Introduction

International standards are an important mechanism contributing to facilitation of trade, spread of knowledge, sharing of technological advances, and management practices. The standards address a large spectrum of issues ranging from highly technical issues (i.e., normalized screws, procedures for material testing methods) to issues that are social and environmental in nature (i.e., standards for fair trade, social responsibility or sustainable development). The research into standardization has investigated various aspects of standardization including standards development (Castka and Balzarova 2008a; Perry and Noelke 2005; Richardson 2009), standards adoption (Balzarova and Castka 2008; Beck and Walgenbach 2005; Finnemore 1996; Power and Terziovski 2007) and the benefits of standardization (Corbett et al. 2005; King and Lenox 2000). Inarguably, our understanding of established standards is relatively rich.Footnote 1 Yet what seems to be the most recent, the most critical and comparatively less covered by researchers is the development of international standards for social and environmental issues (Tamm Halström and Boström 2010).

It has been suggested that for social and environmental types of standards, constructive partnerships of multiple stakeholders are beneficial for their potential of effective consensus building, knowledge sharing, interest representation, and achievement of legitimacy (Fransen and Kolk 2007; Dejean et al. 2004) resulting in better addressing the complexity of social and environmental issues. Indeed, a lot of relatively recent social and environmental standards were developed through constructive partnerships (Rangan et al. 2006) between private and public players. Such standards are referred to as multi-stakeholder standards (Fransen and Kolk 2007).Footnote 2 Examples of these multi-stakeholder standards include: Standards for sustainable forestry (FSC—Forestry Stewardship Council), sustainable fishing (MSC, Marine Stewardship Council) or fair trade (Fair Trade Mark by Fairtrade Labeling Organization International). These standards typically emerged through a partnership between non-governmental organizations (NGOs) and large multinational corporations. FSC standards, for instance, were created through a collaborative network including Greenpeace, Rainforest Alliance, and Home Depot (Conroy 2007). Leadership in Energy and Environmental Design is another example of a constructive partnership of the Natural Resources Defense Council, the Rocky Mountain Institute as well as others.

The multi-stakeholder standards continue to evolve (Smith and Fischlein 2010). In fact, various experimental approaches are being explored and new multi-stakeholder arrangements are emerging. Scholars recognize the increasing diversity of approaches and have begun to provide taxonomies of multi-stakeholder standards. For instance, Fransen and Kolk (2007) suggest that multi-stakeholder standards can be distinguished based on their approach to stakeholder participation: Some standards are developed by broad involvement of stakeholders while others adopt a consultative approach. Standards setting organizations, such as ISEAL, also recognizes “involvement” and “consultation” as different models for governance of standards (ISEAL 2010). Tamm Halström and Boström (2010), furthermore, note that there is a difference between standards setting organizations: Some standards are developed through a new committee established in an existing standards setting organization (i.e., Committees within International Organization for Standardization—ISO), while others are developed through new standards setting organizations established for one specific task (i.e., FSC)

Given that this multi-stakeholder standardization seems to be the preferred approach to develop social and environmental standards, what then is our current understanding of this phenomenon? The research suggests that these standards represent a new form of private governance, which acts as a counterpart to state authority (Mörth 2004). This, typically emerges in areas, where conventional state-based legislation is ineffective (Garrett 1998) and where multiple and often globally dispersed players are involved (Brunsson and Jacobsson 2000; Djelic and Quack 2003). Standards developers are portrayed as multiple actors with specific interests, perceptions, strategies, and goals (Tamm Halström and Boström 2010), who search for legitimacy, power, reputation, and resources (Boström 2006), and who agree to new standards based on contestation and conflict (Botzem and Quack 2006). Several other studies have provided insight into stakeholder involvement in particular standards (i.e., FSC, MSC, and Fairtrade standard setting is described in Conroy 2007; Tamm Halström and Boström 2010; Smith and Fischlein 2010; Cashore et al. 2006).

Despite the growing number of studies, multi-stakeholder standardization research, and our understanding of stakeholders’ involvement in these standards is still in its infancy (Fransen and Kolk 2007). First, most studies in this area are conceptual works that provide useful, yet only, preliminary ideas and concepts. There is, however, a lack of empirical evidence to support the concepts. Some researchers are also recognizing that more conceptualization and theory building is necessary (Smith and Fischlein 2010). Second, the studies mainly analyze multi-stakeholder standards, which were developed through newly established standards setting organizations. Comparatively less attention has been paid to standards that are developed through established standards setting organizations. The exceptions are the studies by Castka and Balzarova (2008a, b, c) and Tamm Halström and Boström (2010). Due to the fact that this is a relatively recent phenomenon, even these studies draw upon limited empirical evidence or are merely conceptual works themselves. Hence, this study aims to address these two obvious gaps in the literature. Therefore, this study provides an empirical investigation of a multi-stakeholder standards development of ISO 26000 as an example of a multi-stakeholder standard that is developed through a new committee established in existing standard-setting organization. For this reason, we have posed the question:

How do multiple stakeholders engaged in standards development influence and contribute to the developmental process of a new standard?

The article is organized in the following manner. First, we discuss standards development in general. In the background of this discussion, we present the evolution of standardization within the ISO. We do so to demonstrate the unique nature of our case and to describe the differences in multi-stakeholder standardization. Second, we explain our methodology. Third, we present the results of our research and demonstrate how stakeholders influence and contribute to multi-stakeholder standards development. Fourth, we discuss our results, provide a set of propositions, discuss the limitations of our study and provide suggestions for further research.

