Information Value of Sustainability Assurance: Impact of Assurance Provider and Influence of Corporate Reputation

Information Value of Sustainability Assurance: Impact of

Assurance Provider and Influence of Corporate Reputation

Thesis MSc Accountancy

Suradj Soerdjbali

Molenstraat 10

9936EC Farmsum

s.v.soerdjbali@student.rug.nl

Telephone number: +31644978380

Student number: s1903659

Date: 17.08.2020

Word count: 9555

Supervisor: Prof. T.A. Marra

University of Groningen

2

ABSTRACT

As sustainability reporting became the theme of the 21st century, stakeholders interest is also changing

towards the firm’s credibility of the sustainability activities. This study empirically examines the relation of assurance providers and corporate reputation to the information value, as a proxy for quality, of sustainability assurance reports. To improve credibility a voluntary assurance is performed by an independent third party, namely accountancy firms, consultancy firms and engineering firms. This research aims to provide new insights in differences of these assurance providers to the information value of sustainability assurance. Further, this research tests if firms with a good reputation have a higher information value in the sustainability assurance reports. In total 1862 firms from the GRI-database in a 4 year period from 2014-2017 are tested with multiple linear regressions. Also, Fortune’s Most Admired Companies list is used to select firms with a good reputation. The results provide evidence that a good corporate reputation indicates a higher score for information value in the assurance report. Furthermore, evidence is found that engineering firms score the highest and have a positive significant relation to the information value of an assurance report comparing to accountancy firms. For consultancy firms no significant evidence is found to prove that relation. The results indicate that engineering firms and firms with a good reputation have a more complete and informative sustainability assurance report. These findings complement the current CSR literature, with an emphasis on sustainability disclosure and the assurance valuation, which is at a limited but evolving stage.

.

Keywords: Sustainability assurance, Information value, Corporate reputation, Stakeholder theory,

3

Table of Contents

3.3.1 Information value of assurance statement ... 14

3.4 Independent variables ... 15 3.4.1 Assurance provider ... 15 3.4.2 Corporate reputation ... 16 3.5 Control variables ... 16 3.5.1 Firm size ... 16 3.5.2 Profitability ... 16 3.5.3 Industry ... 16 3.5.4 Country ... 17 3.5.5 Years ... 17

4

3.6 Statistical test ... 17

4.1 Results ... 19

4.1.1 Descriptive statistics ... 19

4.1.2. Information value score ... 19

4.2 Correlations ... 21

4.3 Regression Analysis ... 21

4.4 Additional Analysis ... 23

5. Discussion and conclusion ... 25

References ... 27

5

1. Introduction

Nowadays, the world is changing to a more sustainable environment. This is reflected by the increasing numbers of sustainability reports. Besides, according to Gürtürk & Hahn (2016), these reports more often do include an assurance statement (Junior, Best & Cotter, 2014). As Odriozola & Baraibar-Diez (2017) stated, the demands of stakeholders have increased to a level that companies are enforced to make additional efforts to meet stakeholder’s needs. One of the most important efforts to fulfill these needs is voluntary disclosures of sustainability reports which include a voluntary assurance report (Odriozola & Baraibar-Diez, 2017). These assurance statements are provided by an independent third party, mainly an accountancy firm, an engineering firm or a consultancy firm (Global Reporting Initiative, 2013; Al-Shaer & Zaman, 2017). Protecting reputation, gaining credibility and trust with stakeholders are the most important benefits a company obtains from the assurance reports (Briem & Wald, 2018; Junior et al, 2014). In order to obtain these benefits, the company’s management is trying to build a long-term relationship with their stakeholders, a relationship that allows the company to adapt to changes from the public sentiment more quickly (Harjoto & Salas, 2017).

External assurance is defined as “activities designed to result in published conclusions on the quality of the report and the information contained within it” (Global Reporting Initiative, 2013). However, the external assurance of sustainability reports is still in the development stage and does not include obligatory requirements. Sustainability reporting is not regulated and mandatory in the majority of the countries (Junior et al., 2014). Nevertheless, the quality of such an assurance statement is important as the information is used for decision-making and business strategies (Zorio, García-Benau & Sierra, 2013). Although the quality of sustainability assurance has been measured differently in various studies, there is no clear guideline or standard in order to measure quality. I will use information value as a proxy for quality in this study, as it is important for users to be able to understand the information provided in the reports.

There are mainly three types of assurance providers that differ in their expertise, their background and their approach (Perego & Kolk, 2012). Due to these differences, I will test the impact of the type of assurance on the information value provided in the assurance reports. In the current literature, the information value reported by the assurance providers is strongly questioned and criticized by, among other things, independence and knowledge of the providers, but also the content and scope of the

assurance (Boiral, Heras-Saizarbitoria, Brotherton & Bernard, 2019; Kraten, 2019; Gürtürk et al., 2016). Nevertheless, external users rely on the information provided in the assurance statements and can affect their decision making process. The users’ decision making process can also be influenced by the scope of the assurance, which may be a result of the different standards used. The most used standards are the ISAE 3000 and the AA1000AS (Hassan, Elamer, Sobhan & Fletcher, 2019). Besides, the sustainability reports are not all assured completely as companies are able to choose to assure specified sections only (e.g. GHG) (Gürtürk et al., 2016). Moreover, there is a lot of discussion about the different assurance providers in the current literature, of which mostly accountants and non-accountants are mentioned. The big four is already known for their reputation and expertise in assurance, however this does not apply for consultancy and engineering firms. This is confirmed by Hodge, Subramaniam & Stewart (2009) who found in their study that users believe assurance provided by an accountant is more reliable than assurance provided by non-accountants. In addition, Cheng, Green & Ko (2012) reported that non-professional investors are

6

more likely to invest when the assurance of the sustainability report is provided by an accountant rather than by a consultant or engineer. This is also reflected in the price to be paid for an assurance as accountancy firms are generally more expensive than consultancy and engineering firms (Farooq & Villiers 2019). Due to the fact that users have various prejudices about the different assurance providers, the relationship of these providers to the content of the assurance report is questioned in the current literature and hence my first research question is:

“What is the impact of the type of an assurance provider on the information value provided in the assurance report?”

Besides the type of assurance provider, I will investigate the impact of corporate reputation on the information value of the assurance report. From the company's perspective, reputation is an important factor influencing the continuity of the company. If the reputation is harmed, it can impact a company’s long term performance in a negative way (Martín-de Castro, Amores-Salvadó & Navas-López, 2019). Therefore maintaining and retaining the reputation is an important activity for the management. One of the impacts of reputation is the rise of social media, which affects the companies’ control of their marketplace perception. Deloitte (2013) has investigated in their report ‘Exploring Strategic Risk’ that the reputation risk is the biggest risk concern since the last few years. And as a result of the importance of managing the reputational risks, I expect companies with a higher reputation will inform their stakeholders in a thorough manner. In order to measure corporate reputation in this study, I will use the published list from Fortune’s Most Admired Companies. In the academic research for finance and management, this list is the most used proxy for reputation, which is publicly available (Cao, Myers & Omer, 2012).

