Creating Legitimacy in the ISO/CEN Standard for Sustainable and Traceable Cocoa: An Exploratory Case Study Integrating Normative and Empirical Legitimacy

Abstract

The paper assesses the legitimacy of the ISO/CEN standard for sustainable and traceable cocoa during the standard-setting process and thereby to establish the degree of legitimacy achieved and to explore new sources of legitimacy in the development of sustainability standards for agricultural commodities. The paper examines the normative and empirical legitimacy concerns involved in the development of the ISO/CEN standard for sustainable and traceable cocoa (ISO 34101 series). The findings suggest that while the standard-setting organisation is establishing normative legitimacy, empirical legitimacy is lacking. Absence of empirical legitimacy is a serious concern for a successful and just implementation.

1. Introduction

In the context of governing global commodity chains through voluntary sustainability standards, a centrepiece of the academic debate is the question of how legitimacy is created and claimed by standard-setting initiatives and organisations [1,2,3,4,5,6,7,8,9,10,11,12,13]. Legitimacy is questioned because, in contrast to state authorities, non-governmental organisations have no official mandate to define rules and no enforcement capacities on the compliance with standards [14,15].

Most research examining the legitimacy of standard-setting initiatives is based on empirical case studies. Often cited examples are private sustainability standards like Fairtrade, UTZ or industry-wide initiatives addressing sustainability challenges in commodity sectors, e.g., the Roundtable for Sustainable Palm Oil or the Marine Stewardship Council for Sustainable Fishing [2,12,16,17]. Even though they target different commodity sectors, they share many standard features in the way they claim legitimacy to set and govern the sustainability agenda in the respective sector. For example, they include a broad base of stakeholders and establish standards based on expertise, knowledge and consensus. However, many authors studying multi-stakeholder initiatives have argued that smallholder producers from developing countries are not sufficiently represented in the development of the standard [18]. Another common criticism challenging equal and balanced stakeholder representation in multi-stakeholder initiatives concerns the superior influence of powerful actors in the standard-setting process leading to a reinforcement of existing power inequalities along global value chains [18,19]. This criticism can pose a threat to the legitimacy of developing a voluntary sustainability standard and justifies the need to explore the sources of legitimacy of the ISO/CEN standard for sustainable and traceable cocoa during the standard-setting process.

The standard examined in this paper is unique because it is the first ISO/CEN standard developed for the sustainable production of an internationally traded agricultural commodity. Moreover, it marks a relevant case for understanding legitimacy creation during the development of sustainability standard in a multi-stakeholder setting [20]. Finally, the ISO/CEN standard for sustainable cocoa draws interest for other agricultural commodities which are produced by unsustainable practices. Thus, it can be seen as a forerunner case for other agricultural commodities such as the cotton and rice sector where nearly 100 million farmers are employed and where there is a need for sustainable production to become mainstream (Informant A, CEN secretariat, 2016).

The cocoa sector depends on nearly 14 million cocoa farmers in cocoa-producing countries around the world, growing the crop on smallholder farms [21]. In the last two decades, international NGOs and journalists have revealed that the sourcing of cocoa leaves significant environmental footprints [22], violates international labour rights by making use of child labour and child trafficking, and is [23] also prone to gender issues [24]. Additionally, there is a significant discrepancy in the profit distribution along the global cocoa value chain. The smallholder cocoa farmers, for example, earn less than USD 1.25 per day, which typically equates to receiving 6.6% of the value of the cacao (measured as outlet prices in the West). Castañeda-Ccori et al. document that cocoa farmers make only pocket money [25]. Mars, a leading player in the cocoa industry, has revenue of USD 35 billion [26] (Mars’s revenue includes the sale of non-cacao-based products. The number still indicates the distribution of economic gains in the cacao industry). These and other social, economic and environmental concerns have led to increasing pressures on chocolate companies and manufacturers to engage in more corporate social responsibility (CSR) along their supply chain [27]. Voluntary market-driven initiatives and private sustainability standards have been the predecessors of ISO/CEN trying to prescribe rules in the form of voluntary standards for more sustainable cocoa production and trade.

The empirical data collected in this explorative case study is based primarily on semi-structured interviews with the CEN secretariat, as well as purposefully selected participants and non-participants in the standard-setting process (see Appendix A: list of interview partners). The sample thus encompasses internal participants sampled based on their position and role in creating internal legitimacy, and external stakeholders considered relevant for assessing the external legitimacy. As this paper focuses on the standard-setting process rather than the use-phase of the standard, empirical legitimacy is delimited to internal legitimacy. The internal legitimacy was captured by the views of the participants involved in the development of the standard.

The purpose of the paper is thereby to assess the legitimacy of the ISO/CEN standard for sustainable and traceable cocoa during the standard-setting process, to establish the degree of legitimacy achieved and to explore new sources of legitimacy in the development of sustainability standards for agricultural commodities. The paper’s main contribution is empirical and focusing on the legitimacy creation process as the actors perceive it during the creation process. The findings of the paper should be used by standard-setting organisations to increase awareness about legitimacy impeding factors and enhance legitimacy creation through the outlined strategies in the standard-setting process (see Section 2.1.1). Moreover, by applying these strategies to the empirical case study of this paper, it becomes evident which practical challenges still need to be overcome in future standard-setting processes. For example, we argue that there is a need for redesigning the process of inclusion to ensure that the standard will be considered legitimate and adopted. The paper also makes a theoretical contribution to the field by incorporating hereto ignored academic sources concerned with legitimacy in neighbouring fields. More specifically, the paper develops a model encompassing both normative and empirical legitimacy (and its subcomponents) that can be useful as a structuring device for established literature. In addition, this the paper holds important practical contributions concerning for standard setting organizations about how to avoid marginalizing critical stakeholders whose exclusion can negatively impact the legitimacy of the standard and for critical stakeholder who can insist on that the standard setting organizations introduce means that addresses the exclusionary issues.

