Voluntary Sustainability Disclosures in Non-Listed Companies: An Exploratory Study on Motives and Practices

: The aim of this study is to investigate the reasons for the voluntary preparation of a sustainability report and to identify sustainability disclosure practices in the context of non-listed companies. For this purpose, a multiple case study methodology involving ﬁve manufacturing Portuguese companies was employed. Data collection consisted of a mixed approach involving interviews with the sustainability report preparers, and content analysis of sustainability reports and company webpages. The results show two main reasons to initiating voluntary sustainability reporting: requirements from speciﬁc customers and the parent company, and the need to communicate with stakeholders, in particular customers, the local community, and suppliers. Companies reveal knowledge of sustainability-related concepts and apply frameworks such as GRI Standards and SDG in preparing their sustainability reports. Several departments are involved in preparing the sustainability report (sustainability, quality, environment, human resources, marketing, and/or communication departments), but no company mentioned the ﬁnancial department. Moreover, on the webpages, sustainability reports and annual reports tend to be presented in separate sections, suggesting that companies still have a way to go in integrating sustainability information with ﬁnancial information. Overall, the results show that institutional, stakeholder, legitimacy, and signaling theories can provide explanations for the motives and practices adopted by non-listed companies in voluntarily disclosing sustainability information.


Introduction
Although governments in some countries have been requiring or encouraging the disclosure of sustainability information disclosures, and despite the recent trend towards a mandatory sustainability reporting regime in Europe following the Directive 2014/95/EU [1,2], the expansion of sustainability information disclosures across the globe has mostly occurred under a voluntary regime. In this context, we have seen a growing interest from scholars on content, practices, and reasons for companies to engage in voluntary sustainability disclosures, materialized in studies applying several theoretical frameworks and methodological approaches [3][4][5][6][7][8].
Non-listed (private) companies represent a large proportion of economic activity in most countries and have different stakeholder groups, ownership structure, financing strategies, and utility functions from their public peers that may justify different motivations and practices towards sustainability reporting [9]. However, in a similar way to other areas of corporate reporting, research on sustainability disclosures has focused mainly on listed companies, the rationale being the greater availability of public information on these entities and the accountability pressures they face [8]. With few exceptions, little is known about sustainability reporting in non-listed companies. One of those exceptions is the study of Chi et al. [9] that compares listed and non-listed companies as to the determinants for preparing a sustainability report. The other exception consists of a strand of studies that analyze the implementation of sustainability reporting in Small and Medium Sized Enterprises (SME), which in most countries tend to be non-listed [10][11][12].
Portugal is a small country where about 99 per cent of companies are SMEs. The stock exchange lists less than 50 companies, meaning that most of Portuguese companies are private, with the banking system as their main source of funding. In Portugal, only large companies that are public interest entities are subject to the mandatory sustainability reporting imposed by Directive 2014/95/EU, which requires the preparation of non-financial information to be presented in a "non-financial statement" or a separate report. For non-listed companies with limited liability, the Portuguese Companies Law requires the preparation of a management report that must include, when appropriate, non-financial performance information, namely, information on environmental and employee matters. Thus, in the context of Portuguese non-listed companies, the preparation of a sustainability report is voluntary.
Several theories, including stakeholder theory, legitimacy theory, signaling theory, agency theory, among others, have been used to justify why firms disclose or do not disclose voluntarily sustainability information [13][14][15]. However, these theories were developed and applied mainly in the context of listed companies, so their validity in explaining voluntary disclosures of non-listed companies, where legitimacy concerns and external pressures are less strong, is not fully known.
Regarding the methodological approaches, the studies on the motives of voluntary sustainability reporting can be divided in two groups. The first includes the studies that test whether certain internal and/or external factors, considered proxies for institutional or stakeholders' pressures, legitimacy concerns or other motives for voluntary sustainability reporting-the "determinants" to be tested-affect voluntary sustainability reporting [3,6,7,16]. Content analysis of corporate reports or webpages is the most used method of data collection in the studies on the "determinants". The second group corresponds to the studies where companies (the preparers of sustainability information) are directly observed and/or questioned about the reasons for voluntary sustainability reporting. These studies, fewer in number, due to the costs and complexity involved, intend to provide a deeper understanding of the phenomenon under analysis, adopting more qualitative methodological approaches, such as case studies, and data collection methods such as interviews, questionnaires, and observation [5,8].
Based on these considerations, this study aims to investigate the motives for the voluntary preparation of a sustainability report and to identify sustainability disclosure practices of non-listed companies. To this purpose, a multiple case study methodology was employed, involving five manufacturing Portuguese companies and a mixed approach of data collection from three sources: interviews, sustainability reports, and webpages. The contributions of this study are several. Firstly, we expand the knowledge on the motives and practices of sustainability reporting to a less explored context, that of nonlisted companies, demonstrating that these companies already identify their stakeholders and show some sophistication in the frameworks applied. Secondly, we analyze these motivations and practices in the light of several theories, concluding that non-listed companies, like listed ones, also have concerns related to communication with stakeholders and the need to be perceived as legitimate in the view of the community where they are located. Thirdly, we contribute to a line of more qualitative studies, which allows us to observe other dimensions of the voluntary reporting phenomenon, responding to the call of more in-depth studies, in different geographical contexts, that capture perceptions, beliefs, attitudes, and experiences of different parties involved in sustainability reporting [8]. Fourthly, we expand the empirical evidence on the reasons for voluntary sustainability disclosures in the Portuguese context. In Portugal, except for Eugénio et al. [17], who employed a single case study methodology, empirical evidence comes from studies on "determinants" of voluntary disclosure carried out with listed companies [18,19]; or listed and non-listed [20][21][22]; or listed banks [23]. Finally, considering the existent proposal for a new Directive on Corporate Sustainability Reporting [24], which will impose sustainability reporting to all large companies (listed and non-listed) and all listed companies (except listed micro-enterprises), our study makes an important contribution providing evidence on the degree of preparedness of non-listed companies for the not-so-distant future in which sustainability reporting will be an obligation for them.