Many studies already found a positive relationship between reputation and financial performance

(Galbreath, 2012), however in the last few years the focus of stakeholders has changed to the sustainability disclosures (Larrinaga et al., 2020). Therefore the second research question is:

“What is the impact of corporate reputation on the information value provided in the assurance report? “

Business sustainability has emerged as the theme of the 21st century (Ng & Rezaee, 2015). Still, there is little and scarce research on the information value of the assurance report and process (Hahn & Kühnen, 2013; Martínez-Ferrero & García-Sánchez, 2017). The value of these independent assurance reports is questioned in many different papers (Boiral et al., 2017), where Cheng et al. (2012) found that non-professionals are more likely to invest if an assurance is provided by an accountancy firm. Larrinaga et al. (2020) and Hummels, Schlick and Fifka (2017) explored that the credibility not only depends on the assurance company, but also on other actors, like the assuror self. Altogether, there is not yet an unambiguous conclusion about the information value of the different assurance statements. The current research is at a limited but evolving stage (Al-Shaer et al., 2018), to which this study contributes. This paper complements the current CSR literature, with an emphasis on sustainability disclosure and the assurance valuation. Also, the evidence in this study can be used to understand the role of corporate reputation in the choice of assurance providers. As Omar and Williams (2005) mentioned, most of the prior research for corporate reputation is to relate reputation to markets, pricing and debt & equity financing activities, but the effect on reporting and assurance quality is very limited. The focus of this paper is to provide a conclusion of differences in information value disclosed in the assurance statements

7

of a sustainability report and beside that, also to examine whether and how corporate reputation influences the disclosed information value for the stakeholders. These relationships are investigated on the basis of stakeholder, legitimacy and signaling theory, which are based on proving the credibility and transparency to the stakeholders (Junior et al., 2014; Martïnez-Ferrero & Garcia-Sanchez., 2016; Briem et al., 2018) The remainder of this paper is organized as follows: in the next section the theoretical framework will be presented for the relationship with the information value of sustainability assurance, and also the

development of the testable hypotheses will be elaborated. The third section will describe the research method used for this study. Following, the fourth section will reveal and present the empirical results of the hypotheses. Finally, the last and fifth section will discuss the findings and clarify limitations for this study.

2. Theoretical Framework 2.1 Theory

The objective of this study is to explore a way to improve the information value by discovering what the effects are of the types of assurance providers and corporate reputation on the information value provided in the sustainability assurance reports. The relationships between these variables can be explained by relying on the stakeholder, legitimacy and signaling theory.

2.1.1 Stakeholder Theory

First of all, the disclosure and assurance over the sustainability activities are on a voluntary basis. In order to answer the question why reputation could affect the information value in the assurance report, an understanding of the stakeholder theory is important. According to Casey & Grenier (2015), the stakeholder theory defines that the firm’s key stakeholders need to be satisfied. This indicates that managers should change the focus from the shareholders, to the stakeholders. By voluntary disclose and assure the sustainability reports, a special focus to the stakeholders is created. By doing this, the firm invests in their stakeholders by voluntarily giving information and being transparent in their behavior. The reason firms do so is because stakeholders can be seen as a foundation of competitive advantages to create and maintain relations for an extended period of time (Hillman & Keim, 2001 ; Martín-de Castro et al., 2019). Another definition for the stakeholder theory is that firms are responsible to protect stakeholders rights, by also helping them in decision making (An, Davy & Eggleton, 2011). With voluntary assurance over the sustainability reports the managers provide the stakeholders with additional information which can be used for decision making. Zhou et al. (2016) emphasizes the importance of meeting stakeholders’ interest to be successful in their long term performance. The pressure from stakeholders will impact the reputation of firms, if they are not able to meet the informational needs of the stakeholders. Based on this, if firms attach value to their reputation they are more likely to disclose and assure their sustainability report. Firms are more likely to request their assurance provider to assure the entire report, to include meaningful findings, conclusions and/or recommendations in their assurance report. By doing so, the risk of not fulfilling stakeholder’s needs is mitigated.

8

2.1.2 Legitimacy theory

Also, a well-known theory in the current literature for explaining the relationship of reputation and assurance providers with the information value provided in the sustainability assurance is the legitimacy theory. This theory describes the way individuals, but also firms are aiming to act in accordance with social values to ‘legitimize’ their work. For example the sustainability disclosure decisions that favors society (Deegan, 2002). According to Martinez-Ferrero et al. (2016) and Kolk & Perego (2010), the increasing credibility of sustainability information is the main demand for voluntary assurance, because it represents the constructive dialogue between the firm and their stakeholders which is in line with the legitimacy theory. Boiral et al. (2017) discussed the role of an assurance provider, where the reports reflect the companies’ interest more instead of stakeholders’ interest. The management of a company could prefer a certain scope to publish for reputational impact, where many of the assurance reports do not cover the entire sustainability disclosures, but a certain or specific section of the disclosures (Gürtürk et al., 2016). Companies are able to request their assurance provider to assure only categories in which they excel in and therefore act in favor of their interest instead of the stakeholders’ interest. The legitimacy theory can be used to explain the choice of an assurance provider. According to Martinez-Fererro et al. (2018) consultancy and engineer firms, with the use of AA1000AS, refer more to an organizations’ total performance in contrast to accountancy firms where more specified sections are assured which impact the information value of the assurance report. To mitigate this difference, the scope of the assurance should be the entire sustainability report, since the assurance of sustainability reports are improving the

legitimacy of the disclosed information and stakeholders should have a full understanding of the entire firm, instead of small parts (Boiral et al., 2017; Simnett, Van Straelen & Chua, 2009). From the assurors point of view, their reputation, expertise and background could explain the scope and standards they use to report their findings. According to Briem et al. (2018) the perceived reliability from the stakeholders is already higher if a big four firm assured the sustainability report. The assurance of sustainability reports should decrease the expectation gap from the investors (stakeholders) to the firms’ sustainability issues and continuity.

In addition, according to Martín- de Castro et al. (2019) legitimacy and corporate reputation are also closely connected to each other, as legitimacy is to prove quality and truth under the laws, rules and social values. Just as corporate reputation, Martín- de Castro et al. (2019) relates organizational legitimacy to the impression the society has of a firms’ behavior. That is because legitimacy and corporate reputation both are approved or disapproved perceptions of a firm’s behavior (King & Whetten, 2008). According to the legitimacy theory managers legitimize sustainability reporting by adopting assurance processes for the key stakeholders in general (O’Dwyer, Owen & Unerman, 2011). It is therefore conceivable that firms that attach great value to their reputation are stakeholder-oriented and therefore are willing to increase the scope of the assurance in such a way to ensure transparency and validate their legitimacy. Hence, a higher information value of the sustainability assurance reports is provided as the central element of the

legitimacy theory involves adhering expectations of society’s norms and values and thus, stakeholders needs (Deephouse & Carter, 2005).

9

2.1.3 Signaling theory

Lastly, the signaling theory is elaborated in order to understand the firm’s choice for an assurance

provider, their reputation and the relation with the information value of the assurance. This theory assumes that disclosing information affects the support a firm receives from its stakeholders by revealing the firm’s true performance, such as sustainability performance (Touboul & Kozan, 2020). However, the signaling theory also assumes when a firm is able to disclose information in favor of their activities, they are more likely to do so immediately in order to improve and remain their trust and confidence of the stakeholders as well as the firm’s transparency, reputation, credibility and accountability (Hassan et al., 2019).