The remainder of this paper is structured as follows: The first section provides a brief overview of the conceptual framework based on selected concepts of legitimacy, which are considered relevant in the development of international standards. The second section presents the underlying methodology of this explorative case study, including a presentation of the selected informants. The third section introduces the case study. The fourth section presents the analysis and the fifth section, the overall assessment and conclusion.

2. Voluntary Standards—A Literature Review and A Conceptual Framework for Assessing the Legitimacy

This conceptual section provides a brief overview of the academic debate on the role of legitimacy about voluntary sustainability standards set by non-governmental standard-setting organisations (SSOs). The aim is merely to provide a conceptual framework based on existing theories, which will be used for the subsequent empirical analysis. It builds on Hallström and Boström’s work on transnational multi-stakeholder standardisation and includes additional dimensions needed to analyse the empirical case study [28].

Space prevents us from reviewing the literature on global value and commodity chains, and we focus only on the meaning of the building blocks in the model. The model nevertheless complements the global value and commodity chain research on standards by evoking streams from abject fields that can help to structure the concept of legitimacy and also provide subcategories not used in the existing literature. The model is, therefore, not a critic of the existing literature but developed to create a complementary framework that can be useful in addressing the impact of legitimacy in a more structured manner.

2.1. Defining Legitimacy

Standard-setting organisations strive for legitimacy to enforce the acceptance and implementation of the developed standard [12]. The reason is that “contrary to governments who can use violent force to ensure law compliance, voluntary private governance initiatives rely mostly on legitimacy to ensure compliance” [12], for critical reflections see [29]. The term legitimacy has numerous definitions and encompasses multiple dimensions, depending on the author and context, but there is a common distinction between normative and empirical legitimacy [12,14,30,31,32].

Normative legitimacy refers to the question of whether an “authority possesses legitimacy” and whether its claim of legitimacy is valid, typically justified through, e.g., democratic elections [14]. Normative legitimacy is usually claimed by transnational standard-setting organisations such as ISO and CEN. However, SSOs cannot adhere to the same normative principles inherited in democratisation processes as national authorities. This strengthens the importance of empirical legitimacy which can be framed as the “people’s perceptions of the rightfulness and appropriateness of authority for the acceptance and support for political and social order” [8] (p. 20). Similarly, other scholars describe empirical legitimacy as the societal acceptance of authority and political orders [30,31].

The definitions of empirical legitimacy imply that SSOs can gain legitimacy by the perceived acceptance of a specific audience. A distinction between internal and external legitimacy has been introduced. Bierman and Gupta, for example, explain internal legitimacy as the “acceptance of norms by participants in an institution, for instance, the members of an organisation or supporters of a rule-making mechanism” while external legitimacy refers to the “acceptance of a rule by non-members or non-participants” [32] (p. 1858).

To further break down the notion of normative legitimacy, several authors have differentiated between input, procedural and output legitimacy to assess the justification and credibility of processes and the effectiveness of outcomes of private and public governance [32,33,34]. Hallström and Boström take the approach of distinguishing between three pillars of normative legitimacy: Input legitimacy refers to “balanced representation of stakeholders and the inclusion of expertise”; procedural legitimacy encompasses whether the standard-setting process is transparent and accountable and output legitimacy investigate the stringency of standards and whether they are effective in their goal to increase sustainability and impact the market, people and environment positively [28] (p. 24). Hallström and Boström conceptualisation constitute the foundation (see item 1–7 below) and is supplemented by incorporation of two components being ‘creating a network of partnership’ and ‘initiation of standards’ (item 8–9 below). The two components addressed two crucial gaps identified through the literature review and the empirical analysis (see the methodological section below concerning why the latter is incorporated in the presentation of the framework).

2.1.1. Seven Legitimacy Strategies

Hallström and Boström have suggested seven legitimacy strategies which should be considered when setting up a transnational multi-stakeholder standardisation process (see

Table 1. Legitimacy strategies.

Legitimacy Strategy	Type of Legitimacy
(1) Creating stakeholder categories and balanced representation	Input
(2) Including high-profile actors	Input
(3) Using and referring to expertise	Input
(4) Enabling transparency	Procedural
(5) Facilitating active participation through sufficient language skills and the right vocabulary	Procedural
(6) Practising consensus as a decision- making principle	Procedural
(7) Balancing market impact versus stringent standards and auditing	Output
(8) Initiation of standard	Input
(9) Creating networks of partnerships	Input/Output

Source: Adapted from Hallström and Boström [28] (p. 141).

).

The strategies have been identified by the researchers based on a comparative analysis regarding the legitimacy of three well known international multi-stakeholder initiatives: The Roundtable for Sustainable Palm Oil, the Marine Stewardship Council for Sustainable Fishing and the ISO 26,000 for Corporate Social Sustainability. The strategies will be discussed by reviewing and including other scholars’ views on the sources of legitimacy for transnational standard-setting initiatives to strengthen the relevance of these strategies. In this way, this framework provides sufficient rigour to be used as a reference and analytical framework to structure the analysis of primary data (see Section 5. Analysis).

2.1.2. Literature Review on Legitimacy Strategies

(1)

Creating stakeholder categories and balanced representation

SSOs are expected to include a broad base of representative stakeholders to account for the lack of traditional normative legitimacy gained through, e.g., democratic elections. Hahn and Weidtmann bridge this strategy to the concept of deliberative democracy which can replace the democratic pillar of normative legitimacy when “all parties that are affected by an instrument of policy have the opportunity to participate in its development through deliberation. The centrepiece of legitimacy thus rests in communication processes that aim at building consensus (or at least mutual understanding) among the actors involved” [13] (p. 99). Actors from developing countries are frequently overlooked or play only a marginal role in this process. Regarding this issue, the literature stresses the critical role of local NGOs in representing local actors, e.g., farmers at the international meetings and “voicing” their concerns [18].