Voluntary Sustainability Disclosures: Concept
Voluntary disclosures, in general, refer to the discretionary release of financial and non-financial information beyond compliance with standard-setting body regulation, law, or contractual obligations [16,25,26]. They are "disclosures in excess of requirements" and represent "free choices" by companies' management to provide accounting and other information deemed relevant to the decision needs of users of their annual reports [27] (p. 555). From an economic perspective, companies only disclose information voluntarily when the expected benefits exceed the direct and indirect costs of doing so.
Sustainability disclosures, also labelled as "corporate social responsibility" (CSR) or "triple bottom line: people, planet, profit" (TBL) or "environmental, social and governance" (ESG) disclosures, have originally focused on environmental information, broadening their scope to include ethical and social issues, usually employee and community matters, as well as corporate governance aspects [28]. There are several channels through which companies may disclose sustainability information, such as the annual report, standalone reports, or corporate webpages. Sustainability reporting refers to a more formal process, since reporting involves measurement, disclosure, and communication of information [16]. Consistently, sustainability-related reports include standalone reports covering only one dimension, as social reports or environmental reports, or more comprehensive reports, as CSR or sustainability reports. Integrated reports combine sustainability information with traditional financial information in a single report to provide a holistic picture of value creation over time [5,7]. For sake of simplicity, throughout this work, "sustainability reporting" and "sustainability disclosure" are used as synonymous for the release of sustainability information by companies through any channel, and "sustainability reports" refers specifically to standalone reports including CSR reports.
The research on sustainability disclosures deals with the "adoption", the "extent", and the "quality" [7,29]. The "adoption" refers to the decision or likelihood to engage in sustainability disclosures through one or more of the described channels. The "extent" generally addresses the "volume or amount of reporting (i.e., the quantity of disclosed information based on keyword, sentence or page-counts in order to identify major themes discussed in sustainability-related reports)" [7] (p. 10). The "quality" of sustainability reporting concerns more specific aspects of the information disclosed, for example, if it is merely narrative and descriptive (not easily verifiable) or if it is specific, quantifiable, and monetary data (objective) [7]. Other aspects related to the "quality" of sustainability disclosures are the voluntary adoption of sustainability standards, such as GRI Standards [30] or the voluntary assurance of sustainability information [31].

Motives for Voluntary Sustainability Disclosure
Empirical evidence shows that no single theory seems to be sufficient to explain the motives and practices of voluntarily disclose sustainability information. Alongside with studies adopting a single theory, such as the stakeholder or legitimacy theory (e.g., [9,17,20]), there are others that adopt several theories simultaneously (e.g., [32][33][34]), while other studies refrain from mentioning a theory (e.g., [35]). In the following sections, we present the explanations provided by empirical literature for voluntary sustainability disclosure, as well as reasons for non-disclosure, framing them within the respective theories and illustrating them with empirical evidence.

Meet Stakeholder's Information Needs (Stakeholder Theory)
Stakeholder theory is based on the relationship between an organization and a set of "interested parties", the stakeholders, these being defined as "those groups who can affect or are affected by the achievement of an organization's purpose" [36] (p. 49). According to this theory, managers have a fiduciary duty to satisfy the expectations of all stakeholders and not only those of shareholders, as in traditional shareholder theories [36,37]. This is because stakeholder theory recognizes the existence of other interests in the company beyond those related to its economic and financial performance, such as sustainability performance.
From an ethical perspective, the company has the responsibility to provide information to all its stakeholders, regardless of their power. However, from a resource management perspective, the company will be more motivated (or pressured) to report to the most important stakeholders in terms of the resources they provide to the company [14]. For this reason, when defining the content and format of sustainability reporting, companies evaluate the "stakeholder salience" prioritizing their expectations according to their strategic importance (power) to the company [38,39].
Sustainability disclosures are addressed to various groups of stakeholders as suppliers, customers, capital providers, employees, and community [40]. Studies on the "determinants" of voluntary sustainability disclosure often use the company's size as proxy for the pressures exerted by stakeholders. In this vein, the empirical evidence has shown a positive relationship between the companies' size and the extent or quality of voluntary sustainability disclosure in general (e.g., [18,20,41,42]); and on specific matters such as carbon disclosures and environmental disclosures, in particular (e.g., [32,43]). Some studies found a positive association between the consumer proximity of the companies' sector and specific sustainability disclosures, suggesting an orientation of these companies towards specific stakeholders' groups, such as customers, community, and society [18,20].
Stakeholders' orientation also shapes the channels used by companies to communicate voluntarily sustainability information. Formal channels such as sustainability reports, integrated reports, and annual reports are mainly addressed to investors, and informal channels such as webpages are used to communicate to local communities [18,44].

