Introduction

The social and environmental effects of multinational corporations (MNCs) and their supply chains garner increased attention, with business practices increasingly judged against the United Nations’ Sustainable Development Goals (SDGs) (Burritt et al., 2020; Kelling et al., 2021). In the sphere of human and labour rights, many Western countries recently enacted legislation to combat modern slavery, which include supply chain elements (Australian Government, 2018; UK Parliament, 2015; Islam & Van Staden, 2022). For example, in the UK commercial organisations that fall within the remit of the 2015 Modern Slavery Act must prepare an annual slavery and human trafficking statement, detailing the steps the organisation took during the financial year to ensure that slavery and human trafficking is not taking place in any of its supply chains, and in any part of its own business (UK Parliament, 2015). The statement may include information on the parts of its business and supply chains where there is a risk of slavery and human trafficking taking place; the steps the company has taken to assess and manage that risk; and its effectiveness in ensuring that slavery and human trafficking is not taking place in supply chains (UK Parliament, 2015). In the EU, businesses face mandatory supply chain due diligence requirements, necessitating greater attention to their supply chain partners’ business practices (European Parliament, 2023). Specifically, the EU Directive on Corporate Sustainability Due Diligence envisages that companies falling within its scope “should consider establishing through contractual provisions with a partner with whom they have a business relationship that it will ensure compliance with the code of conduct and, as necessary, a prevention action plan” (European Parliament, 2023, p.36). This includes measures relating to occupational health and safety, working hours and workload, labour exploitation including child labour, fair trade, production of waste, and the sustainable use of natural resources (European Commission, 2022). In addition to these legal requirements, NGOs and the media take increasing interest in the practices of MNC’s supply chains with growing consumer pressures on managers to ensure that their whole supply chain acts in a responsible and sustainable manner (Vestergaard & Uldam, 2022). Those that fail to respond to growing societal concerns risk consumer, employee, media, and investor backlashes (Tian et al., 2021; Vestergaard & Uldam, 2022), weakening organizational legitimacy (Bitektine & Haack, 2015).

In response to these pressures, MNCs increasingly implement social evaluations assessing working practices and human rights, linked to standards which their suppliers must follow (Bartley & Egels-Zanden, 2015; Locke et al., 2009, 2013; Soundararajan et al., 2021). Exchange relationships, and hence the legitimacy of suppliers, depend on an evaluation of the latter’s social practices. Although codes of conduct and private industry standards have a long history, the former traditionally governed conflict of interests, bribery, and corruption (Bondy et al., 2008). Industry standards, on the other hand, addressed technical specifications, safety, and hazard procedures (Baines, 2010). Notwithstanding some notable exceptions like Fair Trade, the expansion of private industry/third-party standards to cover human and labour rights within supply chains is relatively new (Wettstein et al., 2019).

A growing body of literature considers workers’ rights within international supply chains (Anner, 2020; Bartley & Egels-Zanden, 2015; Vaughan-Whitehead & Caro, 2017). Much of this research takes a Global Value Chains (GVCs) approach, considering the degree to which the engagement of suppliers from emerging economies in international supply chains results in social upgrading. Social upgrading refers to ‘improvements in the rights and entitlements of workers as social actors by enhancing the quality of their employment’ (Rossi, 2019, p.273). This literature identifies a set of metrics for measuring social upgrading and explores the relationship between economic (improvements in value added) and social upgrading (Barrientos et al., 2011; Gereffi & Lee, 2016). It finds that economic upgrading in GVCs can be conducive to, but insufficient for, social upgrading to transpire (Henry & Rivera, 2019; Rossi, 2019). Specifically, engagement with global supply chains sometimes results in deteriorating working conditions due to imbalances in power and the intensifying cost pressures exerted by focal organisations (Anner, 2020; Bartley, 2018) as well as corruption and cultural factors (Lin-Hi & Blumberg, 2017). The potential decoupling of economic and social upgrading raises concerns regarding the outcomes of lead firms’ labour-related initiatives in their supply chains (Locke et al., 2009, 2013). However, the primary concern of the social upgrading literature is with the outcomes of GVCs rather than MNCs’ resource dependencies and legitimization strategies for social evaluations per se. Consequently, theories relating to the external control of organizations (Pfeffer & Salancik, 2003), and legitimacy (Bitektine & Haack, 2015; Suchman, 1995; Suddaby et al., 2017) can help enrich our understanding of the enactment and implementation of MNCs’ social evaluations. Drawing on resource dependence and legitimacy theories, this paper conceptualizes MNCs’ implementation of social evaluations as a means to reduce environmental contingencies (McWilliam et al., 2020) and maintain organizational legitimacy (Bitektine & Haack, 2015), protecting themselves from negative legitimacy spill-overs (Haack et al., 2014), while meeting Corporate Social Responsibility (CSR) objectives, and retaining autonomy for independent action.

This paper addresses the research question: how do MNCs legitimise and implement social evaluations in their supply chains? To address this question the paper draws on Resource Dependence Theory (RDT) as well as strategic approaches to organizational legitimacy (Suchman, 1995; Suddaby et al., 2017). Both RDT and legitimacy theories analyse the relationships between organizations and their environments, where legitimacy and resource dependencies are constraints on organizational behaviour (Bitektine & Haack, 2015; Dowling & Pfeffer, 1975; Pfeffer & Salancik, 2003). Both theories seek to identify strategies organizations can employ to mitigate these constraints (Bitektine & Haack, 2015; Pfeffer & Salancik, 2003). RDT (Pfeffer & Salancik, 2003) argues that organizations rely on external actors, for financial and physical resources (resource dependencies), which they seek to co-opt, to obtain and preserve autonomy and maintain their ability to pursue organizational interests. Augmenting RDT, resources critically include Corporate Reputation (CR), which rather than being an owned asset or process, depends on the perceptions of stakeholders (von Berlepsch et al., 2022). CR is, thus, not under the direct control of an MNC and represents an environmental dependency. Further complicating this dependency is the potential for the actions of suppliers to spill over and negatively affect the CR of a MNC (Haack et al., 2014).