(2)

Including high-profile actors

Another dimension referring to input legitimacy is Hallström and Boström’s note suggesting that “Standard setters trying to establish legitimacy for their organisations and processes need to borrow legitimacy from others by inviting stakeholders with strong legitimacy in the eyes of a broad audience” [28] (p. 145). Schouten and Glasbergen consider local governments as high-profile actors in SSOs and argue that the recognition of initiatives by state regimes is one way to enhance their legitimacy [12]. Key participants should be able to influence the decision-making process using their position in their organisation and by their capabilities, knowledge and scope of influence [35].

(3)

Using and referring to expertise

Ponte and Cheyns [36] (p. 463) summarise the literature on normative legitimacy and suggest that standards can be seen as “expert knowledge stored in the form of rules”. Due to the international nature of non-governmental SSOs, the standard-setting process can also be regarded as a “vehicle for the international circulation of values, knowledge and technologies” [37]. During the process of developing the standard, knowledge is not only provided by the participating actors but also generated through their interaction [36]. Without the incorporation of expert knowledge, the input legitimacy might be reduced as it will reflect mainly practitioners’ situated knowledge (while acknowledging that experts’ voice should not be privileged).

(4)

Enabling Transparency

Enabling transparency of governance processes is a crucial strategy to allow for procedural legitimacy [11]. Practically, it may refer to how well stakeholders and the public are informed about on-going activities [28,36]. Disclosing information about the decision-making process to the public is another means to enhance accountability and acceptance of the established standard [11]. Moreover, accountability might ensure that “decisions are based on reason rather than on the sole basis on the asymmetric power resources of the participants” [13] (102). Auld and Gulbrandsen also strengthen the importance of outcome transparency for enhancing legitimacy [11]. This refers to the quantity and quality of disclosed information during and after the implementation of the standard, to ensure that the users can be held accountable for their actions.

(5)

Facilitating active participation through sufficient language skills and the right vocabulary

The standard development process is often constrained by unequal opportunities for participation, e.g., if participants cannot understand the spoken and written language of the meetings and documents. Thus, language barriers have adverse effects on procedural transparency and the balance of stakeholders’ influence on the standard-setting process. It should also be considered that the standard-setting vocabulary is very technical and may cause problems of understanding when stakeholders (e.g., farmers) are not familiar with it [28]. Moreover, several issues which were raised concerning their everyday life and more civic aspects of sustainability were said to be not within the boundaries of standards [37].

(6)

Practising consensus as a decision-making principle

Consensus formation is a crucial component to enhance procedural legitimacy of voluntary sustainability standards [37]. Van de Hove stresses the need to acknowledge that individuals and stakeholder representatives have different and sometimes even contradictory viewpoints, which are not always possible to resolve [38]. He suggests focusing on common interests which can reduce the divergence of standpoints and increase the likelihood of building consensus. An important distinction is further made between consensus and compromise. Van de Hove argues that “participatory approaches always combine some degree of consensus-oriented cooperation and some degree of compromise-oriented negotiation” [38] (13). This may be regarded as a problem because as Hahn and Weidtmann explain “sustained efforts to achieve consensus might additionally lead to a situation in which the individual value for the users is rather low if the achieved consensus builds on too many compromises” [13] (120).

(7)

Balancing market impact versus stringent standards and auditing

Once the standard is published, there are many challenges with measuring its effectiveness in the market. Ambiguities about the distinction and measurement of output and outcome, direct and indirect effects and economic and non-economic factors can exacerbate the assessment of the standard and its output legitimacy. Furthermore, there are different views on whether the aim should be to have low entry requirements for sustainability standards to reach the mainstream market or whether more stringent standards should raise the bar so that only the most socially and environmentally responsibly producers are certified [28].

(8)

Initiation of standards

Inspired by the Habermasian approaches to CSR and communicate rationality, there is a need to complement Hallström and Boström’s seven dimensions with a component covering the dialogical dimensions of the initiation of the standards [19,28]. In the context of standard-setting, this refers to deliberately creating dialogue spaces incorporating multiple stakeholders on equal terms before the initiation of the formal standard development process. These dialogue spaces should ensure an open and transparent dialogue on stakeholders’ legitimate interests and a process ensuring not only that all have the opportunity to voice their concerns but that these concerns are adequately reflected in the subsequent steps of the standard development process. In more practical terms, relevant stakeholder groups should agree about the need for the standard and the related process even before the development, and the actual formulation of the standard begins. This can reduce power imbalances throughout the process and increase the input legitimacy of the standard.

(9)

Creating networks of partnerships

Bitzer, Glasbergen and Leroy argue that the second step after the establishment of partnerships for making the sector more sustainable is the recognition of shared goals and activities leading to the development of linkages between these initiatives and thus forming a “network of partnerships” [39] (p. 359). Bitzer et al. have studied how these partnerships can be linked and suggest two options: either through stakeholders participating in multiple partnerships and creating so-called affiliation networks or through direct institutional linkages between partnerships (i.e., output legitimacy).

The analysis of networks of partnerships is unique for the agro-food sector and is considered fundamental for this study and not addressed by Hallström and Boström and hence included here [28]. In the light of the numerous partnerships and initiatives that emerge as a response to the current sustainability challenges in the cocoa sector and other commodity markets, the creation of networks of partnerships can lead to the avoidance of duplicated efforts and better distribution of tasks and responsibilities. We suggest that the creation of networks of partnerships can be regarded as an additional normative legitimacy strategy.

2.2. Conceptual Framework

In order to analyse the case study of this paper, we have developed a conceptual framework based on but extending the legitimacy concepts suggested by Hallström and Böstrom [28]. Figure 1 below presents the relevant theoretical components and shows how they are related. Important to note is the reciprocal relationship shown by the arrows between normative and empirical legitimacy. Most authors investigating the question of legitimacy in transnational governance, acknowledge the fact that there is a link between normative and empirical legitimacy [13,14].