Manage Image and Reputation (Legitimacy Theory)
Beyond meeting stakeholder needs, voluntarily disclosing information could also serve to legitimize the companies' actions concerning sustainability and this is supported by legitimacy theory. The main difference between the stakeholder theory and the legitimacy theory lies in the "actors" of each: stakeholder theory focuses on the needs and pressures of specific stakeholders, whereas legitimacy theory focuses on society as a whole [32].
According to legitimacy theory, to operate, a company needs to be accepted by society, that is, it must have legitimacy in the sense of a social "license to operate" [45] as only then will it have access to the resources needed to conduct business successfully. Thus, the theory is concerned with the relationship between the organization and society at large, without considering individuals separately [45]. Suchman [46] considers that "Legitimacy is a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions." (p. 574). Legitimacy management is based on communication between the organization and its audiences, and it is oriented by one of three strategies: gain, maintain, or repair legitimacy [46].
Communicating the companies' sustainability practices can be used not only to meet stakeholder expectations, but also to legitimize their operations and gaining social acceptance [47]. Companies that fail to do this risk losing access to resources that are afforded to them by society [48]. Thus, the legitimacy theory predicts that managers engage in sustainability practices and voluntarily disclose them as a strategy to maintain, gain, or restore legitimacy [14].
Larger companies are more visible and, therefore, more pressured to disclose information to manage image and reputation. Along with size, the company's activity sector (industry) may also give rise to legitimacy concerns and is often pointed to as a "determinant" of voluntary sustainability disclosure [49]. Empirical evidence has shown that companies belonging to more environmentally sensitive sectors disclose more sustainability information in general [20,41] and/or on environmental issues, in particular [20,50,51]. Byrd [52] found that companies in "controversial" sectors (such as alcohol, tobacco, and firearms) disclose more on social and community actions than companies in "noncontroversial" ones. Grougiou et al. [53] observed that companies in so-called "sin" industries are more likely to provide sustainability reports.
Voluntary sustainability disclosures may also occur for legitimation reasons after specific events, as the BP oil spill in 2010 [54]. Based on a case study of a Portuguese cement company, Eugénio et al. [17] identified several legitimacy strategies carried out through sustainability reports after controversies involving the company, including "don't panic", "create monitors", "justify", "disassociate", and "explain".
A strand of studies analyzing the relationship between sustainability performance and sustainability disclosures also documented voluntary sustainability reporting for legitimacy reasons. For example, Luo [55] found a negative relationship between voluntary carbon disclosures and carbon emission performance, among Global 500 companies participating in the Carbon Disclosure Project, concluding that companies with lower carbon emission performance (higher level of carbon emissions) provided more detailed carbon disclosures, as a tool to regain its legitimacy. It should be noted that there are arguments for either a positive or a negative relationship between sustainability performance and sustainability disclosure. Legitimacy theory predicts a negative relationship but signaling theory can explain a positive relationship [56].

Signal the Quality of the Company (Signaling Theory)
Signaling theory is based on the presence of information asymmetries between insiders (management) and outsiders (shareholders and other stakeholders) and explains the process through which those asymmetries are reduced. In this process, the party with more information (the signaler) conveys information (sends the signal) to the other party (the receiver) and this one interprets that information and reacts to the signal (the receiver's feedback) [57,58]. Signaling theory implies that the sender intends to signal its true quality to the receiver and the receiver stand to gain from making decisions based on information obtained from those signals. Thus, signaling is a viable strategy when the receiver knows that the signal is sent only by "high quality" companies, and this happens when two conditions hold: (i) for the "high quality" companies, the gains from signaling outweigh the gains from any other strategy; and (ii) for the "low quality" companies, a nonsignaling strategy provides a bigger payoff than does signaling [59]. For a signal to be effective, it must be observable by the receiver and have a cost that can be supported by the sender [57].
In the context of sustainability reporting, information asymmetries arise because it is difficult for the parties outside the company to gain credible information on the company's sustainability strategies, performance, and risks, and voluntary sustainability reporting can play a role in reducing those asymmetries. Thus, signaling theory predicts that companies with better sustainability performance or more engaged with sustainable development (the signaler) have incentives to voluntarily disclose information on these issues or voluntarily prepare a standalone sustainability report (the signal) to demonstrate its unobservable characteristics to the less-informed stakeholders (the receiver). For "low quality" companies, no action is expected since these companies need considerably more change and resources to implement sustainability strategies and reporting practices [57].
Studies on the "determinants" of voluntary sustainability disclosures use profitability as a proxy for the companies' quality and a positive relationship between profitability and voluntary disclosures confirms the use of sustainability reporting as a signal of the companies' quality [32,41,42].
The strand of studies examining the relationship between sustainability performance and sustainability disclosure have also documented that voluntary sustainability reporting can act as a signal of the companies' quality. For example, Karaman et al. [60] concluded that companies in the logistic sector with better "green performance" were more likely to prepare a sustainability report. Mahoney et al. [61] tested two competing explanations for the preparation of a sustainability report by U.S. companies-signaling and greenwashing-and found evidence consistent with the signaling explanation, as companies issuing voluntarily sustainability reports had higher corporate social responsibility scores.
Adams and Frost [35] interviewed four British and three Australian companies actively engaged with sustainability reporting on the reasons why they have started this process voluntarily. The reasons given were the high impact of their operations on the environment; to be accountable to, and build trust with, key stakeholders such as NGOs and local communities; to influence business leaders and key opinion makers; to differentiating themselves from competitors; to follow competitors; and, as a result of the influence of tools such as GRI Standards. These reasons are consistent with stakeholder, legitimacy, and signaling theories.

