, S., Bozzolan, S. (2008). Quality versus Quantity: The Case of Forward - Looking Disclosure. Journal of Accounting, Auditing & Finance, 23, 333-376.

Boesso, G., Kumar, K. (2007). Drivers of corporate voluntary disclosure: A framework and empirical evidence from Italy and the United States. Accounting, Auditing & Accountability

Journal, 20, 269 – 296.

Carroll, A.B. Business and Society: Ethics and Stakeholder Management; South-Western College Pub: Cincinnati, 
OH, USA, 1993. 


Chen, E., & Gavious, I. (2015). Does CSR have different value implications for different shareholders?. Finance Research Letters, 14, 29-35. doi: 10.1016/j.frl.2015.07.001

Cowper-Smith, A., & de Grosbois, D. (2011). The adoption of corporate social responsibility practices in the airline industry. Journal Of Sustainable Tourism, 19(1), 59-77. doi: 10.1080/09669582.2010.498918

Deegan, C. (2002). Introduction: the legitimizing effect of of social and environmental disclosures –a theoretical foundation. Accounting, Auditing and Accountability Journal, vol. 15(3), 282-311.

Donaldson, T.; Preston, L.E. (1995). The Stakeholder Theory of the Corporation: Concepts, Evidence, and 
Implications. Acad. Manag. Review, 20, 65–91.

European Commission, 2010. Europe 2020. A strategy for smart, sustainable and inclusive growth, COM(2010) 2020 final, Brussels.

European Commission – EC (2014). Directive 2014/95/EU of the European Parliament and

of the Council on non-financial disclosure. http://eur-lex.europa.eu/legal- content/EN/TXT/?uri=CELEX%3A32014L0095

Elkington, J. (1998). Cannibals with forks: The triple bottom line of 21st century business. Gabriola Island, Canada: New Society.

Fassin, Y. (2008). SMEs and the fallacy of formalising CSR. Business Ethics: A European

Review, 17(4), 364-378. doi: 10.1111/j.1467-8608.2008.00540.x

Freeman, R. E.: 1984,Strategic Management: A StakeholderApproach. Pittman, Marshfield. Gray, R., Owen, D., and Adams, C. (1996), ‘Accounting and Accountability: Changes and Challenges in Corporate Social and Environmental Reporting, London: Prentice-Hall.

Golob, U., & Bartlett, J. (2007). Communicating about corporate social responsibility: A comparative study of CSR reporting in Australia and Slovenia. Public Relations

Review, 33(1), 1-9. doi: 10.1016/j.pubrev.2006.11.001

June 2019

Hill, C., & Jones, T. (1992). STAKEHOLDER-AGENCY THEORY. Journal Of

Management Studies, 29(2), 131-154. doi: 10.1111/j.1467-6486.1992.tb00657.x

Hillman, A., & Keim, G. (2001). Shareholder value, stakeholder management, and social issues: what's the bottom line?. Strategic Management Journal, 22(2), 125-139. doi: 10.1002/1097-0266(200101)22:2<125::aid-smj150>3.0.co;2-h

Hąbek, P., & Wolniak, R. (2015). Assessing the quality of corporate social responsibility reports: the case of reporting practices in selected European Union member states. Quality &

Quantity, 50(1), 399-420. doi: 10.1007/s11135-014-0155-z

Hörisch, J., Freeman, R., & Schaltegger, S. (2014). Applying Stakeholder Theory in Sustainability Management. Organization & Environment, 27(4), 328-346. doi: 10.1177/1086026614535786

International Air Transport Association (2009).“A global approach to reducing aviation

emissions” November 2009. Available at www.iata.org.

Jones, T., Wicks, A. (1999). Convergent Stakeholder Theory. Academy Of Management

Review, 24(2), 206-221. doi: 10.5465/amr.1999.1893929

Kakabadse, N., Rozuel, C., & Lee-Davies, L. (2005). Corporate social responsibility and stakeholder approach: a conceptual review. International Journal Of Business Governance

And Ethics, 1(4), 277. doi: 10.1504/ijbge.2005.006733

Kim, H., Hur, W., & Yeo, J. (2015). Corporate Brand Trust as a Mediator in the Relationship between Consumer Perception of CSR, Corporate Hypocrisy, and Corporate Reputation. Sustainability, 7(4), 3683-3694. doi: 10.3390/su7043683

Kramer, M. R., & Porter, M. (2011). Creating shared value. Harvard Business Review, 89(1/2), 62-77.