In their critical examination of RDT, Hillman et al. (2009) identify three areas for future research. Firstly, to go beyond the existing strategies identified by Pfeffer and Salancik (2003) and explore ‘other actions firms can take to manage/reduce environmental dependencies’ such as outsourcing, as ‘research applying RDT to supply chain management is minimal’ (p.1419). Secondly, to refine RDT, recognizing the multiplexity of resource dependencies and how the ‘rapid pace of globalization in the past three decades also raises new opportunities for organizational forms designed to reduce interdependences across different institutional settings’ (p.1419). Finally, Hillman et al., (2009, p.1420) echo the call of Pfeffer and Salancik (2003) ‘for more research considering the boundary conditions of RDT.’ This agenda informs the current paper, which focuses on a revelatory, single case study (Yin, 2018), that of the introduction of social evaluations for fresh fruit and vegetables suppliers by a large, multinational grocery retailer. Fresh fruit and vegetable supply chains represent an exemplary case as the external resources relevant to retailers are reframed from merely financial and physical (e.g., safe, fresh produce at good prices) to assurances regarding workers’ rights in a highly labour-intensive production process. Assurance of workers’ rights is challenging in fresh fruit and vegetable supply chains as employment is geographically diffused, often seasonal, with short peaks of physical work, often relying on temporary immigrants, some of whom may be undocumented (Augère-Granier, 2021). Ensuring compliance with due diligence legislation is, thus, problematic.

This paper contributes to the social evaluation literature by conceptualizing how the evaluation of suppliers’ social practices by third parties emerged as a novel strategy employed by focal organizations to reduce their environmental dependencies while preserving autonomy. In so doing, we extend RDT by conceptualizing external resources as not merely financial and physical in nature but also perception (Suddaby et al., 2017) and credence based (Darby & Karni, 1973). In so doing, we refine and extend RDT’s in house assumptions regarding the strategies employed by MNCs to reduce their environmental dependencies (Alvesson & Sandberg, 2011). Secondly, we contribute to the organizational legitimacy literature through an understanding of how third-party standards bodies, as judgment validation institutions (Bitektine & Haack, 2015) independent of the focal institution, are the preferred means for implementing social evaluations. Specifically, such third-party social evaluations offer a means for the focal organization to reconcile both moral and pragmatic concerns, in a manner which allows it to mitigate resource dependencies without ceding control over enforcement and enabling actions.

The next section of the paper draws on RDT and legitimacy theory as foundations for understanding how MNCs legitimise and implement social evaluations in their supply chains, before turning to the methodology. The findings section analyses the implementation of social evaluations for suppliers in the case study MNC. The discussion evaluates the relationships and tensions between different forms of legitimacy, with the focal organisation designing processes that seek to reconcile moral and pragmatic concerns. The paper concludes with a discussion of limitations and suggestions for future research.

Resource Dependencies and MNCs’ Social Evaluations of Their Supply Chains

RDT conceptualizes how the non-dependable environments in which they operate, constrain and affect the ability of organizations ‘to create acceptable outcomes and actions’ (Pfeffer & Salancik, 2003, p.11). Accordingly, a function of management is to enact or create a more favourable organizational environment (Hillman et al., 2009; Pfeffer & Salancik, 2003). Yet, the ability of managers to control and manipulate their environment depends on the nature of the organization’s interdependencies with external actors and the degree to which they are symmetric. Asymmetric interdependencies explain variations in organizational power, with the former arising from differences in the availability of resources relative to their demand (Casciaro & Piskorski, 2005). An organization is relatively weak and lacks autonomy when, for example, it depends on suppliers which provide critical and important inputs, with a lack of alternative providers (Pfeffer & Salancik, 2003).

According to RDT, the solution to problems of interdependence and uncertainty is to increase coordination, and hence, mutual control over external actors’ activities (Casciaro & Piskorski, 2005; Hillman et al., 2009). Pfeffer and Salancik (2003) identify five main strategies for achieving this: vertical and horizontal integration (extending organizational control over exchanges vital to its operation, increasing power in exchange relationships, and reducing uncertainty generated from competition), association with interdependent others with the partial pooling of assets (e.g., joint ventures, cartels), interlocking Boards of Directors to stimulate tacit co-ordination amongst interdependent organizations, political lobbying, and executive succession where selection impacts on managers’ ability to cope with critical contingencies. Subsequent empirical work investigates the validity of these strategies, providing supporting evidence for their ability to aid organizational performance via retaining or improving their autonomy (Drees & Heugens, 2013; Hillman et al., 2009).

While influential and persuasive, two important caveats to RDT are evident. Firstly, Pfeffer and Salancik (2003) conceptualize resources as financial or physical. This downplays the importance of intangible assets such as Corporate Reputation (CR). CR is a ‘unique, intangible, status-based asset’, that represents ‘the aggregated opinion of the stakeholder community and is co-created by the interplay of organizations, their stakeholders, and the competitive environment’ (von Berlepsch et al., 2022). As it is determined in part by the perceptions of actors external to the organization, CR is thus a resource dependence. For many organizations, their most valuable assets are brand names linked, inevitably, to CR.