The public, and thus stakeholders, are increasingly aware of the firm’s sustainability components, which also increases pressure on the management of the firms to share this information, for example in the form of a sustainability report (Gürtürk et al., 2016). The transparency to stakeholders of valuable information plays an important role, and firms are increasingly involving a third party to require credibility for this disclosed information by having an assurance performed (Junior et al., 2014; GRI, 2013). This is in line with the signaling theory to also disclose an assurance statement from a third party, which is not

mandatory, but it does send a positive signal to the stakeholders since the perceived credibility increases (Gürtürk et al., 2016; Briem et al., 2018; Martínez-Ferrero & Garcia-Sanchez., 2018). Signaling theory argues that stakeholders actually are aware about the firm’s behavior and actions regarding sustainability issues, because of the presence of an assurance statement which test the firm’s actions (Clarkson, Richardson & Vasvari, 2011).

Moreover, signaling theory can also be used to link the corporate reputation to the information value of the assurance statements. As Omar et al. (2005) stated in their study, corporate reputation is used to signal a firm’s characteristics for their social status, which is an outcome of the firm’s competitive processes and behavior. According to Galbreath (2012), sustainability disclosures and reputation are positively linked to each other and to maintain this reputation status, the socially responsible behavior should maintain or improve too. That is because the firm’s behavior is responsible for the judgments a stakeholder has, which is fundamentally the foundation of a firm’s reputation. That is the representation of how the public’s opinion is, which is important for firms for meeting the expectations of the public (Galbreath, 2012). Disclosing high level sustainability information will signal the stakeholder in meeting their expectations and thus influence the firm’s reputation positively (Brammer & Pavelin, 2006). Therefore firms that put great value their reputation are more likely to request their assurance provider to assure the entire report and to include meaningful findings, conclusions and/or recommendations in their assurance report in order to signal their stakeholders. By doing so, firms are trying to maintain or improve their reputation by showing their transparency.

2.2 Hypotheses Development

This section will elaborate and define the terms of the research questions. In addition, the hypotheses will be developed regarding multiple variables that affect the information value of sustainability assurance.

10

2.2.1 Sustainability assurance

In prior research, studies underlined the heterogeneous content of voluntary assurance statements. The reason for this is the current standards are vague and there are no existing regulations to determine the quality (Gürtürk et al., 2016). Quality is relatively a large concept and difficult to measure. Therefore, as a proxy for quality I will investigate the information value of the assurance measured by the model used by Gürtürk et al. (2016), which will be explained in the method section. The information disclosed in the assurance statement is for great importance to the stakeholders, especially for the investors since the information can affect the investors’ decisions (Zoiro et al., 2013; Dilla, Janvrin, Perkins & Raschke, 2019).

The information disclosed in the assurance statements can be prepared by three different independent parties, namely the accountancy firms, engineering firms and consultancy firms (GRI, 2013; Al-Shaer et al., 2017). All these firms have their own specialties in conducting the assurance. As Martinez-Ferrero et al. (2016) mentioned in their study, engineering firms and consultancy firms have better substantive expertise for sustainability issues. Besides, the statements provided by the aforementioned firms contain recommendations and advice for current processes and systems that can be improved in the future, which is mostly not included by accountancy firms (Deegan et al., 2006). Hasan, Roebuck & Simnett (2003) therefore state that the assurance statements of engineering and consultancy firms are more clear and complete, but also more informative. Also, engineering and consultancy firms are more involved in sustainability issues, partly because they have more specific knowledge and believed to have climate change expertise (Datt et al., 2019 ). Accountancy firms on the other hand are professional auditors and are worldwide known for their independency that are paid to verify numbers and follow standards, laws and regulations more narrowly, but their core business is still the audit of annual reports. Moreover, consultancy and engineering firms have more technical expertise to support the management and can help improve and resolve sustainability activities (Datt et al., 2019; Al-Shaer et al., 2019 & Martinez-Ferrero et al., 2016). This suggests accountants have more general knowledge versus consultants and engineers with specific knowledge. Rossi et al. (2017) indicate in their study that the assurance statements from big 4 accountants therefore are incomplete. In addition, the dataset used for this study shows that the scope of the assurance at accountancy firms was done in 32.5% of all assurance reports for the entire sustainability report. For consultancy and engineering firms this is a lot higher, 53.5% and 55.9% respectively.

Besides the aforementioned differences, the assurance firms also follow their own standards to perform the assurance of the sustainability reports. The difference in methods and formats variety of the assurance is partly because there is no generally accepted approach for sustainability assurance (Perego et al., 2012), therefore different standards can be used. The two most professional standards used for assurance are the Accountability Assurance Standard (AA1000AS) and the International Standard on Assurance

Engagements (ISAE 3000) (Hassan et al., 2019). These assurance standards both have a different way to express the level of the assurance, where ISAE 3000 expresses ‘reasonable’ or ‘limited’ and AA100AS expresses a ‘high’ or ‘moderate’ assurance level (Martínez-Ferrero et al., 2016). Moreover, Hassan et al. (2019) mentioned, that because of this non-regulated approach, it can be discussed that the current assurance is missing specificity, robustness and transparency. Nevertheless, Gürtürk et al. (2016) found that an approach including the AA1000 standard has a wider and more diverse use of methods and scores better on the information value the assurance contains. This is quite interesting, since the use of the

11

ISAE3000 is increasing and gaining more popularity over the AA1000 standard, which therefore seems to influence the procedures during the assurance process (Gürtürk et al., 2016). According Deegan et al. (2006), Briem et al. (2018) and Martínez-Ferrero et al. (2018) the ISAE3000 is more of a financially used standard, which is therefore mostly used by accountancy firms, and the AA1000AS is mostly used by non-accountants, e.g. engineering firms and consultancy firms. In contrast, we note in the literature that the ISAE 3000 highlights the limitations of corporate reporting, and that the use of the AA1000AS more focuses on the strengths and weaknesses in the statements (Briem et al., 2018). According to Stones (2014) both standards have a significant different approach:

“ISAE 3000 is issued by public accountants firms and stresses the importance of data quality, reporting procedures, controls and evidence gathering procedures. AA 1000 AS is generally used by specialist bodies and emphasizes the need for organizations to demonstrate effective stakeholder engagement, the identification of material sustainability issues and the existence of a responsible business strategy which is aligned to the identified issues. AA1000AS is purpose-built and developed by sustainability professionals in response to stakeholder concerns about narrow focus of the ISAE standard and the limited effectiveness of assurance statements that are written without meaningful findings, conclusions and/or

recommendations resulting from the assurance process. AA1000AS focus on materiality- making sure the organization is reporting on what is important”

Due to these differences I assume the following hypothesis:

H1: The information value of the assurance statements is higher if the assurance is provided by an engineer or consultancy firm.