Reduce Information Asymmetries and Agency Costs (Agency Theory)
Agency theory is based on the agency contract that is established between managers (the agent) and capital providers (investors/shareholders and lenders). This relationship is characterized by the existence of information asymmetries and agency conflicts that originate the so-called agency costs [62,63]. Agency costs and information asymmetries are reflected by investors and lenders in the returns required and, consequently, affect the cost of capital [64,65].
Reducing agency costs and/or information asymmetries to thereby achieve a lower cost of capital may be an incentive for companies to voluntarily disclose information in general [66] and sustainability information, in particular. Empirically, this motivation could be tested directly, through the relationship between the cost of capital and voluntary sustainability disclosure [67,68], or indirectly, through the relationship between the presence of information asymmetries or agency costs and voluntary disclosure. As proxies for the presence of information asymmetries and agency costs, empirical studies often use leverage, the age of the companies, ownership structure, and variables related to the quality of corporate governance, as board independence or the existence of a sustainability committee [32,51,69].
Empirical evidence supporting the reduction of cost of capital as a motivation for voluntary sustainability reporting is limited and not conclusive. Carp et al. [67] found no significant relationship between the decision to disclose sustainability information and the cost of capital in Romanian listed companies considering that investors, lenders, and business partners interpret sustainability reporting as insufficiently documented and as having a low capacity for integration within the decision-making process. Gjergji et al. [68] found a negative relationship between environmental disclosure and the cost of capital in Italian SMEs, suggesting that debtholders may see disclosure as a potential risk of release sensitive (proprietary) information. Chi et al. [9] documented that non-listed companies in Taiwan are less likely to issue sustainability reports compared with their listed peers because only listed companies enjoy more favorable credit ratings and a lower cost of debt due to sustainability reporting, while private companies do not reap similar benefits. Further analysis suggested that debtholders only consider information on sustainability when the reports issued by private companies are assured, suggesting that reports issued by private firms may lack credibility and be viewed as "window-dressing" by debtholders.

Avoid Political Costs (Political Costs Theory)
The political cost theory states that companies are subject to political processes through which a potential wealth transfer may occur, for example, through the payment of taxes, the reduction of Government subsidies, or other regulatory action (the political costs). The theory predicts that, in this context, managers adopt accounting procedures or make accounting choices that reduce their wealth transfer [70][71][72]. In addition to politicians, non-governmental entities and other stakeholders can lobby for wealth transfers from companies in their own interest. Voluntary sustainability reporting may be used to distract other's attention from aspects that may give rise to political costs or to create a reputation that avoids or reduces political costs. Size and profitability are often used as proxies for political costs, because larger and more profitable companies are more visible to the public and tend to be subject to greater political and regulatory pressures from external interest groups [27,39,50]. Moreover, the activity sector, when subject to specific regulation or attention from lobby groups, may be a proxy of political costs that may induce voluntary sustainability disclosures [27,73,74]. Other proxies for political costs are media attention, number of employees, and listing status [50].

Respond to Institutional Pressures (Institutional Theory)
Institutional theory examines organizational forms and explains why organizations within the same "organizational field" have homogeneous characteristics or forms [14] (p. 162). The "organizational field" consists of "those organizations that, in the aggregate, constitute a recognized area of institutional life: key suppliers, resource and product consumers, regulatory agencies, and other organizations that produce similar services or products" [75] (p. 148). Institutional theory views organizations as operating within a "social framework of norms, values, and taken-for-granted assumptions about what constitutes appropriate or acceptable economic behavior" [76] (p. 699). Conformity to social expectations and predominant norms or traditions define socially acceptable conduct, which is essential to organizational success and survival. The organizations' tendencies toward conformity lead to homogeneity among structures and management practices of organizations in the same organizational field, regardless of their actual usefulness or organizational efficiency [75,77,78]. Thus, once an organizational field is structured, various powerful forces emerge within society, which cause organizations within the field to become more similar to one another [75]. This homogenization process is known by "isomorphism", that is, "a constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions" [75] (p. 149). There are three mechanisms through which institutional isomorphic change occurs: (1) coercive isomorphism that stems from formal and informal pressures exerted on organizations by other organizations upon which they are dependent and by cultural expectations in the society within which organizations function; (2) mimetic isomorphism results from uncertainty and consists on imitation, that is, when organizations face problems with ambiguous causes or unclear solutions they may model themselves on other organizations perceived to be most successful; and (3) normative isomorphism that stems primarily from professionalization and is related to the pressures emerging from common values to adopt a particular institutional practice, coming from formal education or professional networks [75].
Institutional theory provides a framework for analyzing the process through which voluntary sustainability reporting practices are embedded in the organization and become widespread across companies. De Villiers et al. [79] explain this process as follows. When companies face uncertainty concerning the practices to adopt, they try to mimic other successful companies to overcome uncertainty, benchmarking or copying the best practices. The uncertainty underlying voluntary sustainability reporting favors this "mimetic isomorphism". Over time, pressures from new regulations and demands from influential stakeholders create a "coercive isomorphism", forcing the company to comply with rules. Mimetic and coercive forces act in a first moment, introducing a "field" and institutionalizing it. For the field to reach maturity, "normative isomorphism" must occur through professionalization driven by similar training and social interaction between professionals, which leads to conform to taken-for-granted norms and the internalization of new norms.
In the context of voluntary sustainability reporting, mimetic isomorphism is present when multi-national companies follow the practices of their peers; smaller companies benchmark the practices of industry leaders; companies follow the practices of competitors to gain or retain specific customers; or when companies interact with each other to establish the extent of disclosures, in order not to do anything that is considered too much [80][81][82]. Coercive isomorphism may result from specific demands of influential stakeholders regarding the implementation and disclosure of sustainable practices, as the International Organization of Standardization (ISO), the parent company, customers, and suppliers [82][83][84]. Normative isomorphism is present when professionals inside (e.g., managers and directors) or outside the company (e.g., external consultants and audit firms), due to similar educational background, training, or experience, adopt similar rules and practices such as the use of GRI standards [80].