Kuo, T., Kremer, G., Phuong, N., & Hsu, C. (2016). Motivations and barriers for corporate social responsibility reporting: Evidence from the airline industry. Journal Of Air Transport

Management, 57, 184-195. doi: 10.1016/j.jairtraman.2016.08.003

Neu, D., Warsame, & H., Pedwell, K. (1998). Managing public impressions: environmental disclosures in annual reportsa. Account. Organ. Soc, 23, 265-282.

Leitoniene, S. and Sapkauskiene, A. (2015). Quality of Corporate Social Responsibility Information. Procedia - Social and Behavioral Sciences, 213, pp.334-339.

Lock, I., Seele, P. (2016). The credibility of CSR (corporate social responsibility) reports in Europe. Evidence from a quantitative content analysis in 11 countries. Journal of Cleaner

Production, 122, 186- 200.

Ogrean, C. (2017). The Directive 2014/95/EU – Is there a “New” Beginning for CSR in Romania?. Studies In Business And Economics, 12(1), 141-147.

Stieb, J. (2008). Assessing Freeman’s Stakeholder Theory. Journal Of Business Ethics, 87(3), 401-414. doi: 10.1007/s10551-008-9928-4

Szwajkowski, E. (2000). Simplifying the Principles of Stakeholder Management: The Three Most Important Principles. Business & Society, 39(4), 379-396. doi: 10.1177/000765030003900403

Whittington, M., Ekara, A., (2013). Assessment of corporate reporting quality: a review of the literature. European Accounting Association. 36th Annual congress in Paris, France.

Appendix

European airlines

Name Country of Origin Number of employees

Luthansa Germany 129424

KLM/Air France Netherlands/France 36498

IAG Great Britain X

SAS Sweden 10146

Finnair Finland x

Easyjet England 14000

Thomas Cook England 21000

Wizzair Hungary 3686

Non-European airlines

Singapore Singapore 14729

QATAR Qatar 45633

American Airlines USA 107003

Air New Zealand New Zealand 11900

B.Coding sheet

V1 Organisation’s name Open code

V2 Reporting year Open code

V3 Country of origins Open code

V4 Number of employees Open code

V5 V5.1 Employees V5.2 Government V5.3 Customers V5.4 Investors V5.5 Media V5.6 Neighbours V5.7 NGOs V5.8 Science V5.9 Suppliers V5.10 Environment Stakeholder groups addressed Nominal:0-1 V6 V6.1 Carbon offsetting V6.2 Economic direct benefits V6.3 Economic indirect benefits V6.4 Air quality V6.5 Emissions to air V6.6 Emissions to air - GHG emissions V6.7 Noise V6.8 Labour conditions V6.9 Labour management relations V6.10 Cabin personnel health and safety V6.11 Cabin air quality V6.12 Unlawful sex tourism

industry- specific subjects addressed

V6.13 Persons’ with special needs access to services and facilities V6.14 Emergency

preparedness

V6.15 Fleet technological improvement

V6.16 Corporate governance V6.17 Sourcing strategy for aircrafts and components

V6.18 Materials sourcing and use

V6.19 Renewable/alternative energy sources: 0-not present/1-partially present/ 2 – fully present

CONTENT VARIABLES V7 Readability/Representation

V7.1 Clarity and simplicity Nominal 0-1

V7.2 Language quality Nominal 0-1

V7.3 V7.4 V8 Data quality Communication technique/data visualisations Timeline Nominal 0-1 Nominal 0-1 V8.1 V8.2 Coherence Nominal 0-1

V8.3 Relevance – goals and

future development

Relevance – stakeholder dialogue

Nominal 0-1

Accuracy Nominal 0-1

C. Krippendorff's Alpha values for each individual quality-scale item

Krippendorff's Alpha’s for respective for respective items of quality scale

Clarity and simplicity .58

Language quality .73 Communication technique/data visualisations .5 Timeline .65 Coherence .59

Relevance – goals and future development .51 Relevance – stakeholder dialogue .66