The analysis of the case will be based on the combination of the two dimensions of legitimacy to assess the overall legitimacy of the ISO/CEN standard-setting process for sustainable and traceable cocoa. The normative elements of the legitimacy of the SSO will be presented along with the strategies that facilitate input, procedural and output legitimacy as suggested by Hallström and Boström [28]. Empirical legitimacy will be examined based on the participating stakeholders’ views. The collected data deepens the understanding of what kind of strategies and elements are essential to establish legitimacy during the development of the standard. Before proceeding to the analysis of input, procedural and output legitimacy concerning the case study, the next section will outline the underlying methodology.

3. Methodology

This section briefly outlines the research design behind the paper. This paper relies on an abductive exploratory case study research strategy for unpacking the process dimensions of developing legitimacy within the context of sustainability standards. According to Eriksson and Kovalainen, the abductive approach is especially suitable for conducting an exploratory research study as it allows for iterations between theory and empirics [40]. The explorative nature of our study further helps to concretise the meaning behind existing normative legitimacy strategies and identify new strategies based on the empirical data. The abductive approach lends itself to specific presentation challenges as the framework is a result of an iterative process, yet the framework is presented before the empirical analysis in the paper. This implies that the framework incorporates elements emerging from the ‘inductive’ parts of the abductive process.

Aligned with the literature in the global value and commodity chains research, we rely on a case study approach. Following Yin, a case study is a preferred strategy when the research aims to explain a complex contemporary research phenomenon with limited control over the boundaries of the study object, as it is the case in this paper [41]. More specifically, we apply a holistic case study approach, see Figure 2. The CEN secretariat gives a holistic view of the organisational structure. The embedded units of analysis are the Danish, Ghanaian and Ivorian committee with their respective participating stakeholders. The part of the study allows for analysing the process related to internal legitimacy. ISO/CEN is embedded in the external context, and hence the research includes interviews with experts not participating in the development of the ISO/CEN standard to assess the external processes about the creation of legitimacy. All informants are thus purposefully selected.

We regard the stakeholders as independent entities grouped by their national committees and who are all participating in the standard-setting process. The eight embedded cases document the participants’ views on the process, the internal empirical legitimacy. The following reasons can explain the focus on the Danish, Ghanaian and Ivorian committee: The Danish committee has been chosen since Dansk Standard, the Danish national standardisation organisation has initiated the development of the standard, a senior consultant at Dansk Standard is the leader of the CEN secretariat for the standard at focus (he is one of the informants). Toms—a Danish chocolate producer—played the driving role in the initial phase. Following the ISO requirements, each standard can be developed by a standardisation body from a developed and developing country under the so-called “twinning arrangement” [42]. The Ghana Standards Authority (GSA) has entered such an agreement and constitutes a core member of the ISO committee for the standard on sustainable and traceable cocoa. Five Ghanaian participants have been contacted for an interview. In the end, only two out of five participants, the representatives of the International Cocoa Farmers’ Organization (ICCFO) and the World Cocoa Foundation (WCF) accepted to be interviewed. This attests to the challenges of undertaking research concerned with standard-setting processes.

The single embedded case for Ivory Coast has been included because it provides some valuable data from a representative of the largest cocoa economy in the world and gives additional insight into the farmers’ views in West Africa. The Operational Manager of ICCFO has suggested the interview partner from Ivory Coast.

Data Collection and Data Analysis

An exploratory study as this cannot rely on established data sets and/or survey data, and the complexity of the research also suggests a need for applying an open-ended research methodology allowing for iterations between theory and empirics (i.e., an abductive approach). Hence, we collected qualitative data via semi-structured interviews with a total of twelve informants [43]. The structure of the interview guide corresponds to the triple composition in input, procedural and output legitimacy, which has been explained in more detail in the literature review. Different emphases have been chosen depending on the interviewee. The interviews were transcribed and coded. Coding refers to the systematic classification of themes in empirical data [44]. The qualitative data analysis software Nvivo has been chosen as a tool for coding. In line with the abductive research approach, we used “sensitising concepts” in the form of input, procedural and output legitimacy and the corresponding seven suggested legitimacy strategies by Hallström and Boström to structure the primary qualitative data [28,40]. In Nvivo coding is stored in so-called “nodes” which have been created for the pre-selected concepts. In the next step, each transcribed interview has been coded according to the established nodes, but also regarding possible subcategories, so-called “child nodes” in Nvivo, see Figure 3 for the presentation of the tree of codes.

This approach enabled to concretise what the seven theory deduced legitimacy strategies meant in the specific ISO/CEN case and what interviewees regarded as relevant. The literature review on the suggested legitimacy strategies has further helped to make sense of the data. Moreover, two new nodes have been created for the newly identified strategies, one based on the literature review and data (i.e., creating networks of partnerships) and one based on data only (i.e., initiation of standard). Each parent node consists of a hierarchy of child nodes which have been identified based on multiple rounds of coding. Figure 3 illustrates the tree of codes; the figure illustrates procedural legitimacy only, but a similar procedure has been used for each parent node.

The primary qualitative data are complemented with participant observations and secondary data. As argued by Robson, participant observation is especially useful when the purpose is to get an idea of the on-going debate in the field of research and as a means for established a network used for snowball sampling of informants [45]. One of the authors participated in Copenhagen’s Chocolate festival and gained access to industry-specific knowledge and established contacts which were later transformed into interview appointments; the participant observation was especially useful for identifying informants related to external legitimacy. Snowball sampling for internal legitimacy was conducted with the assistance of the CEN secretariat. Various secondary data sources have been used to complement the primary data. Easterby-Smith argues that secondary data can especially be useful to get a historical perspective on the chosen research topic [46]. For this purpose, numerous reports about the evolution, characteristics and sustainability issues of the cocoa sector in West Africa have been reviewed including but not limited to reports from financial institutions (e.g., World Bank) and the United Nations (e.g., UNCTAD), international consultancies (e.g., KPMG), non-governmental organisations (e.g., Solidaridad, VOICE Network), research institutes (e.g., Südwind institute), the standard and certification industry (e.g., Fairtrade, UTZ) and cocoa initiatives (e.g., WCF, ICCO). When conflicting data appeared we assessed the validity and reliability of the data on a case-by-case approach and incorporated the findings from secondary sources, which had the highest quality.