Development of International Standards—Toward Multi-Stakeholder Standardization

Standards are developed based on a set of rules that are determined by a standard-setting organization. For instance, the ISO determines its standard setting in ISO/IEC Directives and also offers standards for “the development of standards” (i.e., ISO/IEC Guide 59). Similarly, World Trade Organization (WTO) provides a code of good practice for preparation, adoption, and application of standards. Other standards setting organizations follow similar processes and in some instances even mirror ISO and WTO’s guidelines (for instance, ISEAL builds substantially on and refers to both ISO Guide 59 and WTO’s Technical Barriers to Trade Agreement Annex 3). The actual developmental process then consists of several stages typically including preparatory stage, committee stage, enquiry stage, approval stage, and publication stage.

Despite a presence of well established guidelines for standards development, standards setting organizations were often accused of a lack of transparency and a lack of involvement of multiple stakeholders. ISO standards are a good example of this common criticism. ISO is a worldwide federation of national standards bodies (NSB) and develops international standards that are required by the market for the purpose of facilitation of trade, spread of knowledge, sharing of technological advances and management practices (ISO/TMB/WG/SR 2006). ISO was founded in 1946 by delegates from 25 countries and has grown to a network of 163 national member countries. According to Bryant (2007), ISO offers a portfolio of some 16,500 standards that cover product specifications, services, test methods, conformity assessment, as well as management, and organizational practices. Yet despite this expertise and global outreach, the standards setting process by ISO has often been criticized for under-representing minority groups and NSB from developing countries (Tamm Hallström 2000). ISO standardization has also been portrayed as a platform to serve the industry, rather than creating standards for public good (Ecologia 2007).

Recent developments in environmental and social standardization are indicative of ISO’s inability to mobilize a wider stakeholder base for its standards development. Indeed, new standards setting organizations entered the field of standardization and introduced multi-stakeholder approach for the development of social and environmental standards. For instance, several greenhouse gases (GHG) emissions standards were developed through multi-stakeholder dialogue, such as: The gold standard initiated by a number of NGOs, including World Wild Fund and SouthSouthNorth or the Voluntary Carbon Standard initiated through the International Emissions Trading Association and World Economic Forum. Smith and Fischlein (2010) report that over 20 voluntary standards for GHG are now in operation. ISO has also developed a set of standards for GHG emissions (ISO 14064/1/2/3, 14065). Yet ISO used its existing committees for the development—committees that were previously criticized for the lack of multiple stakeholder involvement. Coincidently, the uptake of ISO standards for GHG emissions has been marginal in comparison to other ISO standards and in comparison to other GHG standards.

In response to multi-stakeholder standards setting efforts elsewhere, the ISO has also introduced multi-stakeholder standards development. A representative example is the development of ISO 26000 standard for social responsibility. The build-up process spanned between 2002 and 2004, and included an analysis of stakeholders’ viewpoints by ISO Advisory Group on social responsibility (ISO/AG/SR 2004a, b; ISO/COPOLCO 2002). The analysis revealed controversial positions of stakeholders regarding certification (Castka and Balzarova 2008a).Footnote 3 Consequently, certification was dropped in favor of creating a guiding standard—a move that provided an acceptable platform for multiple stakeholders. After this build-up, ISO created a new multi-stakeholder called Working Group on social responsibility, which was charged with developing the “International Guidance Standard on Social Responsibility—ISO 26000”. Established in 2005, the SR Working Group involved approximately 300 nominated experts from fifty-four ISO member countries and thirty-three liaison organizations, which represent six main stakeholder groups: Industry, Government, Consumer, Labor, NGOs and Service, Support, Research & Others (SSRO) representatives. This Working Group is considered as one of the biggest and most diverse groups ever created by the ISO (ISO/TMB/WG/SR 2006).

The development of ISO 26000 represents another milestone in multi-stakeholder standards development. Social and environmental standards, which were developed before ISO 26000 (such as FSC, MSC, etc.), provide a platform for engagement with multiple stakeholders, where the developmental process was typically initiated and controlled by a relatively narrow set of actors. These standards were also developed through new standards setting organizations that were established for one specific task (Tamm Halström and Boström 2010). ISO 26000, on the other hand, provides a much broader platform for stakeholder engagement as the scope of the standard is broader and the development is handled within an established standard-setting organization. Furthermore, ISO 26000 standard is a guiding standard without certification. The certification requirement has been incorporated into the development and reinforcement process of other standards (such as FSC; Conroy 2007) but is absent from the development of ISO 26000. The differences are summarized in Table 1 and further elaboration on this topic is provided in the discussion section of the article.

Table 1 A comparison of recent multi-stakeholder standards development
Full size table

Research Enquiry

Our enquiry is designed as a single-case holistic study. Yin (1994) argues that there are three rationales that justify a single-case design: It should be critical, extreme or unique and revelatory. Our research enquiry conforms to each of these requirements. It is revelatory because it allows the study of a phenomenon that was previously inaccessible to scientific investigation. ISO 26000 provided a public access to comments that were submitted by stakeholders during the development. Before this, no such data has been made publicly available. ISO is one of the most prominent standards setting organizations and ISO 26000 standards development is ISO’s first attempt to develop standards with multiple stakeholders. The inclusion of multiple stakeholders in ISO standardization is a remarkably liberal shift in what has historically been conservative institutional behavior. Given ISO’s global outreach and influence, this case also exhibits the unique features of a critical case (Yin 1994), coupled with a potential to represent a significant contribution to knowledge and theory building in the standards development arena.