Secondly, in considering social evaluations, it is useful to distinguish between search, experience, and credence resources. Exchange relationships between supply chain partners traditionally focused on search (apparent prior to purchase) or experience (apparent during and after use) attributes (Pullman & Dillard, 2010). For instance, when diaries buy milk from farmers, standard testing for somatic cell counts, as well as fat and protein content, ensure procurement relates to search and experience qualities. Credence attributes (Darby & Karni, 1973) cannot be verified easily by the purchaser either prior to, during, or post use. In the dairy case, the welfare of animals or workers on a suppling farm are credence attributes, which the buyer cannot deduce from traditional tests for measuring intrinsic quality, such as that employed in testing for somatic cell counts or measuring fat and protein content. The evaluation of credence attributes, thus, depends on a dedicated infrastructure.

The credence nature of social evaluations has important implications for understanding resource dependencies and management’s attempts to preserve and expand organizational autonomy in the face of environmental uncertainties. Consumers and wider publics increasingly care about the social outcomes of supply chains and revelations regarding the poor treatment of workers can lead to a loss of trust in brands and tarnish CR (Narula, 2019). For focal organizations, poor treatment of workers by its suppliers, thus, presents a critical risk (Locke et al., 2007). However, the treatment of workers by their suppliers is not easily observable, especially when production is geographically fragmented within complex supply chains. Moreover, extensive monitoring increases transaction costs substantially (Short et al., 2016). In the case of fresh fruit and vegetables, retailers face intensive price competition, and cost and flexibility pressures in supply chains increase the attractiveness of casual labour and informal employment arrangements (Henry & Rivera, 2019). The latter places workers most at risk of exploitation and are difficult to monitor (Locke et al., 2009; Narula, 2019). Consequently, managers of focal organizations thus face an external environment characterized by actors with multiple and conflicting demands (Pfeffer & Salancik, 2003).

To address these critical risks, lead firms increasingly impose on their suppliers’ social evaluations and codes of conduct, governing the treatment of workers (Locke et al., 2007, 2013; Short et al., 2016). Typically, it is the responsibility of suppliers to prove compliance with the focal organization’s code of conduct, reducing the latter’s financial and administrative burdens. Third-party social auditors monitor compliance, but the extant literature provides ‘little light on how it works in practice’ (Short et al., 2016, p.1880), with several authors questioning the legitimacy and effectiveness of the social evaluations implemented by focal organizations for their supply chains (LeBaron & Lister, 2015; Locke et al., 2009; Rossi, 2019).

Legitimacy refers to ‘a generalized perception or assumption that the actions of an entity are desirable, proper or appropriate within some socially constructed system of norms, values, beliefs, and definitions’ (Suchman, 1995, p.574). As legitimacy depends on the perceptions of others, it is a resource dependency; a dynamic constraint that alters as organizations adapt to their environment and the norms and values which define legitimacy change (Dowling & Pfeffer, 1975). While never in full control of their legitimacy, organizations can adopt legitimation strategies (Bitektine & Haack, 2015; Haack et al., 2014; Suchman, 1995).

Legitimation strategies seek to help gain, maintain, and/or recover legitimacy, so that key stakeholders perceive that an organization and its specific activities are appropriate. While several typologies exist, two important forms are pragmatic legitimacy, which relies on the self-interested calculations of an organization’s stakeholders, and moral legitimacy, based on normative approval (Bitektine et al., 2020; Suchman, 1995). Pragmatic legitimacy is exchange related, depending on key stakeholders accepting the usefulness of an organization’s actions (Bitektine & Haack, 2015; Bowen, 2019). On pragmatic grounds, other supply chain actors may accept social standards as a mechanism to protect the reputation of the retailer or the supply chain, identify and remove from the supply chain ‘rogue actors’, prevent further state regulation, and to learn from others (Bowen, 2019). However, legitimacy may not solely rest on transactional interest alignment.

Moral legitimacy focuses on the ethical foundations of an organization or its activities (Bowen, 2019). Two important forms of moral legitimacy relate to outcomes and what is accomplished (consequential legitimacy) and the perceived soundness of procedures (procedural legitimacy) (Bitektine et al., 2020; Suchman, 1995). The moral legitimacy of social standards would thus depend on their effect on the treatment of workers within supply chains, the good-faith efforts of the focal organization, and the propriety of monitoring, auditing, and processes (Suchman, 1995). Understanding the nature and effectiveness of different legitimation strategies can help understand the implementation of social standards, addressing calls for further research on social evaluations of supply chains (Wettstein et al., 2019).

Methodology

The study adopted an inductive, interpretive approach, focusing on a revelatory, single case study (Gioia et al., 2012). The unit of analysis is the implementation of social evaluations for fresh fruit and vegetables by a large multinational retailer. In 2020, the retailer’s turnover exceeded €75 billion, and it had more than 350,000 employees. It operates in 23 European countries, with food retailing being its core business. The case study focuses on the period from 2011 to 2020, as in 2011 the company’s Board of Directors first announced its intention to ensure fair working conditions in its supply chains. The case finishes in 2020, with four years of data relating to the implementation of social evaluations of suppliers. An inductive, interpretative approach focusing on a specific case helps generate novel insights which are grounded in practice (Ciulli et al., 2020).