2.2.2 Corporate reputation

In the current literature corporate reputation has many different definitions. Michelon (2011) referenced reputation to strategic management and to sociological perspectives. Corporate reputation can be defined as an organizational feature that mirrors the range to which stakeholders see the firm as a good corporate citizen, which is a collective valuation of the reliability and trustworthiness (Michelon, 2011). According to Omar et al. (2015) reputation has four core elements, namely 1.objective specific elements, which states if a firm is well known, 2. Net effective or emotional reactions, which are based on the expectations to which a company should perform according to the stakeholders, 3.Past actions, where reputation is

influenced by the economic and non-economic activities that a company has been involved in the past, and finally 4. Information cues, which is the result of direct and indirect information received and what people directly and indirectly experience. In this study these four elements are leading in maintaining and/or improving the current reputational status. Reputation is also an important driver for social reporting. Reputation increases the awareness of companies about the need to manage a wide range of

environmental, social and ethical risks and to demonstrate externally that they are actually doing it (Friedman & Miles, 2001). According to the article by Boesso & Kumar (2007), voluntary publication of social and environmental information is not limited to satisfying the information needs of investors, but rather a tool to manage broad relationships with stakeholders. The most important effort to fulfill these needs and manage the relationships with stakeholders are voluntary disclosures of sustainability reports which includes a voluntary assurance report (Odriozola & Baraibar-Diez, 2017). These efforts are needed

12

to protect and maintain reputation and to gain credibility and trust from the stakeholders (Briem & Wald, 2018; Junior et al, 2014). Datt et al. (2019) discusses that managers make strategic choices with regard to the size and scope of the assurance in order to achieve certain CSR goals. For companies with a positive reputation, this means that they exert influence to share the information in the assurance as transparently and extensively as possible with the stakeholders because they do not want to harm their reputation and want to maintain or improve their reputation. To test the relation of corporate reputation to the

sustainability assurance, hypothesis 2 is developed:

H2: Firms with a positive corporate reputation have a higher information value in their sustainability assurance.

Figure 1 Conceptual Model

+/-

+/-

Information value of the sustainability assurance report

Corporate Reputation Assurance Provider

13

3. Methodology

3.1 Research approach

In this research paper, I will explore the relationship of assurance providers and corporate reputation to the information value of the sustainability assurance. In prior research, many researchers measure the relation according to the quality of an independent assurance, where quality is measured in many different ways. In this study, I will concentrate on the information value of the disclosed assurance statement, based on the Sustainability Assurance Index from Gürtürk et al. (2016). The official model of the Sustainability Assurance Index was proposed by Perego et al. (2012), where 19 worldwide accepted scoring criteria were used to define the assurance statement quality. This index criteria is expanded by Gürtürk et al. (2016) with recommendations, progression, reservations and examined work. By using this expansion, the content of the assurance can be assessed, but also an indication of the information value of the entire assurance, which enables me to quantify the information disclosed in the assurance statements.

Further, this section will describe the methods used for data collection, which statistical models are used and how validity is guaranteed. This chapter is structured as follows: first I will describe the sample I have used. Secondly, a brief description of the variables used in this study will follow. The last part will

describe the statistical models used for the analysis in this study.

3.2 Sample

My sample contains archival data with 1862 observations in a 4-year period, from fiscal year 2014 through 2017, of worldwide firms adopted in the GRI-database. From this sample only firms with an assurance on their sustainability activities are included. Not every firm has data for each year available in my sample. I especially use firms in the GRI-database, because of the GRI standards, which is the first standard accepted and used worldwide for sustainability reporting (GRI, 2013). The assurances therefore are based on the same environmental standard used for the sustainability disclosures. For the information value of the assurance statements and information about the assurance providers I have hand collected the data together with a group of bachelor and other master students according to the model of Gürturk et al. (2016), where we manually scored the assurance statements.

Furthermore I will use Fortune’s Most Admired Companies to measure corporate reputation. The firms included in my sample are the firms in the Fortune’s list from fiscal year 2014 thorough 2017. For the validity of this sample, I have compared this list with The Newsweek Green Ranking, where

environmental reputation is measured. The Newsweek Green Ranking is used worldwide, especially in the media sector (Cho, Guidry & Hageman, 2012; Neumayr, 2016). The Fortune’s list provides a total of 544 firms to be top ranked in reputation in the years 2014-2017. Yet, not every firm is included in the GRI database or has an assurance performed for their sustainability disclosures. Of these 544 firms, a total of 437 do not have the information in the GRI database needed for this sample. Another 10 firms are

excluded because of the comparison with the Newsweek Green Ranking. This leaves 97firms (appendix

1) with a total of 254 assurance statements in my sample with a high reputational score. As shown in table 1 there are a total of 1862 assurance reports included in this sample divided in 4 years. 254 assurance

14

reports of this sample are related to firms with a positive reputation according to Fortune’s Most Admired Companies list. 1608 reports are not included in Fortune’s Most Admired Companies list.

As last, the control variables are measured using data from the ASSET-4, which is added to the list of firms I use from the GRI-database. The firms that are missing data needed for this study are already eliminated from the sample.

Table 1. Sample firms

Year 2014 2015 2016 2017 Total Assurance Reports GRI Firms 355 425 444 384 1608 Assurance Reports Reputation Firms 52 74 65 63 254 Total GRI Assurance Reports 407 499 509 447 1862 3.3 Dependent variable

3.3.1. Information value of assurance statement

To measure the information value of the sustainability assurance, I will use the sustainability

assurance index used by Gürtürk et al. (2016). In this index the assurance statements are scored based on 23 scoring criteria, where points can be scored for each part. The potential maximum score is 33, and how these points are awarded can be seen in Appendix 2. The more points an assu rance statement has scored, the higher the information value in that statement is. According to Gürturk et al. (2016), this coding scheme can demonstrate differences and similarities, but also the content and transparency of the assurance. In table 2 can be seen that although a score of 33 can be reached, the actual scores are relatively low: between 14 and 21.This indicates that the overall quality is relatively low.

15

Table 2. Information value frequency

3.4 Independent variables

3.4.1 Assurance provider

The provider of the assurance in this study is the independent variable, which distinguishes between three different providers: 1. Accountancy firms, 2. Consultancy firms and 3. Engineering firms. I have added three dummy variables for respectively Accountancy, Consultancy and Engineering firms. All these variables are tested separately in the regression section. The main used standards these firms use are ISAE 3000 for accountants and AA1000AS for consulting and engineering firms. These assurance types are not measured separately, since the type of provider already suggests that accountancy firms use ISAE 3000 and engineering firms and consultancy firms use AA1000AS. As AA1000AS is generally used by consultancy firms and engineering firms that emphasizes the need for organizations to demonstrate effective stakeholder engagement, the identification of material sustainability issues and the existence of a responsible business strategy which is aligned to the identified issues, together with their expertise in sustainability matters I assume that assurance providers who use this standard are more likely to provide more information in their assurance report. Both assurance providers and the standard they use are included in the GRI Data list and hand collected data.

Information

Value Freq. Percent Cum.

Information

Value Freq. Percent Cum.