Our research aims to determine how multiple stakeholders engaged in standards development attempt to influence and contribute to the developmental process. The focus of the research and the empirical context is the standards development through an established standards setting organization (the ISO). We use longitudinal in-depth archival data from the development of ISO 26000—a guidance standard for social responsibility. To offer a critical perspective, we have endeavored to explore six stakeholder groups involved in the ISO 26000 standards development: Industry, Government, Consumer, Labor, NGOs, and SSRO (Table 2).

Table 2 Stakeholder Representation in ISO 26000 development
Full size table

The data, which was publicly available during the developmental process, is available in the form of “comments”. Hence, our unit of analysis (Weber 1985) is a comment submitted by a stakeholder. During the developmental process, each participating stakeholder had an opportunity to submit such comments in response to a Working Draft of a standard. As such, the comments represent stakeholders’ attempts to influence as well as to contribute to a new standard. For this reason, we have analyzed the comments from two angles. First, we have analyzed the acceptance/non-acceptance of comments. The acceptance/non-acceptance was determined by tracking whether the comments influenced the subsequent Working Draft. Second, we investigated the type of comments and by clustering similar types of comments we have determined key processes in standards development. The more detailed research process is discussed in the following sections.

Commenting Process in Standards Development

There were typically thousands of comments submitted for each Working Draft. To accommodate these comments, the ISO/TMB/WG SR established a procedure to integrate the comments into each subsequent Working Draft. The process of integration of comments is described in two procedures; N79 (The process of Drafting standard text in TG4, TG 5 and TG 6) and N108 (Editing Committee Drafting Guidance). The commenting process is described in Figure 1.

Fig. 1
figure 1

Commenting process for ISO 26000

Full size image

The Data Set

For the purposes of this study, we have tracked stakeholders’ comments on Working Draft N55. We chose one Working Draft because by the time of N55, the number of participating stakeholders stabilized, the representation of various stakeholder reached an equitable level and operational procedures were be agreed upon. Moreover, during this time, the overall design of the standard was decided. This is the period in time where the stakeholder groups were most likely to have diverse positions hence providing us with a good understanding of the processes stakeholders used to influence and contribute to standard’s development. Our data set includes all of the stakeholders’ comments submitted on Working Draft N55 (see Table 3).

Table 3 The data set
Full size table

Data Analysis—Analysis of Comments’ Acceptance

The analysis of comments’ acceptance was performed on all comments from our data set. First, we studied a comment and then investigated the draft upon which the comment was submitted to understand the context of the enquiry. Second, we investigated the new draft to assess whether the issue was accepted or not.

Following the common practice of media research, we coded each comment as accepted, neutral or non-accepted and assigned each comment a value of 1, 0 or −1 (Deephouse 2000; Pollock and Rindova 2003; Pollock et al. 2008; Weber 1985, Janis and Fadner 1965). An “Accepted” comment was defined as an instance in which the proposed change specified in a comment was accepted entirely or the comment had a significant influence on the subsequent Working Draft. The following instances were considered as acceptance:

  • Stakeholder proposed a wording for a clause or a new section in the standard, which appears in the subsequent draft;

  • Stakeholders were asked to select their preference from a list of options in N55 and the selection was carried forward;

  • Stakeholder supported propositions from N55 and those propositions were carried forward;

  • Stakeholder proposed several changes majority of which was accepted.

A “neutral” rating received comments, which had some influence on the draft, but were not considered fully. Neutral comments were rated according to the following procedure:

  • Stakeholders’ suggestion was accepted partially (some issues were accepted and some were not; the balance of issues was about equal).

  • Stakeholder comments that did not suggest any change; for instance, some comments encouraged further dialogue or asked for further clarification without proposing an actual change

On the opposite side of the spectrum, “non-accepted” comments were those rejected entirely or in its majority. The instances include:

  • The proposed change was clearly rejected (i.e., stakeholder proposed new heading or section of the standard, which is not accepted).

  • A comment contains a mix of accepted and non-accepted issues, where the majority of issues having been rejected.

The coding was performed by both researchers. Each researcher has been assigned a set of comments to code. A random sample of 20% of the comments was assigned to both researchers, whereas the rest of the comments were shared between the two researchers. The sample of 20% was used to determine inter-rater reliability (IRR) of the coding. The IRR result was 0.89, which exceeds the common threshold reported in similar studies (Deephouse 2000; Pollock and Rindova 2003; Pollock et al. 2008).

The acceptance of comments from stakeholder groups was determined by the coefficient of acceptance—calculated using the Janis–Fadner coefficient of imbalance:

$$ {\text{Acceptance }} = \, \left( {a^{2} - a\; * \;n} \right)/\left( {total} \right)^{2} {\text{if }}a > n; \, 0{\text{ if }}a = n; \, \left( {a\; * \;n - n^{2} } \right)/\left( {total} \right)^{2} {\text{if }}n > a $$

where a is the number of accepted comments from each stakeholder group, n is the number of non-accepted comments from each stakeholder group, and total is the total number of comments from the stakeholder group. The range of this variable is (−1, 1), where 1 indicates all accepted comments and −1 all non-accepted comments.