Data Collection

We collected data from multiple sources including semi-structured interviews, internal company documents, reports from Südwind e.V., an environmental and labour rights NGO, and non-participant observation. Table 1 provides an inventory of data, detailing sources and intended audiences. Data include a purposive sample of 12 semi-structured in-depth interviews with key actors. Interviewees were selected based on their ability to provide informed explanations (Patton, 2015). Interviewees included: three managers from the retailer, a consultant to the retailer, a representative of GlobalG.A.P, an auditor for SEDEX / SMETA, an NGO manager working in the field of labour standards, a member of the Fair Trade International Symposium steering committee, a producer, and three suppliers. This provided a range of views from relevant stakeholders, allowing for triangulation (Welch & Piekkari, 2017). Table 2 provides brief portraits of interviewees but to preserve the anonymity of interviewees, reflecting the small circle of key actors involved, we omit detailed descriptions (Pratt et al., 2019). The researchers’ judgment guided the sample size (cf. Baker, 2002) and given the extensive expert knowledge of the interviewees in this field, the sample size was appropriate for our purposes.

Table 1 Data inventory
Full size table
Table 2 Profile of interviewees
Full size table

Interviews occurred between Autumn 2020 and Spring 2022 and lasted, on average, between 60 and 90 min. All interviews were conducted with a proviso of preserving the respondents’ anonymity and following common guidelines in management case studies (Gioia et al., 2010) we do not reveal the name of the retailer. Interview questions varied according to the role of informants—to illustrate, Appendix 1 provides an example of the protocol for interviews with social evaluation auditing bodies. Notwithstanding variations according to role, all interviews incorporated questions relating to the company’s implementation of social evaluations, compliance procedures, experienced impacts and challenges, and future expectations. If agreed by participants, interviews were recorded and transcribed for analysis. If audio recording was not agreed, as in three cases, the interviewer(s) took detailed notes. One interviewer took the lead in interviews, but two other authors participated in some interviews, taking notes, and asking supplementary questions. As interviews took place during the Covid-19 pandemic, all occurred remotely using video conferencing software.

Data Analysis

Although the literature provided us with a theoretical sensitivity to legitimacy and resource dependencies, there was no explanation in the literature as to how they may be related in the case of social standards. This lack of theoretical insight therefore, meant that we did not develop an a priori coding framework. Rather, analysis of secondary and primary data occurred inductively, following the approach of Gioia et al. (2012). We began with in vivo coding of interview transcripts and other evidence, generating first-order codes (Gioia et al., 2010). These codes reflected the language and meaning of informants and documentary evidence. Subsequently, axial coding occurred, to identify second order themes, seeking to draw connections between codes, connecting subcategories or types of concepts (Bazeley & Jackson, 2019). Attention focused on understanding the legitimization and implementation of social evaluations of suppliers. In this second stage, we interrogated evidence, informed by legitimacy and resource dependence theories, which yielded insights for understanding patterns within the data but without coding data according to any pre-set theoretical framework (Ciulli et al., 2020). Data analysis proceeded through a discursive process involving all the researchers (Gioia et al., 2010).

Ensuring Trustworthiness

Reflecting the four quality criteria for qualitative research identified by Lincoln and Guba (1985), we undertook steps to ensure trustworthiness, as summarised in Table 3. Regarding credibility, Gioia et al. (2010) recommend giving voice to a knowledgeable insider who can best articulate the rationale for the conceptions and actions that constitute organizational processes and assist in crafting managerially relevant knowledge. Consequently, we engaged a consultant to the retailer, who was integral to the retailer’s implementation of social evaluations, as an “inside” researcher. The “inside” researcher was, thus, both an actor and analytical observer of the process, given his academic and environmental NGO background. To avoid insider bias, all the other authors were researchers with no direct connection to the retailer apart from through the research project (Gioia et al., 2010). The “outside” researchers checked the interpretations of the “inside” researcher against documentary and interview evidence from other informants.

Table 3 Data analysis procedures
Full size table

The inside researcher contributed to the analysis and provided post-analysis comments, acting as a form of member check, to ensure it was accurate and credible. Furthermore, the inside researcher shared our key sensemaking and findings with actors in the focal organization, providing them with an opportunity to identify any false conclusions, misinterpretations, or knowledge gaps that may have eluded us. This collaborative process fostered a comprehensive understanding of their perspectives and ensured the accuracy of our analysis. Prolonged engagement occurred through the inside researcher’s association with the retailer (including 7 years in relation to social evaluations), and insider—outsider author interactions occurred over a four-year period as partners in an EU-funded project. Interview and documentary evidence were triangulated (Silverman, 2017).

Regarding transferability, we provide a thick description of the case, drawing on multiple sources, with the intention of allowing readers to assess whether the findings are transferable to other settings with which they are familiar (Korstjens & Moser, 2018). To ensure dependability and conformability, we employed purposeful sampling (to capture and control variation), verified the interpretation of conclusions though insider–outsider researcher dialogues, and maintained records of data collection, data management, analysis, and reporting (Ciulli et al., 2020; Korstjens & Moser, 2018; Sonenshein, 2014).

Findings

Figure 1 presents the data structure, the aggregate dimensions of which inform a process model of the implementation of social evaluations for suppliers (Fig. 1). Table 4 provides representative quotations from interviews and documentary evidence to substantiate the identified second-order themes.