0 4 0.21 0.21 16 179 9.61 42.43 1 1 0.05 0.27 17 168 9.02 51.45 2 2 0.11 0.38 18 131 7.04 58.49 3 4 0.21 0.59 19 139 7.47 65.95 4 1 0.05 0.64 20 154 8.27 74.22 5 6 0.32 0.97 21 109 5.85 80.08 6 7 0.38 1.34 22 91 4.89 84.96 7 7 0.38 1.72 23 83 4.46 89.42 8 8 0.43 2.15 24 65 3.49 92.91 9 21 1.13 3.28 25 48 2.58 95.49 10 22 1.18 4.46 26 32 1.72 97.21 11 41 2.20 6.66 27 21 1.13 98.34 12 54 2.90 9.56 28 21 1.13 99.46 13 85 4.56 14.12 29 3 0.16 99.62 14 161 8.65 22.77 30 7 0.38 100.00 15 187 10.04 32.81 Total 1,862 100.00

16

3.4.2 Reputation

In this study reputation is measured by coding the reputational firms and the non-reputational firms. The firms with a good corporate reputation are mentioned in the list of Fortune’s World's Most Admired Companies from 2014 till 2017. Therefore a dummy variable is created. If a firm has a good corporate reputation the dummy variable takes value 1 and firms which are not mentioned in this list will take value 0. In Fortune’s list, the reputation is measured according to 9 key attributes to define reputation: 1. Innovation, 2. People management, 3. Use of corporate assets, 4. Social Responsibility, 5. Quality of management, 6. Financial soundness, 7. Long-term investment value, 8 Quality of products/services and as last 9. Global competitiveness. These 9 key attributes are annually assessed by over 8000 financial analysts, senior executives and outside directors, who each rates 4 to 10 firms in their specialized industry (Omar et al.,2012). Their scores result in a World’s Most Admired Companies list, which annually

contains a list of approximately the top 300 firms. The listed firms from 2014-2017 are used in my sample.

3.5 Control variables

Not only the independent variables do have an impact on the information value in the assurance statements, but there are other variables which could impact the dependent variable. To control these effects on the information value of the assurance statement I will include the following control variables used are size, profitability, industry, country and year. These variables are not the primary focus in this research, but are taken into account for their impact on the independent variable.

3.5.1 Firm size

First of all I will use the size of a firm, since prior studies find a positive relationship between the firm size and assurance (Simnett et al., 2009; Hassan et al., 2019). Stakeholders also exert more pressure on the firm if they are larger (Datt, Luo & Tang, 2019), and larger firms also have a higher community visibility and are politically more sensitive (Kang & Gray, 2011). Dat et al. (2019) also mentioned that larger companies are expected to show the world their behavior and activities are legitimate, as they are more likely to take additional measures against climate change. Size is measured as a logarithm of total assets (Martinez-Ferreiro et al., 2018).

3.5.2 Profitability

Secondly, I will take profitability into account. Profitability is a variable that could impact a providers choice and if an assurance should be performed. According to Datt et al. (2019), profitable firms have more capacity to have an assurance performed, but also the size of the assurance and the provider choice, because they have more money to spend. Firms who are more profitable are more likely to request their assurance provider to assure the entire report, to include meaningful findings, conclusions and/or recommendations in their assurance report. Profitability is measured by dividing net income by total assets.

3.5.3 Industry

Industry is also an important control variable to take into account. The firms’ industry may face different social, but also political pressures for sustainable disclosures and assurance procedures. Since in some industries it is unavoidable or harder to meet environmental measures than in others, but a violation can directly impact a firm’s reputation (Zou, Zeng & Shi, 2015; Martín- de Castro et al. 2020). Cho et al.

17

(2012) already found a relation between the type of industry and environmental reputation. Some

industries are more often exposed to environmental and social matters and could possibly have more of an incentive to assure their sustainability reports (Van Tendeloo & Vanstraelen, 2008). To take industry into account as a control variable I will distinguish two different groups: higher pollution industries and lower pollution industries defined by Simnett et al. (2009). The higher pollution industries include 1.Basic Materials, 2.Financials, 3.Oil & Gas and 4.Utilities. The lower pollution industries include 1.Consumer Goods, 2.Consumer Services, 3.Health Care, 4.Industrials, 5.Technology and 6.Telecommunications. These industries are classified according to the Asset4 data set and will take a 1 if it measures the higher pollution industries and otherwise a 0.

3.5.4 Country

Political and legal influences could affect different sustainability issues, including the assurance of these matters. Also, a difference is expected between developed countries and less developed countries. For example, it is expected that there will be a more extensive disclosure on sustainability information in a developed country compared to less developed countries, which could impact the information disclosed and assured. Therefor this study also controls for country effects. In the sample firms of 53 countries are included. For each country a dummy variable is added which takes 1 if the firm is located in the country of the dummy variable, otherwise the value will be 0.

3.5.5 Years

Lastly, years are used as a control variable. According to Cao et al. (2012), economic, business and legal environmental issues can differ over time. For example due to regulations or firm performance, that can influence the choice to disclose and assure more or less sustainability parts if a firm can afford it, but also the choice for assurance providers which differ in price for assurance activities. For each year a dummy variable is included which takes value 1 if the year equals the year of the sustainability report.

3.6 Statistical Test

In this paper I will examine the relationship between multiple variables. Therefore I will test my hypothesis performing a multiple linear regression. The following regression model will include the expectations I have of the effect on the information value of the assurance statements. I will test all of the three components of my hypotheses by using three multiple linear regression models.

IV = 𝛼x +β1-3 (Prov) +β4 (Ctrl_Size) + β5 (Ctrl_ROA) + β6 (Ctrl_IND) + β7 (Ctrl_Country) +

β8 (Ctrl_Year) + 𝜀x

IV = 𝛼x +β1 (Rep) + β2 (Ctrl_Size) + β3 (Ctrl_ROA) + β4 (Ctrl_IND) ++ β7 (Ctrl_Country) +

β8 (Ctrl_Year) + 𝜀x

𝛼x = 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡

β= 𝑠𝑙𝑜𝑝𝑒

18

Table 3. Variables

Variable Description of measurement Label

Dependent variable

Information value of sustainability assurance

The total score, resulting from the Sustainability Assurance Index

IV

Independent variables

Assurance provider As a dummy variable, where the firms

receive value 1 or 0

βi (Accountancy)

βi (Consultancy)

βi (Engineering)

Corporate reputation As a dummy variable, where firms

listed in Fortune’s Most Admired Company’s receive value 1 and others value 0.

βi (Rep)

Control variables

Size As a total assets logarithm βi (Ctrl_Size)

Profitability Net income/total assets βi (Ctrl_ROA)

Industry

Country Year

Dummy variable for higher pollutions firms is 1 and lower pollution firms is 0

Dummy variable for each country Dummy variable for each year from 2014 till 2017.

βi (Ctrl_IND)

βi (Ctrl_Country)

19

4. Results

After performing the statistical analyses, this section will provide the results of the analysis. The first part will elaborate the descriptive statistics. Secondly, with the use of Pearson’s correlation coefficient, the correlation matrix is displayed and discussed. After that, the results of the regression analysis will be displayed, followed by additional analyses.