Data Analysis—Determination of Key Processes

The second part of the study involved a qualitative analysis of the comments. We started the analysis by randomly selecting 20% of the data from our data set and analyzed it through the lens of our research question (Santos and Eisenhardt 2009)—how do multiple stakeholders engaged in standards development attempt to influence and contribute to the developmental process? This approach reflected our position that the research topic represents a revelatory, unique, and critical case. Our first step was to identify central themes or strands that would symbolize stakeholders’ efforts to influence or contribute to standard development.

The comments were uploaded to NVivo and we employed an “open” or emergent coding procedure, allowing the data to “speak” (Huxham and Vangen 2000). We used “free nodes” in NVivo to label the central themes and have identified thirteen free nodes. At this stage, we asked an independent coder to code the data to test inter-coder reliability (Miles and Huberman 1994). The comparison of the results from all coders showed a high level of consistency.

In the next step, we have clustered our thirteen free nodes into groups with overarching patterns and similar themes. This process had several iterations and each time we cross-checked the original data for accuracy. As a result of the in-depth analysis, we had five parent nodes, which we labeled as processes. After this exercise, we had a complete coding scheme to code the remaining 80% of the data from our data set. This approach is consistent with similar methodologies used elsewhere (Druskat and Wheeler 2003; Balzarova and Castka 2008).

Results

In this section, we first establish whether the comments from stakeholders had actually influenced the subsequent draft and also whether there were any differences between stakeholder groups in terms of the acceptance of their comments. We debate the findings from two angles: We present the quantitative results as determined by the coefficient of acceptance and also take a qualitative approach and investigate the nature of accepted and non-accepted comments. Second, we describe stakeholders’ involvement in the standards development through a set of five processes. We describe each process in detail and also highlight differences between stakeholders’ positions within these processes.

Acceptance of Comments from Stakeholders Groups

The quantitative results of acceptance are presented in Table 4. The results show marginal differences and each stakeholder group has a fairly similar ratio of accepted and non-accepted comments. The highest acceptance rate is attributed to Labor, NGOs, and SSRO; lowest acceptance of comments was the ones submitted jointly by all stakeholder groups and Mirror Committees. It should be also noted that in terms of the volume of comments, Industry is the most active stakeholder group. Given the fact that the ratio of acceptance is similar between stakeholder groups, we conclude that the higher volume makes Industry the most influential stakeholder.

Table 4 Acceptance of comments from stakeholder groups
Full size table

A qualitative analysis of the results revealed the following findings. First, standards developers were mostly striving to reach a consensus among stakeholders by incorporating their different viewpoints. This finding can be supported by, for instance, a set of comments submitted for Clause 1. The N55 Draft proposed three options for how Clause 1 should be structured (named Option 1, 2, and 3). There were differences between options, but about 80% of the content was similar. Stakeholders were asked to indicate their preferred choice. In general, stakeholders were mostly in favor of Options 1 and 2. The result is that the subsequent draft is based largely on Option 1, but also incorporates some issues from other options. This example clearly demonstrates the search for consensus in the standard’s development.

The example above also underpins our second finding: stakeholder groups were not always unanimous in their preferences. For instance, Industry supported, almost equally, all three Options for Clause 1. In this particular example, stakeholder groups were rather heterogeneous in their viewpoints. However, in other areas, stakeholders appeared quite unified and maintained homogeneous viewpoints (i.e., Industry opposing certification; Consumer proposing conformity assessment).

Third, stakeholders’ sphere of influence and knowledge in particular areas leads to higher acceptance of comments. For instance, Clause 6.7 deals with Consumer issues and Consumer’s recommendations were mainly accepted. Other stakeholders contributed to the shape of the section, but their comments had higher rejection rates. Other examples include the influence of Labor on Labor related issues and NGOs’ influence of organizational governance sections.

Fourth, comments proposing changes to previously agreed format of the standard were mostly rejected. Such comments include, for instance: Changes to the name of the standard; suggestions to rename established sections of ISO standards (a change of “Normative References” to “Informative References); suggestions to incorporate “management systems” approach or continuous improvement into the standard; push for conformity assessment; definition of minimum requirements. The issues related to certification and verification, often raised by Consumer, were mostly rejected. On the other hand, stakeholders also used previously agreed aspects of standards as a leverage and reference points for their suggestions. A good example here is Labor, which used the Memorandum of Understanding between ISO and ILO to strengthen linkages to and alignment with ILO documents.

Finally, comments that proposed inclusion of specific standards or frameworks were not always accepted. The standard developers mostly accepted stakeholders’ comments looking to include intergovernmental instruments and major frameworks (i.e., OECD Guidelines for Multinational Enterprises; UN Global Compact; ILO Declarations; ISO 14000 series; GRI Sustainability Reporting Guidelines; AA1000 and BS8900). However, many other standards and frameworks were not accepted; such as national standards or private CSR tools and instruments.Footnote 4

Processes that Influence and Contribute to Standards Development

Our research suggests that stakeholders’ involvement can be described by five distinguishable processes that cover eliminating, linking, consensus seeking, reinforcing, and standards improving (Table 5). In a nutshell, stakeholders attempted to (1) eliminate issues that are controversial and may lead to consequences perceived by stakeholders as undesirable; (2) link and integrate the standard into a network of other documents and ISO standards; (3) seek consensus by highlighting areas for further dialogue or by addressing the exclusion from the standards development; (4) reinforce issues that are important; and (5) improve the content of the new standard.