Fig. 1
figure 1

Process model of the implementation and legitimization of social evaluations for suppliers

Full size image
Table 4 Representative interview quotations and documentary evidence, underlying second-order themes
Full size table

Phase 1: Focal Organization’s Commitment to Social Evaluations

The company’s Board of Directors approved guidelines in 2011, which declared that the retailer would work with suppliers to develop binding standards to meet ecological and social challenges in food supply chains. A specific pledge to uphold the core conventions of the International Labor Organization (ILO) within its own operations and its business relations with contracted partners was made. The ILO has eight core conventions (of 189 conventions and treaties), which cover collective bargaining, abolition of forced and child labour, and discrimination at work. The company’s commitment applied to all its brand names, subsidiaries, and the countries in which it operated. The 2011 commitment exceeded legal requirements at the time. The retailer’s approach, thus, moved beyond obligatory (Hendry & Vesilind, 2005) or minimalist (Hemphill, 1997) Corporate Social Responsibility (CSR) strategies.

The Board of Directors tasked its sustainability unit, which is based at the company’s headquarters in its domestic market, with developing systems to ensure compliance with ILO core conventions for its fresh fruits and vegetable suppliers. Systems were first developed and implemented for the domestic market, and then refined based on careful consideration of what worked, before being transferred to other countries in which the retailer operates. However, the sustainability unit is largely peripheral to the retailer’s day-to-day operations, both domestically and internationally, and to implement social evaluations required the agreement of larger and more powerful divisions within the retailer, most notably buyers, who operate according to very different performance targets. The sustainability unit’s employees and consultants, who largely had backgrounds in environmental NGOs rather than retail management, thus first needed to secure the legitimacy of social evaluations of suppliers within the organisation.

To secure internal acquiescence, the sustainability unit appealed to three forms of legitimacy. Firstly, the sustainability team promoted a moral legitimacy argument, which followed a prosocial logic. This encapsulated the beliefs that “supply chains should be fair” (Retail manager) and “We have a responsibility to workers in our business partners” (Retail manager). However, to secure commitment from managers in other business divisions, the sustainability team also made pragmatic claims. The latter perceived that the introduction of social evaluations was in the retailer’s business interest, as well as a manager’s own interests, as “violations are linked back to us” (Retail buyer). This recognised that poor treatment of workers by suppliers was a reputation risk so that “a comprehensive analysis of the sustainability risks…formed the basis for strategy development” (Retailer’s sustainability report). Underpinning this combination of moral and pragmatic forms of legitimacy, was also structural legitimacy, based on the Board of Director’s stated commitment to uphold the core ILO conventions inside the organization and its suppliers. Structural legitimacy, based on formal power structures, increased the influence of the sustainability unit, which was often marginal to the retailer’s routine operations, in their negotiations with other business divisions. As a consultant to the retailer noted “The commitment came from the top, so it was not ignored.”

Phase 2: Specification of Social Standards

After securing broad internal commitment for social evaluations, the sustainability unit devised processes for their implementation. In this phase, two forms of legitimation strategies prevailed: consequential legitimacy (outcomes and accomplishments) and procedural legitimacy (soundness of processes). While Suchman (1995) classifies consequential and procedural legitimacy as forms of moral legitimacy, in this case pragmatic elements were also prominent. Specifically, the sustainability team sought to devise a system that would lead to improvements in supply chain working conditions, with robust procedures, but which would not destabilise procurement or place a high administrative burden or costs on the retailer.

To decide on the specific specification of standards, the sustainability team undertook a review of third-party standards schemes in conjunction with an external expert, evaluating them regarding whether they ensured compliance with ILO core conventions. The review concluded that, by and large, seven different standards schemes (GLOBALG.A.P. GRASP, BSCI amfori, SEDEX/SMETA, SIZA, Fairtrade, Rainforest Alliance, SA8000) ensured compliance with ILO core conventions regarding the elimination of all forms of forced labour, freedom of association, and right to collective bargaining, minimum wage, ending the worst forms of child labour, equal remuneration, and elimination of discrimination.

The seven schemes, however, vary considerably in terms of their requirements, objectives, and focus (Table 5). For instance, only SA8000 includes a living wage requirement, that is sufficient to meet the basic needs of workers and provides some discretionary income. Audit frequency also varies from between every 6 months to 3 years. Some schemes operate globally, while others focus on developing economies, or a specific country in the case of SIZA. Following the review, the retailer decided that all fresh fruit and vegetable suppliers must be certified under one of the seven social standards schemes, which thus act as external judgment validation institutions (Bitektine & Haack, 2015). Some suppliers possess more than one social standard certification, particularly if they supply several grocery retailers internationally, as retailers differ in their requirements. For instance, one major UK retail chain asks for BSCI amfori and does not recognize GLOBALG.A.P. GRASP, while a leading competitor in Germany accepts the latter but not the former. Individual evaluators judge a supplier’s procedures against the pre-specified criteria detailed in the standard’s scheme. As the focal organization decides on which standards it accepts, indirectly it controls which norms apply in judging propriety and validity (Haack et al., 2012).