Moreover, annual reports, sustainability reports, power-point presentations and further information gathered from websites of private industry players such as Mars, Toms and Barry Callebaut have been utilised. To stay informed about current developments in the cocoa-chocolate industry, the online newspaper “Confectionery News” has been monitored and read. The leader of the CEN secretariat also shared some internal documents, presentations and minutes from the meetings on ISO-standards.

The main concern in connection to reliability was “participant bias” and refers to the possibility that the respondents are hindered to say what they think [43]. Data bias could occur, considering that the requirements for the standard were still discussed at the time of the research. To reduce this threat, a confidentiality agreement was signed with the leader of the CEN secretariat and offered to other participants as well. Triangulation between different data sources was also used, including “ISO Policy on Communication of Committee Work” [47]. Observation bias was also reduced by dialogue between the authors and by applying stringency in the coding.

A minor issue concerning the validity of the study is the fact that Toms’ representative who was willing to meet in the initial preparation phase of the research dropped out of the Danish committee and left her position as sustainability manager at Toms. Saunders et al. call this problem the “mortality” threat to validity [43] (157). The consequence is that the data gathered from this initial interview is less focused on the ISO/CEN standard than the other interviews. The opposite case occurred with the representative from Mars, who was recently hired and joined the ISO/CEN process only at the end of 2015. Thus, her experience with the standard was significantly shorter compared to the other participants. Despite these shortcomings, the validity concerns are minor.

4. Case Introduction—Sustainability Challenges in the Cocoa-Chocolate Industry

This section firstly presents the stylised facts on the needs of the chocolate industry, motivating them to initiate the process of establishing a cocoa standard. The section provides the backbone for the subsequent analysis.

4.1. The Commercial Need

According to recent statistics, 73% of world cocoa is produced in Africa thereof 70% originates from Ivory Coast, Ghana, Cameroon and Nigeria and 58% by Ivory Coast and Ghana alone [21]. The three largest global confectionery companies by net sales that manufacture chocolate-based products are Mars Inc (McLean, Virginia, USA), Ferrero Group (Alba, Piedmont, Italy/) and Mondelez International (Chicago, Illinois, USA). At the same time, there are hundreds of other international confectionary companies holding a smaller share in the global chocolate market [21]. A shared commercial concern by the industry actors is that in the next decades, the expected supply of cocoa will not be able to meet global cocoa demand [48]. The concerns stem from the current social, economic and environmental challenges that cocoa farmers face, especially in the two most crucial cocoa-producing countries in West Africa, Ivory Coast and Ghana which account for more than half of world cocoa production [21]. In these countries, cocoa is grown and harvested by more than 6 million smallholder farmers on 2–5 hectares of land [49]. On their limited land, their productivity is constrained by ageing cocoa trees producing low yields and affected by pests and diseases.

Moreover, the use of harmful pesticides and the lack of modern agricultural techniques constitute further obstacles to sustainable cocoa production [49].

Furthermore, it has been shown that West African smallholder cocoa farmers are hardly able to earn income above the global poverty line when their livelihood solely depends on the selling of cocoa. Not surprisingly, this bears many related problems: the use of child labour, child trafficking and poor working conditions have been reported [50,51]. Due to the low income that farmers earn from harvesting cocoa, they exit the industry and start growing competitive crops like rubber or palm oil or move from agriculture to other sectors [52]. This development imposes a risk to chocolate companies’ supply of cocoa in connection to meet the expected growing demand in the confectionery industry in the near future.

4.2. The Technical Need

Recognising challenges in the commodity sectors, standards and certification industry for sustainability has evolved in the past two decades. Voluntary sustainability standards (VSSs) can be defined as “standards specifying requirements that producers, traders, manufacturers, retailers or service providers may be asked to meet, relating to a wide range of sustainability metrics” [53]. Socio-economic sustainability (e.g., living wages) is assumed to increase the incentives for the farmers to remain within the industry, and environmental sustainability is assumed to ensure the long-term socio-economic sustainability of the agro-production part of the global cocoa value chain. This is needed to ensure the supply of cocoa beans in demand by the lead companies to ensure their ability to meet market demands at affordable production costs. Without efficient sustainability standards, the industry fears a change in the bargaining position between buyers and suppliers and thus higher prices for the end cocoa products. This, in turn, translates into higher prices (or lower profits) and possible substitution of chocolate by other types of candy or cake.

There is already a booming of standards. The most established and prevalent private standard initiatives for sustainability in the cocoa sector are Fairtrade International (hereinafter: Fairtrade), Rainforest Alliance/Sustainable Agricultural Network (hereinafter: Rainforest Alliance) and UTZ Certified (hereinafter: UTZ). However, the multitude of standards and certifications leads to some practical challenges such as confusion over multiple labels faced by consumers and conflicting demands placed on the cocoa producers, hence not solving the socio-economic sustainability challenge among cocoa producers.