Table 5 Overview of results: key processes in multi-stakeholder standards development
Full size table

Table 5 also quantifies the percentage of comments that fall into each process. From this perspective, the most comments were used to improve the content of the standard; whereas, the fewest comments were used to seek consensus. This result is not surprising. Improving the content of the standard is by far the most detailed aspect of the standards development. In practice, it means that stakeholders comment on each clause of the standard and suggest changes (such changes can include, for instance, different wording, targeting repetitions, etc.). Inevitably, these comments are more detailed and more abundant than other comments. On the other hand, comments seeking consensus are less numerous. This is mostly because stakeholders had other means of communication to seek consensus; i.e., during plenary sessions, formal meetings and informal dialogues (personally or over e-mail lists). We now focus upon each process in more detail. Within each process, we describe the key themes as summarized in Table 5.

Eliminating

The process of elimination involves attempts to moderate controversial issues, which stakeholders perceived as problematic and may lead to undesirable consequences. Based on the data from our study, the controversial issues can be grouped into three central themes: legislative tension, business bias and misuse. These three themes describe the key tension points, which stakeholders commonly raised.

The first theme in this process is labeled legislative tension. This theme deals with the relationship between the standard and its linkage to legislation. Here, there was much variation on the positions of stakeholders, which ranged from the complete exclusion of the legislation to the complete inclusion of the legislation into the standard. Industry, for instance, sought exclusion of the legal compliance. A typical comment from Industry stated that “the relationship between [the standard] and legal compliance must clearly state that [the standard] has no relation with legal compliance.” Conversely, the SSRO stakeholder group held the opposite position, for instance by arguing that “compliance with the law is the baseline obligation for organizations in meeting the expectations of society”. NGOs and Labor stakeholder groups, on the other hand, were more inquisitive and mostly attempted to point at the need to resolve the relationship between the standards and the law. The subsequent Working Draft contains a statement that “a fundamental principle of social responsibility is respect for the rule of law.”

The second central theme in this process addresses the issue of business bias in standardization. In other words, stakeholders were striving to create a standard suitable for all organizations not only for business organizations. All stakeholder groups showed consensus that a bias toward business organizations was not desirable—inclusive of the Industry stakeholder. All comments aimed to shift the standard from being viewed as a tailored product for Industry toward a more generic product. This could be demonstrated by one of the Industry comments stating that “this section has been written with a strong Industry focus, and with a good deal of Industry related terminology. The whole section should be reviewed to take away this bias”. Similar comments were submitted by other stakeholders and such comments clearly show the effort that stakeholders made to develop a standard for all types of organizations.

The third theme centers around potential misuse of the standard. Stakeholders raised various issues: i.e., the misuse of standards to create trade barriers (Industry), credibility of standards users (Consumer and NGOs), using the standard to substitute other international standards and agreements (labor). Here, each stakeholder group was striving to safeguard their agendas. For instance, the Consumer stakeholder group tried to protect Consumers and argued that mechanisms should be in place to verify organizations’ claims in relation to the individual standards. NGOs highlighted the fact that organizations can cleverly adopt only the parts of the standard that suited their purpose.

Linking

Linking represents a process by which stakeholders were striving to create references between the standard under development and existing standards, tools, guidelines or international instruments. This means that a particular aspect of the standard, which has been developed and established previously elsewhere, could be referred to in the standard being developed. For instance, the standard promotes “reporting” as a mechanism to communicate the social responsibility performance of an organization to the wider public. The standard can either explain the fundamentals of reporting or can refer to, for example, “GRI Guidelines” as a relevant methodology to satisfy this particular aspect. It should be noted here that historically international standards developed by ISO have not provided references to other materials except to other ISO standards. However, in ISO 26000 development directives allowed for creation of links to external documents and our data shows that stakeholders showed considerable activism in this aspect of standards development.

The data shows two themes related to “linking.” Firstly, stakeholders strived to create links to other ISO standards. Typical examples include links to ISO 9000 for quality management systems, ISO 14000 for environmental management systems and ISO 14063 for environmental communication. Links to ISO 14000 series were mostly accepted, though this was not the case with ISO 9000 where links were not accepted. Secondly, stakeholders attempted to create links to external documents (tools, instruments, guidelines, intergovernmental instruments, etc.). A large number of intergovernmental, as well as private instruments were put forward, such as: GRI Guidelines, UN Guidelines, ILO Declarations and others. While promoting the links, stakeholders provided various rationales. Some stakeholders pointed at the legitimacy and relevance of certain tools. Multiple comments supported GRI Guidelines, for instance by stating that “GRI Guidelines have been one of the most widely recognized and endorsed voluntary, global, multi-stakeholder initiatives in the area of social responsibility”. While, other stakeholders emphasized the need for the documents to be recognized by intergovernmental bodies (UN and ILO). Yet, other participants ‘offered’ their unique tools; “we are pleased to share a ‘Corporate Social Responsibility Management Model’ developed by Vincular-CSR”. Although the final version of ISO 26000 provides reference to many tools and guidelines, only a few were accepted at this stage of standards development. For example, the subsequent Working Draft N80 contains only few references to external documents, acceptance of which has been discussed in the previous section of this paper.

Consensus Seeking

The third process that typifies stakeholders’ actions is consensus seeking. The ISO standards development process is based on a consensus of involved parties. Voting is not accepted during the development; therefore, stakeholders have to seek consensus. In ISO standards development, consensus is characterized by the absence of sustained opposition to substantial issues.