4.1.1 Descriptive statistics

A summary of the descriptive statistics for the used variables in this study can be found in table 4.1. This table describes the total observations, standard deviation, minimum and maximum of the used variables. The mean of IV for all observations is 17.68, which indicates that in this sample an average of 17.68 points is scored in the assurance statements according to the model of Gürtürk et al (2016). 64,9% of the assurance reports in the dataset are reports from Accountancy firms, 23,5% from Engineering firms and 11,6% from Consultancy firms. 13,6% of all reports are included in the list of companies with a good reputation. Moreover, the size of the firms included in my sample is measured by a natural logarithm of total assets. In the descriptive statistics a range of 12.496-27.123 is measured, which indicates that there is a large variance of a firms size. Additionally, the profitability is measured by calculating the firms Return on Assets, which also variate widely in my sample. From all the firms the lowest ROA is -9.8% and the highest ROA is 9,98% which shows a wide spread of profitability. The independent variables and remaining control variables are measured as a dummy variable. As last, 47,8% of the firms operate in polluting industries.

Table 4.1 Descriptive statistics Variable Obs Mean

Std.

Dev. Min Max

IV 1,862 17.680 4.529 0 30 Accountancy 1,862 0.649 0.477 0 1 Consultancy 1,862 0.116 0.320 0 1 Engineering 1,862 0.235 0.424 0 1 Rep 1,862 0.136 0.343 0 1 Ctrl_Size 1,862 18.616 2.728 12.496 27.123 Ctrl_ROA 1,862 4.009 3.574 -9.800 9.980 Ctrl_IND 1,862 0.478 0.499 0 1

4.1.2. Information value score

The most common assurance provider in the dataset is an accountancy firm (N=1209), the second common assurance provider is an engineering firm (N=438) and the least common provider is a consultancy firm (N=215). In table 4.2 and table 4.3 the results of the T-Test are shown, to test for the significance between the mean values of Information Value. Accountancy firms scored the lowest average (17.30 points) in the sustainability assurance index, followed by the consultancy firms (17.52 points) and the engineering firms (18.80). In comparison with the scores of Gürtürk et al. (2016), the accountancy firms score higher (their sample scored an average of 16.75 points), but the non-accountancy firms (19 points their sample) score lower. This could be due to the limited size of the sample they use (N=61) and the older year (2013) of

20

their sample. Nevertheless, the results in table 4.2 and 4.3 suggests that only a significant relation (at 0.01 level) is found for the information value from engineering firms in relation to accountancy firms, and no significant relation is found in for the information value from consultancy firms in relation to accountancy firms.

Table 4.2 T-Test difference in mean Information Value Accountant-Engineering

Table 4.3 T-Test difference in mean Information value Accountant-Consultancy Obs. Mean IV Std. Err. Std. Dev.

Accountancy 1209 17.299 .120 4.166

Consultancy 215 17.535 .372 5.456

Difference -0.236

Significance 0.467

Obs. Mean IV Std. Err. Std. Dev.

Accountancy 1209 17.299 .120 4.166

Engineering 438 18.804 .230 4.815

Difference -1.505

21

4.2 Correlations

In table 5 the correlations between the used dependent, independent and control variables are identified by using Pearson’s Correlation Coefficient. By calculating the correlation coefficient multiple significant correlations has been found. If the correlation coefficient exceeds 0.8 or lies below -0.8 it will indicate multicollinearity (Blumberg, Cooper and Schindler, 2014). If multicollinearity occurs, it could impact the predictions of the regression unintentionally. Table 5 shows that there aren’t any variables with a

correlation above 0.8 or below -0.8, hence no multicollinearity is detected.

Table 5 Pearson correlation table

(1) (2) (3) (4) (5) (6) (7) (8) (1) IV 1.000 (2) Accountancy -0.115*** 1.000 (3) Consultancy -0.012 -0.492*** 1.000 (4) Engineering 0.138*** -0.755*** -0.200*** 1.000 (5) Rep 0.081*** -0.006 -0.031 0.030 1.000 (6) Ctrl_Size 0.112*** -0.135*** 0.092*** 0.083*** 0.088*** 1.000 (7) Ctrl_ROA -0.032 -0.021 0.012 0.014 -0.004 -0.190*** 1.000 (8) Ctrl_IND -0.019 0.068*** 0.028 -0.097*** -0.139*** 0.141*** -0.291*** 1.000 4.3 Regression Analysis

In this section the hypotheses will be tested using a linear regression for exploring if assurance provider and reputation are related to the information value of the sustainability assurance statement. Table 6 provides the regression results with 4 different models. Besides the independent and dependent variables, all models will include the control variables. Model 1 represents the relation between the dependent variable Information Value and independent variable Accountancy firm. In the results can be seen that there is a negative significant (at 0.05 level) relation between the Accountancy firm and the Information Value of the assurance statements. Model 2 represents the relation between dependent variable

Information Value and independent variable Consultancy firm, where a negative significant (at 0.05 level) relation is revealed. Model 3 represents the relation between the dependent variable Information Value and independent variable Engineering firm. This model shows a positive significant (at 0.05 level) relation with the information value of the assurance statement. The results of these three models are partly in line with hypothesis 1, since it indicate that accountancy firms have a negative relation to the information value of the assurance reports and engineering firms have a positive relation to the information value.For consultancy firms a negative significant relation is found, which is not in line with the expectations. For

22

the control variables, no significant relation is found. Next, model 4 will represent the relation between the dependent variable Information Value and the independent variable Reputation. This model will test the second hypothesis, where can be seen that there is a positive significant relation between the firm’s reputation and the information value. This relation is also in line with hypothesis 2, which indicates that firms with a higher reputation have a higher information value in their assurance report. Besides, also a significant relation is found with the size of the firm in model 4. Furthermore, the Variance Inflation Factor (VIF) is taken into account. When control variables for years and countries are added to the models, the multi-collinearity increases. The highest VIF value is 4.28, which results in no critical values and thus no multi-collinearity, since the VIF values remain under 10. Further, the table shows that all independent variables have a significant relation with the dependent variable.

Table 6. Regression Results

Model 1 Model 2 Model 3

IV Coef.

Std.

Err. Coef. Std. Err. Coef. Std. Err.

Accountancy -0.205** 0.247 - - - - Consultancy - - -0.710** 0.318 - - Engineering - - - - 0.706** 0.260 Ctrl_Size 0.017 0.069 0.002 .069 0.025 0.069 Ctrl_ROA -0.035 0.030 -0.030 .030 -0.033 0.030 Ctrl_IND -0.111 0.221 -0.075 .221 -0.096 0.221 Adj R-squared 0.180 0.182 0.184 F-value 8.07*** 8.16*** 8.21*** Highest VIF 4.00 3.96 3.97 Model 4 IV Coef. Std. Err. Rep 2.310*** 0.327 Ctrl_Size -0.128** 0.071 Ctrl_ROA -0.029 0.030 Ctrl_IND 0.220 0.223 Adj R-squared 0.202 F-value 9.14*** Highest VIF 4.28

23

4.4. Additional Analyses

Table 7.1 and table 7.2 present the T-Test comparison for the significance between the mean values of

Information Value of firms with a higher reputation. The results show that the mean of the information

value increased for all types of assurance providers. The difference between the IV of accountancy firms and consultancy firms increased from 0.24 to 1.24. Also, in comparison of the previous T-Test of all firms, the results show that there is a significant relation (at 0.1 level) between the IV value of consultancy firms and accountancy firms. The relation of engineering firms to accountancy firms considering the

information value remains significant, but at a 0.05 level. According to this test the information value in the assurance reports from consultancy and engineering firms is higher than the information value by an accountancy firms. This relation is in line with hypothesis 1, considering firms with a good reputation only.