Summary of the Motives for Voluntary Sustainability Disclosure
The motives why companies voluntarily disclose sustainability information and the most frequently applied theories to explain those reasons can be summarized as follows: • Respond to sustainability information needs and expectations of specific stakeholders (stakeholder theory); • Legitimize companies' actions related to sustainability (legitimacy theory); • Signal the companies' positive sustainability performance to stakeholders (signaling theory); • Reduce the information asymmetries and agency costs of the relationship between managers and capital providers (agency theory); • Reduce wealth transfers due to political costs that may arise from poorer sustainability performance or disclosure (political costs theory); • Respond to institutional pressures regarding both sustainability performance and disclosure through an isomorphic change process (institutional theory).
It should be noted that some of these motives imply the use of impression management strategies in sustainability information disclosure in a way that influences stakeholders' perceptions of true sustainability performance [85,86].
Although there are several motives for voluntary sustainability disclosure, not all companies do so, so the reasons for non-disclosure are discussed in the following section.

Reasons for Non-Disclosure (Proprietary Cost Theory)
Proprietary cost theory assumes that voluntary disclosure involves two types of costs: cost of preparing, certifying, and disseminating information; and cost of disclosing proprietary information that can be used by external parties, such as competitors, to harm the company [87]. On one hand, this theory predicts that managers have incentives to voluntarily disclose sustainability information only when the expected benefits exceed the costs. On the other hand, it predicts that companies tend to hide information that reveal crucial aspects about the business that could negatively affect the company's competitive position in the market [88].
Stubbs et al. [89] interviewed 23 companies from Australia's top 200 about the reasons why they do not prepare a sustainability report and the reason found was because no one requests that information and companies do not believe they have a duty to account for their environmental and social impacts. Dissanayake et al. [90] found as barriers for sustainability reporting by listed companies in countries of the Indo-Pacific region the (1) the manager's lack of knowledge, understanding, awareness, and education in sustainability; (2) the costs and time involved; and (3) the lack of initiatives from government. Krishnamurti and Velayutham [91] found that Australian listed companies that have both audit and risk management committees disclose less information on greenhouse gas emissions, which is consistent with these companies withholding risky information that could be valuable to competitors or cause adverse market reactions.

Multiple Case Study
The aim of this study is to investigate the motives and practices relating to voluntary sustainability disclosure in the context of non-listed companies. A qualitative research design consisting of a multiple case study of five Portuguese companies was adopted, as it is considered the most appropriate research strategy when the aim of the study is to gain an in-depth understanding of a specific context [92]. Although the study of a limited number of cases does not support statistical generalizations, case study research can be used to make analytical generalizations, representing theoretical conclusions that have the potential to provide a new understanding of an existing phenomena [92].
We used three sources to gather data for our research: interviews, sustainability reports, and corporate webpages, each of which will be discussed below. The companies, located in the same geographical region, were specifically targeted because: (i) they are non-listed companies that voluntarily prepare sustainability reports and have a corporate webpage containing sustainability information; (ii) they belong to the manufacturing sector, with different products, and with import and export activity, which ensures the existence of a range of stakeholders with distinct needs of sustainability information; and iii) they agreed to participate in this study by giving the interview. As can be seen in Table 1, all the companies interviewed are medium and large sized.

Data Collection, Analysis and Treatment
Both primary and secondary data sources were used in this research. The data collection process took place in two stages. Firstly, semi-structured interviews were conducted to understand the reasons for the voluntary preparation of a sustainability report, how the process began and evolved within the organization, which departments were involved in its preparation, and to gain a general perception of the advantages and disadvantages associated with sustainability reporting. The interviews also revealed some of the practices adopted in relation to the format and content of sustainability reports. Secondly, data from sustainability reports and webpages were analyzed in order to complete the picture of the practices adopted by private companies with regard to sustainability disclosures. It should be noted that not all the companies publish their sustainability reports regularly on their webpages, which conditioned its analysis.
The companies were first contacted by phone to identify the person responsible for the sustainability area or, in the absence of such a function, the person responsible for the content of the sustainability report, to ensure a high level of knowledge and close involvement with all the process of preparation and disclosure of sustainability information, necessary to provide relevant and informed views in answering the questions posed. This person was subsequently contacted by phone and/or email to explain the purpose of the research and the scope of the interview. In one of the interviews (company C1), two people participated in the interview, at the company's wish. Table 2 provides details of the interviewee's roles. To ensure anonymity, the companies are identified with a code and no relationship is stablished between the information in Tables 1 and 2. The questions of the interview guide were developed by the authors, specifically to address the purpose of the study and are presented in Table 3. Table 3. Interview guide.
Q1 What were the reasons that led the company to voluntarily prepare a sustainability report?

Q2
How did the process of preparing the sustainability report begin and how has it evolved?

Q3
Which departments are involved in the preparation of the sustainability report?

Q4
Who are the stakeholders to whom the sustainability report is addressed?

Q5
How is the content of the sustainability report defined?

Q6
What are the advantages and disadvantages associated with sustainability reporting?
The final interviews were conducted between the months of October and December 2020, according to the availability of the interviewees. Given the pandemic context, some companies requested that the interview be held by videoconference. Thus, only the interview with company C1 was carried out in person, at the company's premises. The remaining interviews were conducted by videoconference, using the Microsoft Teams software. The average duration of the interviews was 30 min, and all the interviewees allowed the conversation to be recorded without any objections. The recorded content of the interviews was transcribed manually for the purpose of analysis and transcriptions were checked against the recordings and corrected when necessary. The transcriptions were analyzed and coded to draw out key themes [93] that might explain the motives for the voluntary preparation of a sustainability report and illustrate the reporting practices in use.
The companies' webpages were accessed during August 2021 and analyzed at two stages. Initially, the structure of the webpage was examined, and it was checked whether the sustainability reports were available there. After this, a content analysis of information in web browser format (e.g., HTLM) was performed to collect sustainability information. To guide the data collection, a framework similar to those adopted by Branco and Rodrigues [18] and Matuszak and Rozanska [34] was followed. This framework covers four categories of information: environment; human resources; products and consumers; and community involvement. Considering the information found, new items were included in the existing categories, as well as a new category called "Suppliers" (see Appendix A). To quantify the information disclosed, a binary dichotomous scoring system was applied, checking for the presence (1) or absence (0) of each item. Given the reduced numbers of companies and items disclosed by each company, it was not considered relevant for the analysis a construction of a disclosure index. The sustainability reports available on the webpages were considered a separate information source but, since a lack of regularity in their publication was observed, their analysis was limited to identifying the frameworks used in their preparation.