Our data suggests that there are two major themes within this process. Firstly, stakeholders encouraged further dialogue and highlighted the issues that need to be discussed. This often meant that a particular stakeholder group did not hold a strong or definitive position over the issue. Yet, the issue is highlighted and put firmly on the agenda. For example, a comment would state that “this issue has not been debated; further discussions are necessary”. Secondly, stakeholders seek procedural justice in the development process and highlight issues that were excluded without a consensus. For instance, stakeholders used commenting to communicate to other stakeholders that some comments were excluded from the decision making process (“our comments were ignored”) and highlight their dissatisfaction with the transparency of the decision-making. There were however the least amount of comments in this category. It seems that stakeholders use other means to seek consensus or to raise exclusion from the developmental process (such as plenary meetings, TG meetings and communications), which concludes that commenting was used in a limited way for the purpose of consensus seeking. Especially in terms of the exclusion from the developmental process, it seems that commenting was used as the last resort to raise concerns if other methods failed.

Reinforcing

Reinforcing represents a process by which stakeholders promoted their specific position in regards to the design of the new standard. This process is dominated by three themes. Firstly, stakeholders debated about the appropriate level of strictness. Central to this debate was the question of whether or not organizations should be allowed to choose the scope of their socially responsible activities and whether certification (or some form of verification) is even needed. Many stakeholders perceived that the choices related to the scope of activities could be seen as a “downsizing” of the standard and expressed their worries; “… it worries me when it is said that organizations ‘should develop its SR vision statement’ focusing on issues they ‘want to deal with’.” NGOs highlighted the need to include organizational governance in the scope of the standard. Though, this was contradicted by Industry as the following comment demonstrates:

In some parts of the world, organizational governance is highly regulated and it will be very difficult to insure no contradiction occurs between what is written here and the various regulatory regimes. For this reason, organizational governance should be outside the scope of the standard.

While discussing the strictness of the standard, some stakeholders again raised the issue of credibility and significance of a guiding standard, namely Government and Consumer. Consumers again suggested that third party certification should be considered to raise the bar for adopting organizations. The acceptance levels between stakeholders in this process varied: certification related suggestions were not accepted (Consumers for and Industry against), push to substitute ‘supply chains’ with ‘sphere of influence’ was not accepted either (proposed by Industry) and organizational governance became a part of the standard (NGO for and Industry against).

The second theme deals with the issue of looseness. Stakeholders highlighted the need to allow for flexible interpretations based on the standard and supported the idea that the standard should not be too prescriptive. This issue was of particular importance to Industry, who submitted the most of their comments in relation to this issue. Industry argued for practicability and prioritization. A representative comment would state that “It is impracticable for an organization to deal with every entity that may consider to be a stakeholder. A degree of selection or prioritization is necessary”. Other stakeholders focused on making the standard less restrictive, for instance by stating that “Limiting the definition to “core” business processes is unnecessarily restrictive”. The remaining stakeholders reinforced the need to explain what ‘flexibility’ means, rather than purely encouraging flexibility per se.

The third issue in this process is voluntarism. The stakeholders mostly agreed that the standard should be voluntary, yet they pinpoint different aspects of voluntarism. Industry, for instance, called for a clear statement that the standard should be voluntary and not intended for third party certification. They even called for adjustment in the language of the standard; “We suggest to substitute could for should”. Other stakeholders reinforced the need to frame and to explain voluntarism. For Labor, it meant a clear distinction between voluntary initiatives and governmental instruments; “Voluntary initiatives must not be accorded the same status as authoritative intergovernmental instruments and this distinction should always be made throughout the document.” For Consumers, it was important that there was a commitment once a decision was made; “It should be emphasized that while SR is a voluntary activity of organizations, once they commit themselves to engage in SR and make claims about such activities, [then] they need to comply with some core principles.” For NGOs, the relationship with the law was pivotal; “… [to] clarify the nature of the relationship between social responsibility and legal compliance, [a standard must] include a provision for whether social responsibility is strictly a ‘voluntary’ concept or whether it includes compliance with the law.”

Standards Improving

The fifth process embraces activities by which stakeholders were striving to improve the content of the new standard. Our data shows that stakeholders paid a lot of attention to improving the text of the standard—more than half of the comments in our sample dealt with such improvements. These comments can be clustered into three themes. The first theme is the issue of text clarity. These comments typically highlighted areas in the standard that were perceived as ambiguous or incomprehensible. Typically, a comment highlighted a term or a sentence and explained why this could be misunderstood by the users of the standard.

The second theme was the issue of parsimony, where stakeholders argued for a shorter and more succinct standard. The comments highlighted areas of repetition and duplication, called for more effective language and suggestions were made in areas where a conclusion could be made based on the text; “It is not necessary to mention organization’s own interests. It is not likely that an organization would act in contrary to it is own interests.”

The third theme deals with usefulness. Here, the comments highlighted the areas in the text, which were seen as not adding value to the user. For instance, stakeholders pointed at several issues that might be useful to the users: support in decision making (“there should be some information that allows users of the Standard to evaluate the credibility and usability of the mentioned instrument”) or support in addressing various clauses of the standard (“This could be especially useful to organizations working on their SR vision statements”).

Discussion: Stakeholders’ Influence and Contribution to Development of Social Standards

 

Our research demonstrates how multiple-stakeholders influence and contribute to standards development. In our discussion, we firstly mention the influence of pre-standardization agreements, then move on to discuss the influence and contribution of stakeholders during the developmental process. This is concluded with a discussion of multi-stakeholder standardization in general. We observe that:

In multi-stakeholder standards development, through a new committee established in existing standards setting organization, the pre-standardization arrangements create boundaries that limit stakeholder influence during the developmental process.