Table 5 Comparison of the most important social evaluations
Full size table

The most common social evaluation is GLOBALG.A.P. GRASP, covering over 90 per cent of the fruits and vegetables procured by the retailer, although some of the other social evaluation schemes are more prominent for specific products. For example, more than 75 per cent of bananas procured are Fairtrade or Rainforest Alliance certified. GLOBALG.A.P. was founded in 1997 as a European retailer led initiative, focused initially on environmental and animal welfare issues, where G.A.P. stands for Good Agricultural Practices. The impetus for GLOBALG.A.P. was growing consumer concern regarding food safety and changes in European food law, making retailers responsible for ensuring that the food which they sell is fit for human consumption. GLOBALG.A.P. Global Risk Assessment on Social Practice (GRASP) is a more recent, voluntary module which assesses social practices on farms, considering specific aspects of workers’ health, safety, and welfare. It intends to ensure compliance with ILO core labour standards and suppliers should also adhere to national employment law (e.g., on specification of labour contracts, pay slips, working hours, and breaks). As employment laws differ from country to country, national interpretation guidelines exist. In some cases, GLOBALG.A.P. GRASP details rules beyond those enshrined in national law. For instance, the guidelines for Slovakia state: “For GRASP, a recording system of working time shall be included even when is not mandatory by the local law. Records are regularly reviewed by the employees and accessible for the employees’ representative(s).” (GRASP Module—Interpretation for Slovakia 10/15, mandatory from February 2021), whereas the guidelines for Vietnam read “If a daily time recording system is not implemented (e.g., fixed contracts, fixed working hours, fixed monthly salary), alternative way(s) of recording working hours shall be available. Evidence and explanation shall be provided.” (GRASP Module—Interpretation for Vietnam 10 /13, mandatory from February 2021). GRASP audits focus on whether suppliers have appropriate systems in place and usually occur three times a year and take typically half a day. As noted by a GLOBALG.A.P. representative: “the auditor checks the system, not individual complaints”, although a yearly report is also published relating to individual grievances. To date,  GRASP audits have not required interviews with individual workers, although this will change for countries rated as possessing medium and high risks of worker exploitation, as classified by Social Accountability International.

The costs for obtaining the social standards certification rests with the supplier. These can vary from a few hundred euros as part of another audit (like GRASP with GLOBALG.A.P.) and rise to a few thousand euros for a large co-operative undergoing a SA8000 audit. Given the volume of produce required by major retailers, the cost of social evaluations relative to total contract value tends to be small, although there are rare cases where the costs of social standards certification are an issue for suppliers. GLOBALG.A.P. GRASP’s ubiquity in part reflects that it can be “bolted on” to the core GLOBALG.A.P. audit (covering technical specifications and production processes) without additional audit visits, and typically is the cheapest option for the supplier.

Phase 3: Social Evaluations in Practice

The retailer’s implementation of social evaluations began in 2015, consisting of activities which both enforce and enable legitimacy (Belizón & Kieran, 2022). The enforcement element involves checks on compliance with failure leading to, in the case of the worst breaches, sanctions and relationship termination. In addition, an enabling element consists of training courses to explain requirements for new suppliers and production sites, as well as the retailer providing advice on how to prepare for an initial audit.

Regarding enforcement, typically, the retailer’s buyers meet with fresh fruit and vegetable suppliers to evaluate their performance twice a year. Social evaluation is combined with pesticide residue testing. Every week, approximately 50–60 samples are monitored regarding compliance against the retailer’s criteria (traceability, social standards). Results, stored in a database, allow for an analysis of trends and monitoring. In 2019, over 1200 samples were scrutinized. Table 6 details the geographical coverage of sampling, highlighting the global reach of evaluations across very different institutional environments. If traceability checks are passed, the retailer inspects compliance with social standards. The first part of the social evaluation checks whether necessary documentation is in place. Where documentation is missing, suppliers have ten days to respond. Missing documentation could be a minor paperwork issue or may indicate more serious non-compliance problems.

Table 6 Geographical coverage of samples taken, with audit of social standards
Full size table

Regarding enforcement, it is up to the retailer to decide whether to ‘prosecute’, so that only the focal organization possesses the ability to impose sanctions on suppliers. Social evaluations are confidential and only available to the relevant producer, retailers, and the designation agency (e.g., GLOBALG.A.P. or SMETA). There is no obligation on the retailer to act on an evaluation; when a supplier fails a social evaluation or serious problems are uncovered, it is up to the retailer to decide on any actions. In our case, the retailer’s buyers make the decision guided by advice from the sustainability unit. In response to a social evaluation, there are three potential outcomes: relationship maintenance, where there are no compliance issues, formation of a remediation plan, where failings are apparent but not judged a strategic threat to the retailer, and relationship termination, where failings are judged very severe and/or previous remediation plans failed. In very serious cases of infringement of ILO core standards like child/forced/endangered labour, immediate relationship termination may occur even if it leads to disruption of product availability in stores. While unusual, the fear of “knowing about it and then not seeing to act on the information, if it came out” (consultant to the retailer) motivates a speedy response. The retailer’s buyers and sustainability unit both believed delaying action in such circumstances inappropriate, on a combination of both moral and pragmatic grounds (risk to workers and CR). However, in most cases, non-compliance is less stark—the most common cause of failure to comply was employees voluntarily working over 60 h per week, although for virtually all countries, the GLOBALG.A.P GRASP guidelines are very clear in this regard “For GRASP compliance, even when permitted by the law, the total number of hours, including overtime and ordinary, shall not exceed 60 h per week in any week of the year” (e.g., Guidelines for Brazil). In these cases, the retailer agrees a plan of remedial actions with the supplier to ensure compliance. Figure 2 details that since being instigated in 2015, compliance with social evaluations rose from roughly 50% to over 80%, in 2019. Suppliers becoming more familiar with the paperwork requirements accounts for some of the improvement in compliance rates but not all—meaningful improvements in working conditions for some suppliers’ employees also occurred.