The first industry-wide action plan to make the cocoa sector more sustainable was set in the Harkin–Engel Protocol signed by private and public representatives of the cocoa and chocolate industry in 2001 [54]. This protocol laid the foundation to establish the International Cocoa Initiative (ICI), which works with the industry, civil society organisations and national governments in cocoa-producing countries on the implementation of the protocol. Policies have also been formulated on company level. For example, Mars and other major companies like Ferrero and Hershey have officially committed to source 100 per cent sustainable certified cocoa by 2020 [48]. Moreover, major chocolate companies are jointly members of the World Cocoa Foundation (WCF). It is an international membership organisation whose members have recently agreed on a voluntary industry-wide strategy called “Cocoa Action”. The underlying goal is to coordinate and manage sustainability efforts in the cocoa sector (WCF, n.d.). However, according to Bitzer et al., WCF does not engage in rule setting [40]. It neither establishes a sector-wide minimum standard nor selects an already existing standard as a ‘golden standard’. Instead, WCF accepts the coexistence of multiple standards. Thus, there is a perceived need to establish a shared standard to overcome the socio-economic and environmental barriers preventing the industry from being able to sustain its industry in the future.

4.3. The Role of ISO/CEN

The ISO/CEN standard for sustainable and traceable cocoa intends to fill this gap by developing a “universal” standard which has been initially requested by the Danish chocolate industry. The goal is to develop an international standard for sustainable cocoa production, making sustainable cocoa mainstream [55]. This shall be achieved by specifying requirements for a management system for sustainable production of cocoa. In general, a management system in an organisation is implemented to make the use of resources more efficient, improve risk management and increase the satisfaction of customers by meeting their product or service expectations. For the ISO/CEN standard for sustainable cocoa, an organisation can also be understood as a farmer cooperative. These farmer associations also need to manage their agricultural resources, cope with the above mentioned social, economic and environmental risks and satisfy their customers, the cocoa buyers and traders, who may demand sustainably grown cocoa according to standards’ requirements [50]. However, the concept and understanding of a management system comes from the industry and is promoted mainly by ISO standards that help larger organisations structure and manage their operations. The standard refers to the formation process, and the subsequent analysis examines the extent to which the industry-driven process has reached the legitimacy needed for developing a functioning standard in the interests of all stakeholders acting in the cocoa sector.

5. Analysis and Discussion

To address the concern if the standard is conceived as normatively legitimate (left side of the model, Figure 1), normative legitimacy is assessed based on integrating input, procedural and output legitimacy with the Hallström and Boström’s seven legitimacy enhancing strategies [28]. To address if the standard is conceived as empirically legitimate, the empirical internal and external legitimacy is addressed subsequently. The section can only present selected findings, especially for normative legitimacy due to a large number of subcomponents. We refer to

Table 2. Normative legitimacy.

Type of Legitimacy	Strategy	ISO/CEN Approach: Legitimacy Enhancing	ISO/CEN Approach: Legitimacy Impeding
Input	Initiating a standard	Standard development as a response to a need in the market, consultation through CEN’s national standards bodies Subsequent alignment with ISO based on Vienna Agreement	Consultation only through national standards bodies, initially only in Europe SSEAL Alliance and stakeholders outside Europe have not been consulted in the initiation phase of the standard
Input	Creating stakeholder categories and balanced representation	Multi-stakeholder approach: participants from both cocoa consuming and producing countries representing various stakeholder groups “Twinning” arrangement between developed and developing country standardisation body (Netherlands and Ghana) Various channels to get involved: national standardisation bodies, trade/European associations, directly via ISO/CEN	Poor communication between national standard bodies and possible participants (e.g., in Ghana) Some stakeholder groups more influential than others (e.g., industry representatives) Insufficient farmer representation on the international level, farmer organisation only liaison partner with limited voting rights, joined the standard-setting process at a late stage
Input	Including high-profile actors	High-profile representatives from notable chocolate companies (e.g., Mars, Facer, Lindt) participate Producing country, government representatives are involved to some extent (Conseil du Café-Cacao, Cocobod)	Not all major chocolate companies participate (e.g., Nestlé) More engagement between the secretariat and critical decision-makers in cocoa-producing countries is needed Availability and financial capacity of participants not always assured
Input	Using and referring to expertise	National mirror committees, technical committees and working groups ensure expertise on different levels	The process can be influenced by specific kind of knowledge, preparation, argumentation Possible conflicts in participants’ role as experts and representatives of their organisation/country
Procedural	Enabling transparency	Procedural transparency Transparency guidelines are based on WTO principles for standard development ISO guidelines on transparency are available Output transparency Certification scheme document is being developed to ensure accountability of standard compliance	Procedural transparency Some information on participants and content is not publicly available during the standard-setting process Output transparency Complex audit and accreditation system, scheme ownership is still undefined
Procedural	Facilitating effective participation through sufficient language skills and the right vocabulary	Documents are available in four different languages, national meetings are held in the national language, international meetings in English A guidance document is being developed	Not all participants in international meetings understand and speak English Difficulties with understanding standard terminology Guidance document not suitable for the individual farmer
Procedural	Practising consensus as a decision-making principle	Consensus-based approach Good atmosphere among members to facilitate agreements Common overall goal	No clear definition of sustainability Many other controversial topics where consensus is difficult to achieve Compromise rather than consensus
Output	Balancing market impact	Industry-wide mainstream standard Low-threshold/high-bar approach	Minimum standard rather than sustainability standard Meaning of ISO/CEN standard is difficult to communicate to consumers
Output	Creating a network of partnerships	Liaison partnerships are established and enable coordination of sustainability agendas ISO/CEN standard-setting process becomes a platform for collaboration It facilitates the overall effectiveness of joined and individual actions through exchange of information and resources	Unclear distinction between requirements of ISO/CEN standard and existing standards The business model of existing standards impedes one harmonised standard for the whole sector

for additional details and documentation concerning normative legitimacy.