Our research has demonstrated that stakeholders often used the pre-standardization arrangementsFootnote 5 to leverage their positions and arguments. A good example is the argumentation over certification and its role in the reinforcement of the standard, which reemerges in stakeholders’ arguments.

The pre-standardization discussion among stakeholders showed that there is split of opinion, and not a consensus, over certification (Castka and Balzarova 2008a). Through this process, the idea of certifiable standards was abandoned, which raises several issues. The pre-standardization seems more like an effort to embed new structures and practices into already legitimate institutions (Suchman 1995) and to associate ISO as a standard developer with other more respected entities in this context (Dowling and Pfeffer 1975). This is because ISO’s reputation was tarnished by accusation over the dominant Industry influence in the standardization process. Inclusion of other legitimate entities in standards development, such as ILO, helped to address this issue. The non-certifiable nature of the standard also attracted other entities that would not have usually participated, such as NGOs. This made ISO efforts to appear desirable, proper, or appropriate an obvious attempt to legitimize its standardization effort. On the other hand, had the certification been agreed on as part of the arrangement, the outcome would have been a more stringent standard with an embedded verification mechanism—even though resting on a less legitimate developmental process (Castka and Balzarova 2008c).

Under the constrains of the pre-standardization arrangements, we provide several observations about stakeholders’ influence and contribution to the developmental process as discussed below. We observe that:

Stakeholder groups that are the most likely to be directly affected by the standard, will be the most active participants in the standards development through processes of eliminating, linking, consensus seeking, reinforcing and standards improving .

The most directly affected stakeholder is the Industry stakeholder group. It can be expected that if a standard is enforced, it would be the Industry, which needs to comply with the requirements. The impact of the standard can be positive or negative. For instance, an adopting organizations can signal their otherwise hidden qualities to the market place (Graffin and Ward 2010) or reduce information asymmetries with supply chain partners (King et al. 2005). A new standard can also become a requirement from purchasing organizations or governments allowing for an adopting organization to enjoy preferential treatment (Neumayer and Perkins 2005). On the other hand, a new standard can also be burdensome for organizations. A new standard can be used as a trade barrier, or could increase the cost of production as organizations have to absorb the cost of compliance (Delmas 2002). In a more extreme case, the external audit could reveal regulatory non-compliance in an adopting organization (Delmas 2002). As a result, there is a lot at stake for the industry and the Industry is, therefore, incentivized to take collective action (Olson 1971).

Numerous studies do indeed demonstrate that Industry is the most active participant in standards development. For instance, Perry and Noelke (2005) studied the International Accounting Standards Board’s (IASB) standards development. They observed “the non-participation of organizations outside the commercial sector” despite the formal processes for stakeholder consultation. Carmin et al. (2003) studied stakeholder involvement in the development of US Voluntary Environmental Programs. The results from Carmin et al.’s study accordingly suggest that third party sponsored programs suffer from poor stakeholder diversity in their overall contribution. Our study provides similar conclusions and Industry is the most active participant.

Previous studies highlighted that expertise is a prerequisite to exercise power in standards development (Botzem and Quack 2006; Castka and Balzarova 2008a; Mattli and Buthe 2003; Dejean et al. 2004). It has been suggested that the stakeholders who are knowledgeable about standards setting, have advantage over less experienced participants (Tamm Hallström 2000). In the case of multi-stakeholder standardization, the most experienced groups of stakeholders are Industry and NSB. These stakeholders possess various forms of expertise, such as standard-setting experience, experience with collective action (in line with Olson’s (1971) logic of collective action), and leadership experience from various committees and/or other procedural competence. Hence, these stakeholder groups have the power to become particularly influential—even though, in this instance, power does not necessarily mean power over, but rather power to (Clegg et al. 2006). This leads us to the following proposition:

Stakeholder groups with expertise in standardization will be more active participants in multi-stakeholder standards development through processes of eliminating, linking, consensus seeking, reinforcing and standards improving.

The most active stakeholders are also the most influential. The study reveals that stakeholders are treated equally because the ratio of accepted comments is similar for each stakeholder group. However, the volume of the comments shows significant differences between stakeholders (Table 4). Higher volume ultimately leads to a higher number of accepted comments hence greater influence. Therefore, the ability to mobilize resources and actively participate leads to the uneven influence in standard development. We observe that:

In multi-stakeholder standards development, the most active stakeholders will be the most influential stakeholders through processes of eliminating, linking, consensus seeking, reinforcing and standards improving .

Olson (1971) also suggests that individuals in any group attempting to provide collective goods will have incentives to free ride if the group is working to provide public goods. Previous research demonstrated that some stakeholders free ride in the standardization process or, put another way, use standardization to enhance their credibility, learning, and legitimacy (Tamm Halström and Boström 2010). Bowers (2006) observed that stakeholders in ISO 26000 development entered the process with clear positions: Industry to insure a manageable standard; Labor (represented by ILO) to safeguard their international gains; Governments to observe; and groups like consultants, academics, and certifiers hoped that “a social responsibility standard would generate the same level of business that grew up around ISO 9001 and 14001.” Our research has identified similar agendas to those proposed by Bowers (2006).