Fig. 2
figure 2

Source: Analysis based on data from the retailer

Percentage compliance with social standards (Years 2015–2019).

Full size image

While some suppliers reject the legitimacy of social evaluations, and this is a cause of non-compliance, the vast majority accept their legitimacy on both moral and pragmatic grounds. For some “it is the right thing to do” (supplier), while also a form of self-interest (pragmatic legitimacy). For instance, one supplier argued that during the ‘Arab Spring’ previous fair treatment of workers prevented disruption, which affected “those that cared less about their workers.” As an auditor noted: “suppliers having gone through the process…, often welcome the outcomes, as it helps them have better workplace systems and processes.” For most suppliers, the cost of GLOBALG.A.P GRASP is modest when compared against their turnover, and “a small price to pay” (supplier) for maintaining market access. Only in a few instances has the introduction of social evaluations led to the termination of business relations with a supplier.

The most problematic cases for securing compliance with social standards have occurred in emerging economies, namely: legumes from Central Africa and West Africa, grapes from India, garlic from China, and some fruits from Asia such as mango, dragon fruit, and pineapple. Two common problems underpin non-compliance in such cases. First, the implementation of social standards is dependent on governance and the rule of law and is much easier where national labor laws already conform to ILO core conventions and are enforced effectively. In contrast, for example, in Thailand the implementation of social standards has been challenging because the government’s control system falls short of what is required for international standards. Secondly, problems emerge where there is a vast number of suppliers and traceability is complex, as enforcement of social standards relies on the latter. In problematic cases, the retailer increased sampling, and if problems persist sourcing products from other countries is preferred.

For the retailer, relationship termination is, however, not the preferred option for both pragmatic reasons (supply disruption, fewer procurement options) but also a belief that once the relationship ends, the retailer loses its leverage to improve working conditions. Terminating the relationship completely may, thus, do little to help the employees of suppliers. Often, remediation plans achieve their objectives and improve suppliers’ labour practices. For instance, evaluations revealed cases of non-compliance with minimum wages in Columbia and Kenya. The retailer did not immediately terminate the relationship and the subsequent audit revealed that minimum wages were being paid. As a consultant noted “I’m quite confident that if we had immediately terminated the relationship, or not had the social standards, that would not otherwise had happened.”

Remediation plans, however, do not always work. In some cases, suppliers refuse to engage because they believe that the retailer has no legitimate grounds for interfering in how they treat their workers or regard the process as unimportant. In others, the threat or realization of relationship termination is necessary for action to occur. For instance, one supplier of figs from Turkey made no progress on a remediation plan in the 3 years following an audit. Consequently, the retailer terminated the relationship. This led to a change in the supplier’s approach and management practices, so that the relationship resumed later. Sometimes, suppliers make progress in remedying some issues while ignoring others. For instance, a supplier of broad beans from Senegal provided written labour contracts while still allowing security guards on the farm to work more than the permitted 60 h per week. In these cases, the retailer’s buyers, based on an assessment by the sustainability unit, make a judgement as to whether to persist or end the relationship.

Discussion and Conclusions

Recent reviews call for greater attention to be paid to MNC’s implementation and enforcement of social evaluations for their supply chains (Aragòn-Correa et al., 2020; Bowen, 2019; Kano et al., 2020), recognizing “business and human rights (BHR) as a research area with great potential” (Wettstein et al., 2019, p.54). In response, this paper considers the legitimization and implementation of social evaluations for a focal organization’s suppliers, drawing on a case study of a multinational grocery retailer.

The case study highlights that social evaluations of supply chain partners depend on processes that reconcile both moral and pragmatic concerns, in ways deemed acceptable to the focal organization. Without satisfying pragmatic concerns, social standards fail to gain the support of buyers within the focal organization, whose co-operation is essential for implementation. However, without being able to identify some improvements in social standards in supply chains (consequential legitimacy) and appropriate processes for their assessment (procedural legitimacy), the raison d'être of the initiative would be questionable, leaving the focal organization with social standards that lack structural legitimacy (Bitektine & Haack, 2015). Without suitable moral legitimacy, the focal organization thus leaves itself open to claims of mere “greenwashing” and a potential media backlash (Tian et al., 2021; Vestergaard & Uldam, 2022). Consequently, the structural legitimacy of social standards depends on establishing processes that reconcile both moral and pragmatic concerns.

The reconciliation of pragmatic and moral concerns, however, is not without problems. One tension involves trade-offs between procedural and pragmatic legitimacy. The most common form of social standard assessment is GLOBALG.A.P. GRASP, which is typically the cheapest option for suppliers. Given that audit costs relate to the length of time an auditor spends on a farm (including whether they interview workers or not), this could lead to social evaluations with the lowest levels of on-site inspections (weakest procedural legitimacy) becoming the industry norm. An option for the retailer would be to insist on a more restrictive list of approved standards or a single ‘higher level’ standard. However, this clashes with pragmatic concerns—reducing procurement options could affect product availability in store, shift the balance of power to those few suppliers which meet a higher standard, and increase the average costs of fruit and vegetables procured. Given low margins in the grocery retail sector as well as high consumer price sensitivity (Shankar & Krishnamurthi, 1996; McKinsey & Company, 2021) and strong competition (Steenkamp, 2018), such an action if taken unilaterally by one retailer is likely to weaken its competitive position (Lin-Hi & Blumberg, 2017).