5.1. Normative Legitimacy

The standard-setting organisations ISO/CEN have developed a solid grid of procedures to enhance its normative legitimacy which is anchored in the four defined principles of standard development: respond to a need in the market, based on global expert opinion, developed through a multi-stakeholder process and based on consensus (ISO website). On the input level, the two organisations unify 163 national standard bodies and create direct and indirect channels to enable stakeholder participation from different stakeholder groups that are relevant for the respective industry for which the standard is developed. Additionally, liaison partnerships are built to collaborate with other partnerships and initiatives. Moreover, national mirror committees, international technical committees and working groups are established to create platforms for discussion, negotiation and the generation of new knowledge based on expertise. On the procedural level, various policies and documents provide guidelines to ensure transparency and to reach a consensus-based result. On the output level, the standards are made available through the national standard bodies and further guidance documents are developed to facilitate the use, implementation and monitoring of the standard.

These efforts that ISO/CEN puts into these three levels are aiming at ensuring a high level of normative legitimacy, and as Hallström and Boström note, through the implementation of these and other procedures SSO’s have “grown to become authoritative”. However, due to the voluntary nature of standards, “the authority tends to be strong only as long as stakeholders perceive the power of these arrangements as legitimate” [28] (p.26). This has been captured by the reciprocal relationship between normative and empirical legitimacy in the conceptual framework. This implies that without complementary empirical legitimacy, the standard is not considered legitimate (see empirical legitimacy below).

What was perceived as limiting the normative legitimacy despite the considerable effort set in action by ISO/CEN? Concerning input legitimacy, concerns were raised about the limited inclusion of non-European participants, power and influence inequality, and limited farmer representation. The most critical opinions in the Danish committee have been voiced by the NGOs, the Danish Consumer Council and NOAH-Friends of Earth Denmark:

“There is too much industry. It’s not democratic in my view. Sometimes I think the NGO’s are just there to make the standard legitimate.”

(Informant D, Danish Consumer Council, 2016)

“It was the first time that we have decided to join forces with the Danish standard (…), we wanted to see if we could make any changes. I think the conclusion is that our opinions are too far from each other, so we cannot influence the system; our forces are used better outside than inside.”

(Informant F, NOAH, 2016)

For procedural legitimacy, problems stem from information asymmetry, divergent definitions of sustainability and the risk of agreeing on the lowest common denominator regarding what sustainability entails. For example, the Danish environmental NGO eventually decided to exit out of the process because it could not accept the standards’ underlying definition of sustainability which does not anticipate organic farming and sets environmental requirements at what they refer to as ‘minimal level’.

This, in turn, imposes a threat on output legitimacy, which requires sufficiently high requirements to enforce sustainable cocoa production. Moreover, output legitimacy is further constrained by, e.g., unclear communication to consumers about the content of the standard as there is no brand associated with the standard (unlike Fairtrade, UTZ, etc.) (see Table 2 for more details).

The newly added strategies “Initiating a standard” and “Creating networks of partnerships” emerged from the analysis of our primary data and additional literature review. The first has been inspired by ISEAL’s critique on the insufficient consultation with relevant stakeholders about the need for the standard. A multi-stakeholder approach for identifying a need for a standard has, to our knowledge, remained unaddressed. In this case, none of the cocoa-producing country stakeholders has been consulted about the actual need for an international standard. Due to that, informant A (Dansk Standard) has reported that participants from cocoa-producing countries “feared to some extent that it’s a European process”. Thus, we suggest that normative legitimacy can be strengthened if relevant stakeholders are included in the initial phase of deciding upon the need and purpose of the standard.

This links to the second suggested legitimacy strategy of creating a network of partnerships. These networks can decrease the competitive threat perceived by existing standard-setters. Although ISEAL and its members felt overlooked in the initiation phase, the partnership and the inclusion of its members in the standard-setting process has led to accepting the upcoming standard. According to the informant B (Fairtrade), his organisation can align to the new standard and at the same time proliferate with its brand value.

“I would compare the CEN/ISO with the bachelor’s degree and Fairtrade and UTZ and Rainforest to a master’s degree. We are specialised more in some areas and especially when it comes to the brand value, CEN/ISO don’t have a brand value as such and we will, Fairtrade and the other private standards, have to align to the CEN/ISO to some stage. Turning our back to it is not possible.”

(Informant B)

While creating these networks of partnerships can enhance the normative legitimacy of the new standard, it can also be seen as a compromise solution. The ISO/CEN standard has to give up on its goal to be the only standard in the cocoa sector which would make a choice easier for producers and consumers. Moreover, it is questionable whether stakeholders such as cocoa farmers, chocolate companies or consumers will understand the difference between the ISO/CEN standard and the existing standards. If they do and they follow the idea of informant B to regard the new standard as a minimum standard and the existing standards as “more sustainable” standards, the debate about developing an international minimum standard and calling it a “sustainable” standard becomes reignited. Another more positive way of looking at it is that the establishment of a minimum standard to which existing standards and certification schemes can align ensures that minimum requirements are fulfilled, and more smallholder farmers can sell sustainably grown cocoa. At the same time, competition among existing standards will foster innovation for higher and better sustainability standards.

5.2. Empirical Legitimacy Assessment

We shall now turn to empirical legitimacy and unravel the importance of internal and external empirical legitimacy. Without empirical legitimacy, the standard will not be considered legitimate despite being able to claim normative legitimacy.

5.2.1. Internal Empirical Legitimacy

The interviews with the participants from the Danish committee revealed that there is a strong support of the standard from the industry representatives: the chocolate companies were behind the initiation of the standard-setting process, and their ‘voice’ was incorporated early. As argued earlier, Fairtrade represents a particular case because the informants from Fairtrade used to work for Toms and was part of the initiating group of the standard before he joined Fairtrade Denmark. Thus, the informant’s positive attitude towards the standard could be influenced by his former position at Toms. The most critical opinions in the Danish committee have been voiced the Danish Consumer Council and NOAH-Friends of Earth Denmark. Both interviewees have raised concerns about their limited influence on the standard, especially in comparison to the industry’s influence. NOAH even decided to drop out of the process because they could not agree with the standards underlying definition of sustainability which does not anticipate organic farming and sets environmental requirements at what he calls minimal level.