Motives for the Voluntary Preparation of a Sustainability Report
The motives for the voluntary preparation of a sustainability report (SR) were directly asked in the first question of the interview (Q1) and the answers to Q4 and Q6 helped to clarify those motives. Table 4 shows the reasons pointed out by companies for the preparation of the SR (Q1) and Table 5 shows the stakeholders to whom it is addressed (Q4). Table 4. Motives for the voluntary preparation of the sustainability report.

Required by specific customers C1
Required by the parent company C2, C3 On the company's own initiative C4, C5 Regarding the reasons for starting the voluntary preparation of the SR, the answers to Q1 (Table 4) showed two external pressures: requirements from specific customers (C1) and imposition of the parent company (C2 and C3). The companies that prepare the SR on their own initiative (C4 and C5) seek to communicate with the community and keep up with current reporting trends. For instance, C4 mentioned that it has decided to prepare the SR "to keep up with the movement towards publishing information on sustainability, governance, and the environment" and "to give information on the prospects of evolution and on the social or environmental impacts of the company". Company C5 said "the paradigm is changing, in the sense that it is not only about financial performance, but also about all other performance ( . . . ) if a company wants to remain competitive and sustainable in the long term, and relevant for its stakeholders and for society, it must adapt, it must be aware of these changes and it must have the capacity to respond to them". Table 5 presents the stakeholders to whom the SR is addressed (Q4). "Customers" were pointed out by all the companies, followed by "Community, local authorities, associations, scientific and academic community", mentioned by four companies. It is worth noting that capital providers (shareholders and banks) were not mentioned by the majority of companies. Furthermore, employees were not referred by any company, suggesting that the SR is directed towards stakeholders outside the company. Suppliers were mentioned not only as stakeholders of the SR but also as possible providers of this information in the future, as a way of ensuring sustainability along the value chain. For example, C2 said "it is important that we get our suppliers to provide us with sustainability information to maintain a sustainable value chain; the company seeks to work with credible suppliers, who offer confidence both in the products they supply and, in the practices used".
When questioned about the advantages of preparing the SR (Q6), companies agreed that it is an important tool for communication with the wide range of stakeholders and crucial for promoting a positive company's image. Company C1 mentioned that the SR "gives a broad view of the company's policies, vision, actions, and strategies in social and environmental areas". Company C3 said, "this report gives visibility to our projects, internally and externally; and also, a structured vision of what is done". Finally, C5 reported a clear concern with its image: "our company is a chemical industry and its connotation, especially in the community, may not be the best ( . . . ) this type of publication helps the community to better understand the company's activity and performance". Table 6 provides some aspects of the content and format of SRs, based on the responses to questions Q2, Q3, and Q5 of the interviews. The availability of SRs on the companies' webpages and the use of a framework for SRs preparation were verified through the content analysis of both the webpages and the available SRs. As can be seen in Table 6, companies show several reporting practices. Company C1 was the last to start the process, with the preparation and publication on the webpage of its first SR for the 2018-2019 biennium. Conversely, C2 is the company that has been preparing a SR for the longest time, by imposition of the parent company, and does not make it available on the webpage. Companies stated during the interviews that they have not stopped preparing the SR since they started this process. However, making them available on the webpages does not seem to be a priority. For example, when the companies' webpages were accessed in 2021, C1 had not yet made available the report for the year 2020. Company C4 had not published the SR for the year 2019, and in 2020 started publishing integrated reports. Company C5 also publishes integrated reports since 2017 and the last SR available on the webpage was for the year 2016.