Perhaps the most interesting and the most explicit is the process of linking. We proposed that linking represents a process by which stakeholders strive to create links between the standard under development and other standards, tools, guidelines or international instruments. The final version of the ISO 26000 standard indeed contains a large list of examples of voluntary initiatives and tools for social responsibility that were proposed and pushed by the participating stakeholders. We acknowledge the possibility that stakeholders may have a genuine motivation to share these tools. Yet it is also reasonable to conclude that those tools will become points of reference for standardization activities in the future.Footnote 6 Therefore, the ISO 26000 standard-setting process itself can be a legitimating strategy for the groups involved. Similar legitimizing strategies are described in the literature in theoretical terms (Suchman 1995), as enablers for initial public offerings (Pollock and Rindova 2003) or in the standardization literature (Tamm Halström and Boström 2010). In this research, we have identified the process of linking as being an indication of the legitimising efforts. Therefore, we suggest that this mechanism can be attractive to even less active participants as it is linked to potential future benefits. This leads us to the following observation:

In multi-stakeholder standards development, even less active participants will concentrate on activities related to the future benefits (i.e. legitimization) through the process of linking.

Limitations and Suggestions for Future Research

The current study is not without limitations. Our study focuses on a single standard development and relies on a single data type (stakeholder comments on working drafts). Such design allows us to scrutinize the process of standards development and draw conclusions about stakeholders’ influence and contribution to this process. It also allows us to use the data from a large international standardization effort, which was not previously available. However, our design hindered our ability to draw broader conclusions about multi-stakeholder standardization and we propose to enhance our study with further inquiry.

Future studies could investigate standardization efforts other than the standards developments by the ISO. For instance, standards development at the IASB can provide a comparative setting for such inquiry (Perry and Noelke 2005). Such enquiry would provide further generalization of the findings reported in this article and strengthen our understanding of multi-stakeholder standards development through established standards setting organizations.

Multi-stakeholder standards development is not a homogeneous area. There are differences among standardization efforts (Fransen and Kolk 2007). We have summarized the differences in Table 1 and explained that our research is focused specifically at standards setting through an established standards setting organization. Here, we would like to point out that there are differences between the standards setting processes explored in our research and other forms of multi-stakeholder standards development in the hopes of encouraging further research in this area. The differences are discussed in terms of the type of a standard (certifiable standards versus guiding standards), the scope of the standard (narrow versus Industry specific), and the particulars of a standard-setting organization.

The first point of difference is the issue of certification (type of a standard). Our study reports on a multi-stakeholder standard, which does not include certification as part of the standards development process. This specific feature raises an important question: what is the role of guiding standards in social and environmental standardization? A majority of standards in this area were created as certifiable (such as FairTrade and FSC). It is believed that certification supports standards’ credibility and also strengthens standards’ ability to reinforce social and environmental aims (Conroy 2007). The possible roles for ISO 26000 are numerous, yet it is still unclear what purpose the standard will serve in the coming future. More research is needed in this respect, so that it can be determined whether and/or under what circumstances guidance standards are better ways to develop social and environmental standards. Further research should investigate and assess the impact of these two alternative approaches.

The second point of difference is the scope of the standards. ISO 26000 aims to address the social responsibility in the broadest sense. Other standards (such as FairTrade and FSC) tend to be narrowly focused on specific aspects of social responsibility or are aimed at a particular industry sector (Conroy 2007). In the analysis of FairTrade, FSC and other similar standards, Conroy (2007) reports that many standards are initiated by NGOs. NGOs typically target quite narrow social and environmental problems and link them to specific actors (i.e., highly visible violators of such social and environmental problems). The initial steps involve media pressure on the violators, which is followed by some form of agreement and mobilization of resources to achieve change (Smith and Fischlein 2010; Conroy 2007). In comparison, ISO 26000 lacks this focus. The scope of the standard is very broad and accordingly the stakeholder base is broad as well. This raises a question of participation in broadly oriented social and environmental standards: Who and for what reason does get involved in the development of broadly oriented standards such as ISO 26000? We suggest that future studies should investigate this issue further and scrutinize participating and non-participating actors particularly from Industry and NGO stakeholder groups.

The third important point of difference is the background of a standards setting organization. Table 1 has established the difference between standards that are created through newly established standards setters and through established standards setters. Newly established standards setting organizations (such as FSC or MSC) are typically governed by a small number of actors, typically NGOs and Industry partners. Existing standards setting organization such as the ISO, on the other hand, is governed by a membership base consisting of NSBs. NSBs, then, represent their countries. This difference leads to a plethora of interesting questions. For instance, is the role of NSBs (and individual countries) more influential than the stakeholders’ role? Do differences in governance arrangement of a standard setter influence diffusion of standards (Castka and Balzarova 2008b)? Will the countries involved in the standardization use the standards to facilitate social responsibility in their domestic markets?

Practical Implications and Conclusion

Our research offers important practical implications for policy makers, standards setting organizations, and participating stakeholders. The research demonstrates that a broad, non-certifiable standard that is developed through a consensus of heterogenic group of stakeholders (at the platform of an existing standards setting organization) may not necessarily lead to the development of an enforceable instrument. Policy makers should be cautious in supporting such standards because the role of guidance standards is unclear and so is their ability to drive social and environmental change.

The research has also described how stakeholders influence and contribute in standards’ development. This understanding can assist managers, NGO activists, and other stakeholders to evaluate their involvement in the development of social and environmental standards or product labels. Our findings should assist participating stakeholders to carefully assess the agendas of the stakeholder base before committing to the development of a similar standard.