The West African interview partners have granted more acceptance of the process. WCF regards the process as very inclusive and considers the liaison partnership between the World Cocoa Foundation and ISO/CEN essential to align efforts in making the cocoa sector more sustainable. Nevertheless, they also stressed the need to include key decision-makers from the West African governments in order to support the implementation of the standard in cocoa-producing countries. Additionally, WCF highlighted that it takes more engagement by the SSO to communicate the need and usefulness of the standard before it is regarded as legitimate by the local stakeholders.

The two representatives of the farmer organisation ICCFO argue that their participation is necessary to ensure more implementation of the upcoming standard by the smallholder farmers and that they constitute a crucial link between the international standard and the cocoa farmers. However, joining the process after most requirements are written, decreases their influence on the standard and their acceptance of it. They are thus more sceptical about legitimacy.

5.2.2. External Empirical Legitimacy

Finally, the non-participating interview partners have provided their outlook on sustainability in the cocoa sector and the potential use of the upcoming standard. However, since they are not engaged in the standard-setting process, their perception of the standard’s legitimacy is too early to be firmly evaluated. Their centrality in the filed nevertheless lends credibility and legitimacy to their claims. As argued earlier, the SSO should consider involving new stakeholder groups, e.g., journalists, investors and consultants since they provide a more holistic and alternative view on the cocoa industry and can broaden the understanding of what should be considered in a voluntary “sustainability” standard. Furthermore, other non-participating stakeholders, e.g., smaller chocolate companies or consumers, might be easily influenced in their perception about the standard by the media. The interviews with the external stakeholders illustrated two points shedding light on how the legitimacy is limited. An informant representing an investment bureau (active in the cacao industry in developing countries) explained that the ‘their’ part of the industry was not included and that they did not have much information of the process. This suggests that they were not in a position to consider the standard as empirically legitimate. Another actor representing mass media pointed out that the industry failed to provide support for functional upgrading. The actors explained that the standards focus domain was insufficient. The view of the media voice is critical because they are the actors behind the public media representation of the standard. The abovementioned journalist has, for example, been producing internationally recognised productions about cocoa and will, therefore, be consulted by other media concerning their productions. While one can debate if the standards should cover the issues pointed to by the media representative, it is nevertheless critical that he hosts this view because he holds the definitional power on shaping global media coverage and thereby consumers’ and other stakeholders’ perception of the legitimacy of the standard. The empirical legitimacy of the ISO/CEN cocoa standards could be considered limited.

6. Conclusions

Thus, to sum up, the standard development process is perceived to possess a certain internal legitimacy but also lacks to incorporate sufficiently the voice of selected stakeholders (e.g., NOAH) and suffer from lack of inclusion of, e.g., small farmer organisations in West Africa. There is thus a need for redesigning the process of inclusion to ensure that the standard will be considered legitimate and adopted as well as capturing the required incentive structures. Thus far, the standard-setting process has been initiated and primarily driven by chocolate companies like Mars. It can be regarded as a tool to achieve their cocoa sustainability policies and to control the associated risks of insufficient cocoa resources. The technical requirements specifying how farmers should implement good agricultural practices do not capture the full complexity of sustainability challenges or economic and political relationships encountered in the global cocoa sector. Thus, there is a need for more elaboration on the definition, opportunities and boundaries of sustainability standards as opposed to industrial ISO standards.

Additionally, the analysis and discussion suggest that empirical legitimacy is limited in numerous ways. We highlight that the critical stakeholders represented by investment organisations and media hold doubt about the relevance of the standard and the domains it encompasses. Although the normative standards are reached, this suggests that it is challenging to implement the standard because of the limited external legitimacy.

Our empirical findings and model underpinning the empirical analysis complements existing research on the global value and commodity chains by paying more attention to normative and empirical legitimacy. The paper thereby suggests how this stream of literature can benefit from incorporating, testing and nuancing our framework. As mentioned under limitations below, global value and commodity chains scholars need to elaborate on how to assess empirical legitimacy.

Researching standard-setting processes in the actual development phase is demanding and typically comes with limitations in the numbers of informants. This is also the case for this paper and a limitation worth mentioning. However, we have tried to compensate this by including the most critical actors in the process—both critical and supportive actors. The assessment of external legitimacy turned out to be methodologically challenging; the most significant challenges was the limited knowledge about the process and thereby the difficulties related to finding relevant informants. Additionally, we had to infer conclusions based on that some informants had highly limited knowledge about the process.

The study alluded to research off-springs in numerous ways. The standard has only just been released. Follow-up research should especially unpack how the publishing of the standard affects the general acceptance of the standard by companies and farmers and pay special attention to post-publishing dynamics concerned with creating empirical legitimacy. The study has mainly addressed the structural components of legitimacy and selected informant’s perception of the legitimacy of the standard, yet future research needs also to incorporate a behavioural component. There is, e.g., an entire behavioural literature stemming back to Hirschman (1970) on voice, loyalty and exit [56]. This literature addresses how numerous barriers to participation in problem-solving activities ranging from psychological safety over implicit value theories (i.e., taken-for granted-values) to hierarchical values influence the participant’s inclination to participate actively (Latif and Vang, forthcoming). Incorporating these dimensions will provide an even stronger foundation for assessing conditions for real participation.

The findings also have critical implications for how standard setting organizations can create inclusion strategies addressing the factors that reduce the legitimacy of the standards, e.g., exit from the standard creating process. This entails, e.g., ensuring that the stakeholders are adequately included and not used as fig leaves and that the process is organized to ensure that stakeholders without technical expertise can participate on the same level as actors with technical expertise. Stakeholders can also use the findings to device requests to the standard setting organization demanding that the address the factors identified as preconditions for joining in the standard setting activities. This will not solve all challenges but will increase the voice of currently marginalized stakeholders.