Sustainability Reports
The scope of sustainability issues justifies the intervention of various departments in the preparation of the SR, in particular human resources, quality, and environment. Interestingly, the communication and marketing departments are also mentioned by three companies (C1, C3, and C4). The SR is prepared by an external consultant in one company (C2). Only one company (C5), belonging to an environmentally sensitive sector, has a department devoted to sustainability. This department is responsible for preparing the SR, but always drawing on other departments, as said by the company: "the document itself is prepared and carried out by the sustainability department, but there are numerous players who are part of the process ( . . . ) the environment department, the safety department, the commercial department, the purchasing department, the human resources department are all part of this process, so at least one player from each area of the organization takes part in the report ( . . . ) after collecting all the information from those departments, the innovation and sustainability department handles this information and produces the document, which is then also reviewed by all the intervening areas ( . . . ) right now, the sustainability area has as much relevance as the financial area".
Regarding the content and format of the SR, companies show some degree of knowledge of the concepts and frameworks related to sustainability reporting. C1 structures its SR according to the Sustainable Development Pillars (the Triple Bottom Line); C3 and C5 are aligned with the Sustainability Development Goals (SDG) of the United Nations; C2, C4, and C5 adopted the Global Reporting Initiative (GRI) Standards or Guidelines; and C5 mentioned during the interview, and the reports corroborate, the use of a matrix of materiality to identify the topics to be disclosed. C1 mentioned in the interview that the practices adopted also resulted from benchmarking the reporting practices of other companies.
When questioned about the disadvantages of preparing the SR (Q6), more specifically about whether this involved additional costs for gathering and processing information, all companies were unanimous in stating that all necessary information has already been prepared by the company, especially for management purposes. Therefore, it was only necessary to think about how it should be presented. This may have involved some investment of time and resources the first time the SR was prepared, but this was not seen as a disadvantage or cost. For example, C1 stated that, "it consisted of systematizing all the information that existed because it was all here and then trying to present it in a more attractive way, more appealing, and presenting the company as a whole and revealing the concern in the various sustainability areas". Company C2 also mentioned the creation of a "dynamic of continuous improvement, because the fact that every year we must report information, to show others what we do, also pushes us to want to improve". Company C5 corroborates that idea, mentioning that, "obviously the first year was the most challenging, but the truth is that we have seen a brutal improvement in the process, as it has become quite automatic (...) in the beginning there was a lot of reluctance to publish certain information, but now I have the different areas alerting me to new information they want to disclose". Table 7 presents an overview of the structure of the companies' webpages. Companies C3, C4, and C5 have a standalone section devoted to sustainability. The location of the SR on the webpages varies between companies, with C1 presenting it in the "News and Events" tab; C3 and C5 on the "Sustainability" tab, and C4 making it available alongside the "Annual Reports". Information about sustainability matters is spread over several tabs, even in companies that have a specific section for sustainability on the webpage. It is worth noting that C1, C3, and C4 used the "News" section to disclose sustainability information.

Sustainability Disclosures on the Webpages
The results of the content analysis performed to the webpages are presented in Table A1  of Appendix A and Table 8 summarizes this information, allowing for a comparison with the stakeholders identified in the interviews (presented in Table 5). These tables show that the most disclosed category by all companies is "Environment". The only exception is company C3 whose most disclosed category is "Human resources". In the "Environment" category, the most disclosed item is "Environmental management, systems, and audit", mentioned by four of the five companies. However, there are items not mentioned by any company such as "Pollution arising from use of product".  Except for company C3, information on "Human resources" is almost non-existent, with two companies making no mention of this matter on their webpages (C2 and C4). The items disclosed in this category by company C3 show important actions developed in this area such as "Profit share and bonuses", "Employee benefits", or "Information on support for daytime care, maternity, and paternity". These actions can be important to motivate current employees and help attract new talent. Thus, when no information is disclosed on this subject, it may mean that companies do not develop actions in this area or do not consider it important to disclose them. Notably, in the interviews, the employees were not considered a recipient of sustainability information, which is consistent with this lack of information.
All the companies disclose some information regarding "Products and customers/ consumers", specially "Consumer (or product) awards". As can be seen in Table 8, this is consistent with the importance given to customers as a stakeholder of SR. In the same vein, disclosures on "Community involvement" also corroborates the importance given in the interviews to the community as a stakeholder in the SR. Company C5 shows the higher number of disclosures, demonstrating several actions developed as "Charitable donations and activities", "Support for education", and "Support for public health initiatives, projects and campaigns".
The importance of suppliers to ensure a sustainable value chain is also evident in the webpages with some companies providing there purchasing requirements, evaluation criteria, and code of conduct for suppliers (C2, C3, and C5).
The company in the most environmentally sensitive sector (C5) is the one with the highest number of total disclosures and the highest number of disclosures on "Environment" and "Community involvement" matters.