Table 7.1 Reputation Firms T-Test difference in mean Information value Accountant-Engineering Obs. Mean IV Std. Err. Std. Dev.

Accountancy 163 18.196 .305 3.895

Engineering 68 19.309 .516 4.254

Difference -1.113

Significance 0.0277

Table 7.2 Reputation Firms T-Test difference in mean Information value Accountant-Consultancy Obs. Mean IV Std. Err. Std. Dev.

Accountancy 163 18.196 .305 3.895

Consultancy 23 19.435 .895 4.294

Difference -1.239

Significance 0.080

Furthermore, for validity of the reputational firms used in my sample I have compared my list from Fortune’s Most Admired Companies to the list in the same years composed by Newsweek in their Newsweek Green Rankings from 2014 till 2017. To compile this list, they have assessed the companies worldwide based on 8 points: 1.Combined Energy Productivity Score, 2. Combined GHG Productivity Score, 3. Combined Water Productivity Score, 4. Combined Waste Productivity Score, 5.Green Revenue Percent Range, 6.Sustainability Pay Link, 7.Sustainability Board Committee and 8.Audited Environmental Metric (Newsweek, 2014). This results in an environmental reputation score where the top global 500 firms are included. The result of this comparison is that another 10 firms were excluded from the sample, leaving a total of 97 firms (see appendix 1) to measure the relation of reputation and information value of the sustainability assurance.

24

AA1000AS by consultancy and engineering firms (Briem et al., 2019; Stones, 2014). Also the assurance is more complete if a consultancy or engineering firm conducts the assurance. One of the reasons is that they have a wider scope and have more entire reports assured. For the robustness of the assurance providers I will perform an extra regression to see the relation with the information value by replacing providers with two new depended (dummy) variables: scope and standard. For scope the value takes 1 if the entire report is assured and zero if not. For the standard variable it takes value 1 if AA1000AS is used and 0 if

ISAE3000 or another standard is used. The following regression model is used:

IV = 𝛼x +β1 (Scope) + β2 (Standard ) +β3 (Ctrl_Size) + β4(Ctrl_ROA) + β5 (Ctrl_IND) +

β6 (Ctrl_Country) + β7 (Ctrl_Year) + 𝜀x

The results in table 8 show that there is a strong positive significant (at 0.01 level) relation for both scope and standard to the information value of the assurance report, which implies that assurance providers that assure the entire sustainability report will have a positive relation to the information value, and if the assurance provider will use AA1000AS it also will have a positive relation to the information value of the sustainability assurance report.

Table 8. Additional Regression

Model 5 IV Coef. Std. Err. Scope 2.306*** 0.150 Standard 2.058*** 0.222 Ctrl_Size 0.001 0.064 Ctrl_ROA -0.046* 0.028 Ctrl_IND -0.077 0.204 Adj R-squared 0.308 F-value 15.03*** Highest VIF 3.95

25

5. Discussion and conclusion

The final section of this paper will discuss the findings of this study by analyzing the results and

answering the research questions in detail. Subsequently, the hypotheses will be discussed and the possible explanations and implications of the findings will be discussed. Then, the recommendations for future research will be discussed and finally the limitations of this research will be elaborated.

Sustainability assurance is an important mechanism for stakeholders helping in estimate the firm’s value and monitor sustainability performance of a firm and their managers. An effective assurance should be of high quality, correct and complete. This study’s objective is to explore differences in information value by the main three assurance providers, followed by exploring the impact of a good reputation to the

information value in the assurance reports. Little research is done exploring these impacts and mixed ideas about differences in assurance providers are shown in previous research. I predict that the choice of an assurance provider will impact the information value of the assurance and firms with a high reputation have a higher information value in the sustainability assurance report for the transparency to the stakeholders.

Using 1862 firms worldwide in 97 countries from the GRI database, I performed a linear regression. Information Value (IV) is measured following Gürtürk & Hahn (2016) sustainability assurance index in order to measure the differences of assurance providers. I partly found significant evidence to support my first hypothesis; engineering firms score better on information value in their assurance report compared to accountancy firms. For consultancy firms no significant relation is found, except in the additional test, where companies with a good reputation have found a positive significant relation for both engineering firms and consultancy firms. The findings are in line with previous studies. Datt et al. (2019), Al-Shaer et al. (2019) & Martinez-Ferrero et al.(2016) already mentioned the technical expertise and specific

knowledge engineering firms (and consultancy firms) have and together with the sustainability expertise and the use of the AA1000AS standard the assurance reports will be more informative for the users. Previous studies focus on the differences between accountancy firms and non-accountancy firms. This study shows that differences can arise between more assurance providers.

Subsequently, I have used Fortune’s Most Admired Companies list to indicate good reputation of a firm. I have found strong significant evidence that firms with a good reputation have a sustainability assurance report that scores higher on information value than firms with less good reputation.

According Horjoto et al. (2017) and Martín-de Castro (2020) reputation is considered as one of the highest risk, and therefore firms with a good reputation are more focusing on maintaining their reputational status by being transparent to their stakeholders with a higher information value in the sustainability assurance reports.

26

to this study is that the results can be used by assurance providers and regulating bodies. The information value of the assurance reports differ per assurance provider, nevertheless the scores are relative low (between 14 and 21). This could stimulate regulatory bodies to manage certain standards and principles for assurance providers in order to increase the information value, i.e. quality in the assurance reports. For assurance providers, such as accountants, this study shows that the information value is the lowest in their reports. In general accountants charge more than other assurance providers (Farroq & de Villiers, 2019). Regarding accountancy firms, the results in this study can harm the perceived value from a firm’s stakeholder, since it is hard to justify the premium fees asked by accountants. On the other hand

engineering firms and also consultancy firms can compete in gaining more market share. Moreover, this study contributes to the current sustainability assurance literature. The results provide additional insights on the quality of sustainability assurance, since limited research is done on the quality of the assurance reports.

As in several other papers, this study also has recommendations for further research and limitations caused by the resources available and the limited time available to complete this research. Future research could combine more databases or hand-collect data from top reputation firms which are not included in this dataset due to missing data in the GRI-Database or missing data in the ASSET-4 dataset, which is also a limitation of this research. Well-known firms like Apple, Amazon.com, Walt Disney, Starbucks etc. are not included in this sample. From the 544 firms in the Worlds Most Admired Companies list from 2014-2017 only 97 firms with 254 assurance statements in this 4 year period are included in this sample. The same goes for the number of reports from the same companies over the 4 years. Not every company has an assurance report available each year, or uploaded via in the GRI database, which may result in less

significant results. In addition, the score list of Gürtürk et al. (2016) is limited in the number of points that can be given and the relationships in points that are given are also debatable. For example, you get as many points when you put the location of the assuror's office in the statement as for considerations of any reservations of the assuror. Some statements are also much more extensive than others, but only 1 point can be given in some cases. These aspects of the Sustainability Assurance Index are debatable. Future research could implement a new score index for sustainability assurance reports, which first need research to the value certain information gives to the users of the reports. Moreover, data has been collected for the information value of the assurance statements by several people, which may lead to inherent subjective results due to, for example, different interpretations of the score model.