Discussion
This study corroborates that there is no single theory that fully explains the phenomenon of voluntary sustainability reporting [9,13,14,34].
Institutional pressures from specific customers and the parent company were the reasons pointed out in the interviews for the voluntary preparation of a SR by three from the five companies interviewed (C1, C2, and C3). This result adds to the evidence on how institutional pressures at the companies' level determines voluntary sustainability reporting [81][82][83].
Despite the coercive pressure from specific stakeholders, the results of the interviews also show that companies prepare the SR to communicate with stakeholders, in particular customers, the local community, and suppliers. The information disclosed on the webpages also highlights that the local community is an important stakeholder for non-listed companies. This result corroborates the idea stated by Chi et al. [9], that non-listed companies often serve within a local community, which is concerned with how companies affect their living environment and health. Providing evidence on the stakeholders to whom Portuguese non-listed companies address sustainability information, this study adds to previous evidence on the Portuguese context that have identified the size of the company and the consumer proximity and/or the environmental sensitivity of the business sector as determinants of the stakeholders to whom sustainability information is addressed [18,20].
While customers and suppliers are important stakeholders in any company due to the resources they place at its disposal [40], the relationship with the local community may have underlying legitimacy issues and, therefore, the disclosures related with "Community involvement" can also be explained by legitimacy theory. For example, company C5 states during the interview its concern with its image within the community, due to environmental impact of its activity, being the one that shows on its webpage the highest level of disclosure on environment and community issues. The fact that three companies (C1, C3, and C4) involve the communication and marketing departments in the preparation of the SR demonstrates the importance given to how information on sustainability is communicated and the need to manage the image of the company that is conveyed through it. Our results suggest that concerns about how sustainability may affect the company's image and reputation are not only concerns of listed companies or companies operating in environmental sensitive sectors [41,50,51] but also of non-listed companies, complementing the results of Dias et al. [20] and Eugénio et al. [17] for the Portuguese context. It should be noted that the interviews did not reveal concerns with political costs, which tend to be present only in listed companies and/or companies belonging to regulated sectors, which is not the case for the companies analyzed.
One company (C3) mentioned that the SR is used to give visibility to the projects in which it is involved. This is also the company that discloses on the webpage more information about human resources, especially the granting of monetary and non-monetary benefits to its employees. In addition to being a legitimization strategy in the local community, these reporting practices can also be considered as "signals" the company wants to give about its sustainability performance, in particular with regard to human resources, to attract new employees. Previous studies applied signaling theory to explain the "adoption" of voluntary sustainability reporting, more precisely, the voluntary preparation of the SR [35,60,61]. Our results add to this evidence, showing signaling as an explanation for the content of voluntary sustainability disclosures.
Shareholders and banks are stakeholders mentioned by only two companies (C2 and C5), which is consistent with the low importance that seems to be attributed to sustainability information by capital providers, especially in non-listed companies [9], although in Portugal non-listed companies are financed mainly by banks. Considering agency theory, one of the reasons that may justify this result is the lower information asymmetry and the fewer agency conflicts existing in non-listed companies, which reduces the importance of publicly available information [66]. Another reason may be that shareholders and banks do not incorporate sustainability information in their decision-making, as they consider it neither relevant nor reliable [9,67]. A final reason is that sustainability information may convey proprietary information aspects, which are seen by capital providers as an increased risk factor for the company [68]. The analysis carried out does not allow us to distinguish between these reasons. However, the analysis of the location of the SR on the webpages showed that only one of the companies (C4) presented the SR along with the annual report and this was the only company whose interviewee was a person belonging to the financial department. Moreover, no company mentioned the involvement of the financial department in the preparation of the SR. This evidence suggests that companies treat traditional financial reporting and sustainability reporting separately. The former is prepared by the financial department and seems to be mainly directed towards the providers of capital, considered as the "traditional" stakeholders. The latter is aimed at a wide range of stakeholders, and, in the absence of a sustainability department (as in company C5), it emerges from several other departments (namely, quality, environment, human resources, marketing, and/or communication).
Regarding the content and format of the SRs, it was observed that companies are familiar with the concepts and frameworks related to sustainability reporting. Companies also mentioned that the frameworks followed resulted from benchmarking the practices of other companies (C1), or using the frameworks imposed by the parent company and/or by an external consultant (C2 and C3), suggesting the presence of coercive, normative, and mimetic isomorphisms that shape the sustainability reporting practices of non-listed companies.
When questioned about the disadvantages of the preparation of the SR, in particular whether they consider that this has brought about increased costs for the company, the companies are unanimous in saying that the information was already produced internally, and it was only necessary to systematize it and present it in a more attractive way. This evidence is not sufficient for us to conclude that there are no proprietary costs, as it would be necessary to question companies that do not prepare the SR on why they do not do so [89]. Although the companies do not consider that internal changes were necessary to prepare the SR, in the interviews, two companies (C2 and C5) mentioned positive effects of SR preparation at the organizational level, in line with previous evidence [11,12,81]. Company C5 mentioned a change in mentality in some departments that were initially reluctant to disclose certain information, and another company (C2) referred the creation of a dynamic of continuous improvement in terms of sustainability practices.

Conclusions
Stakeholders are at the heart of sustainability reporting for non-listed companies, either because some stakeholders require companies to prepare a sustainability report or because companies want to present themselves as socially and environmentally responsible to the various stakeholders with whom they engage. As in listed companies, stakeholder, legitimacy, signaling, and institutional theories provide explanations for voluntary sustain-ability reporting in the companies analyzed. By contrast, agency and political costs theories do not seem to apply in this context.
The results also showed that companies treat traditional financial reporting and sustainability reporting separately, as sustainability reports preparation involves several departments (sustainability, quality, environment, human resources, marketing, and/or communication), but no company mentioned the financial department. Additionally, on the webpages, sustainability reports and annual reports tend to be presented in separate sections.
As for the practices adopted, the evidence showed companies know and apply frameworks such as GRI Standards and SDG in the preparation of their sustainability reports. The frameworks applied resulted from imposition by the parent company or external consultant, or from benchmarking of other companies, suggesting that coercive, normative, and mimetic isomorphisms may shape the sustainability reporting practices of non-listed companies.
A proposal for a new Directive on Corporate Sustainability Reporting [24] is now on the table, providing for an extension of the current scope of mandatory sustainability reporting to include all large companies (listed and non-listed) and all listed companies (except listed micro-enterprises). In this context, and although our results are not generalizable, our study has important practical implications. It shows that companies that prepare a sustainability report voluntarily do so within a context of motivations that is coherent with the environment in which they operate, having already identified the stakeholders they are targeting. Furthermore, the companies did not consider that the preparation of the sustainability report entailed additional costs, given that the information on sustainability has been already prepared internally and only needed to be systematized. However, there is still some way to go in terms of integrating sustainability information with financial information and in the certification of information required by the new Directive.
Despite its contributions, this study is not without limitations because the results cannot be generalized to all non-listed companies and we did not conduct an in-depth analysis of the sustainability reports, due to limitations in their availability. We suggest for future research a more comprehensive survey of preparedness of companies for the requirements of the new Directive, involving a broader analysis of voluntary sustainability reporting practices of large non-listed companies, through content analysis of sustainability reports and webpages. It would also be pertinent to conduct a study on the reasons why large non-listed companies have not yet started the process of voluntarily preparing a sustainability report, based on interviews. Finally, we suggest a multiple case study in the first year of the new Directive, with both types of companies, to analyze possible differences in the implementation process. Funding: This work was financially supported by the research unit on Governance, Competitiveness and Public Policy (UIDB/04058/2020) + (UIDP/04058/2020), funded by national funds through FCT-Fundação para a Ciência e a Tecnologia.

Conflicts of Interest:
The authors declare no conflict of interest. Environmental policies or company concern for the environment 0 0 1 1 1 3

Appendix A
Environmental management, systems, and audit