Despite of this study’s limitations and partial results in significant relationships, I trust this research to go some way towards improving the understanding of assurance quality in the CSR literature and the differences in assurance providers and corporate reputation, which could be explored by further research.

27

References

Al-Shaer, H. & Zaman, M., (2017). Credibility of sustainability reports: The contribution of audit committees. Business Strategy and the Ennvironment. vol. 27 (7): p. 973-986.

An, Y., Davey, H., & Eggleton, I. (2011). Towards a comprehensive theoretical framework for voluntary IC disclosure. Journal of Intellectual Capital, 12(4): 571-585.

Blumberg, B., Cooper, D., & Schindler, P. (2014). Business Research Methods. McGraw hill education

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

Journal, 20(2): p. 269–296.

Boiral, O., Heras-Saizarbitoria, I., & Brotherton, M.-C., (2019). Assessing and improving the quality of sustainability reports: The auditors’ perspective. Journal of Business Ethics. Mar2019, vol. 155 (3), p.703-721.

Brammer S and Pavelin S (2006) Corporate reputation and social performance: The importance of fit.

Journal of Management Studies. 43: 435–455.

Briem, C.R., Wald, A. (2018) "Implementing third-party assurance in integratedreporting:

Companies’ motivation and auditors’ role", Accounting, Auditing & Accountability Journal. 31 (5), p.1461-1485.

Cao, Y., Myers, L.A. & Omer, T.C. (2012). Does Company Reputation Matter for Financial

Reporting Quality? Evidence from Restatements. Contemporary Accounting Research. 29(3): p. 956-990.

Casey, R.J. & Grenier, J.H. (2015). Understanding and Contributing to the Enigma of Corporate Social Responsibility (CSR) Assurance in the United States. Auditing: A Journal of Practice &

Theory. 34(1): 97-130.

Cheng, M., Green, W. & Ko, J. (2012). The impact of sustainability assurance and company strategy on investors’ decisions. Australian Accounting Review. 34(1): p. 131-162.

28

An empirical investigation of corporate environmental reputation. Accounting, Organizations and

Society. 37(1), 14-25.

Clarkson, P. M., Li, Y., Richardson, G. D. and Vasvari, F. P. (2011), “Does it really pay to be green? Determinants and consequences of proactive environmental strategies”. Journal of Accounting and

Public Policy. Vol. 30(2), p. 122-144.

Datt, R., Luo, L., Tang, Q., & Mallik, G. (2018). An international study of determinants of voluntary carbon assurance. Journal of International Accounting Research. 17(3): p. 1–20.

Datt, R., Luo, L., & Tang, Q. (2019). The impact of legitimacy threat on the choice of external carbon assurance: Evidence from the United States. Accounting Research Journal. 32(2): p.181–202.

Deegan, C. 2002. Introduction: The legitimizing effect of social and environmental disclosures – a theoretical foundation. Accounting, Auditing & Accountability Journal. 15(3): p. 282 – 31,1 Deephouse, D. L., & Carter, S. M. (2005). An examination of differences between organizational legitimacy and organizational reputation. Journal of Management Studies, 42(2): p. 329–360.

Deloitte. "Exploring Strategic Risk: 300 executives around the worl d say their view of strategic risk is changing." 2013.

Dilla, W., Janvrin, D., Perkins, J. and Raschke, R. (2019), "Do environmental responsibility views influence investors’ use of environmental performance and assurance information?", Sustainability

Accounting, Management and Policy Journal.10(3), p. 476-497.

Farooq, M. B., & de Villiers, C. (2019). The shaping of sustainability assurance through the competition between accounting and non-accounting providers. Accounting, Auditing & Accountability Journal. 32(1): p. 307-336.

Friedman, A.L. and Miles, S. (2001). Socially responsible investment and corporate social and environmental reporting in the UK. British Accounting Review, 33 (4): p. 523-48.

Galbreath, J. & Shum, P. (2012). Do customer satisfaction and reputation mediate the CSR–FP link? Evidence from Australia. Australian Journal of Management. 37 (2): p.1-229.

GRI (2013). The External Assurance of Sustainability Reporting. Amsterdam.

Retrieved from https://www.globalreporting.org/resourcelibrary/GRIAssurance.pdf.

Gürtürk, A. & Hahn, R. (2016). An empirical assessment of assurance statements in sustainability reports: smoke screens or enlightening information? Journal of Cleaner Production. 136: p. 30-41.

29

results, trends, theory, and opportunities in an expanding field of research. Journal of Cleaner

Production. vol. 59: p. 5–21.

Harjoto, A.M. & Salas, J. (2017), "Strategic and institutional sustainability: corporate social

responsibility, brand value, and Interbrand listing", Journal of Product & Brand Management. 26 (6): p. 545-558.

Hassan, A., Elamer, A.A., Fletcher, M. and Sobhan, N. (2020), "Voluntary assurance of sustainability reporting: evidence from an emerging economy", Accounting Research Journal. 33(2): p. 391-410. Hassan, M., Roebuck P. & Simnett R. (2003.) An Investigation of Alternative Reporting Formats for Communicating Moderate Levels of Assurance, Auditing: A Journal of Practice and Theory. 22( 2): p. 171-187.

Hillman, A. J., & Keim, G. D. (2001). Shareholder value, stakeholder management, and social issues: What's the bottom line? Strategic Management Journal. 22(2): p. 125–139.

Hodge, K., Subramaniam, N., & Stewart, J. (2009). Assurance of sustainability reports: Impact on report users’ confidence and perceptions of information credibility. Australian Accounting Review. 19(3): p. 178–194.

Holm, C., & Zaman, M., (2012). Regulating audit quality: Restoring trust and legitimacy. In Accounting

forum. 36(1): p. 51-61.

Hummel, K., Schlick, C., & Fifka, M. (2017). The role of sustainability performance and accounting

assurors in sustainability assurance engagements. Journal of Business Ethics. 154(1): p. 733-757.

Junior, R.M., Best, P.J., Cotter, J. (2014). Sustainability reporting and assurance: ahistorical analysis on a world-wide phenomenon. Journal of Business Ethics. 120 (1): p. 1-11.

Kang, H. H., & Gray, S. J. (2011). Reporting intangible assets: Voluntary disclosure practices of top emerging market companies. The International Journal of Accounting. 46(4): 402–423.

King, B.G. & Whetten, D.A. (2008), Rethinking the Relationship Between Reputation and

Legitimacy: A Social Actor Conceptualization. Corporate Reputation Review volume. 11: p. 192–207. Kraten,, M. (2019), Sustainability Reports and the Limitations of ‘Limited’ Assurance. CPA Journal. 89(7): p. 9-10.

Larrinaga, C., Rossi, A. & Luque-Vilchez, M. (2020), Institutionalization of the Contents of Sustainability Assurance Services: A Comparison Between Italy and United States. Journal of