A second tension relates to trade-offs between the self-interested appeals to key stakeholders for pragmatic legitimacy, specifically between control and reputation protection on the one hand and network and learning benefits on the other. Conceptualizations of pragmatic legitimacy typically treat the focal organization as homogeneous, with appeals to self-interest made to external stakeholders within the supply chain (Bowen, 2019). However, the case study illustrates that different functional units within the focal organisation possessed varying incentives and priorities, so that the implementation of social standards required the sustainability unit to make convincing, intra-organizational pragmatic arguments, especially to buyers (e.g., protecting the organization’s CR while retaining strategic control). The pragmatic appeal to suppliers was similar—protecting the supply chain from disruptions, rogue operators which could under-cut them, and insulating them from the threat of more costly and onerous, state regulation. However, the processes designed to achieve these benefits, controlling the nature and direction of information flows, led to few supplier-to-supplier network or learning benefits. The latter, ‘learn through observing others’ benefits (Bowen, 2019, p.261) are difficult to realise when processes prioritise centralised control of information.

A ubiquitous criticism of industry self-regulation is that it allows lead firms to ‘mark their own homework’ (Bowen, 2019, p.267). However, as the process model presented in Fig. 2 reveals, this is not a precise conceptualisation for retailers’ social standards. Specifically, third parties, rather than retailers, assess producers’ procedures, although retailers decide on, to extend the homework analogy, which markers to accept and play a role in shaping the assessment criteria employed by some markers (e.g., in the case of GLOBALG.A.P. GRASP). Of more concern is what happens to homework once it is marked. Specifically, evaluations are subject to confidentiality agreements, so that they are only visible to the retailer, standards body, and the assessed producer. This closed information system (Hong Telvin Goh and Hooper, 2009) grants the focal organizations’ buyers considerable leeway in terms of how they respond—there is no compulsion to act on a negative evaluation. While the retailer’s buyers and the sustainability team in our case were proactive in their response to negative assessments of suppliers, leading to many meaningful improvements in working conditions, other retailers may be blasé, lacking the capacity or will to respond to evidence. Crucially, third-party social evaluations, thus, preserve a remarkable degree of autonomy for the focal organization. Once “homework” is returned, there is no direct requirement to act on low marks or report them to anyone else.

In their analysis of initiatives to promote responsible labour standards in the toy industry, Lin-Hi and Blumberg (2017) found that as many leading brands continued to source from non-compliant factories, there was little incentive for suppliers to engage with industry self-regulation, which could weaken their competitiveness. The case reviewed in this paper documents stricter procedures with the retailer committing to sourcing only from suppliers compliant with ILO-core conventions, with monitoring and sanctioning processes implemented. Consequently, some noticeable improvements in the treatment of workers are evident (consequential legitimacy), which should not be dismissed as merely symbolic (Christmann & Taylor, 2006). However, in giving suppliers a choice between seven standards schemes, the most popular choice has been the cheapest option (typically GLOBALG.A.P. GRASP). Given the acceptability of all seven options, there is no direct incentive for suppliers to ‘upgrade’ to more demanding options such as SA8000, which includes a living wage requirement, that could leave the supplier at a competitive disadvantage. The imposition of social standards while delivering incremental improvements does not yet provide a route to radical, systematic changes in supply chain labour practices.

Finally, beyond the case of social standards, implications for RDT can be drawn. Specifically, the paper identifies how CR can be regarded as a resource dependency with social evaluations a mechanism for co-opting external actors to preserve the focal organization’s autonomy and legitimacy, distinct from the five strategies identified by Pfeffer and Salancik (2003). Specifically, third-party social evaluations offer a means of legitimizing the focal organization’s involvement in suppliers’ human resource practices, both inside the MNC and amongst suppliers, on both moral and pragmatic grounds. Specifically, they offer a mechanism for reducing environmental contingencies while preserving the autonomy of the focal organization for independent action. In contrast to the strategies outlined by Pfeffer and Salancik (2003), third party social evaluations avoid the focal organization’s managers becoming directly involved in the operations of often fragmented and geographically disparate external actors. Consequently, they minimize the MNC’s administrative costs and allocation of management resources. Nonetheless, social evaluations provoke some meaningful improvements for workers in the supply chain, while also acting as a “prophylactic” (Bitektine, 2011, p.157), protecting the focal organization’s CR from negative spill-over effects (Haack et al., 2014).

Limitations and Future Research

Case studies illuminate contemporary phenomena within their real-life contexts and benefit from conceptual simplicity but may lack generalizability (Gioia et al., 2010; Yin, 2018). While this paper focuses on a single case, other European grocery retailers face common environmental contingencies, and responded in a similar manner, introducing social evaluations of their suppliers. While there are some differences across retailers in terms of which third-party social standards they accept, interview and documentary evidence collected suggest that implemented processes are similar. Nonetheless, the topic warrants further case studies to test the analytical generalizability of the process model, and legitimation strategies.

The assessment of social evaluations provides information on rates of compliance, commonly identified problems, and the fortunes of remediation plans. However, none of the social evaluations recognized by the retailer provide the basis for the accurate measurement of socioeconomic effects. This limits our current ability to assess the contribution of social standards to achieving SDGs, which remains an important research task. Moreover, there is a dearth of information on the fortunes of suppliers’ workers after a retailer terminates a business relationship. Investigating this could help better understand the relationship between inclusion and exclusion in a focal organization’s supply chain and social and economic outcomes (Gereffi & Lee, 2016).

Finally, at present, the impact of recent or likely future public regulation (e.g., EU Code of Conduct for responsible food businesses and marketing practices) on MNC’s relationships with their supply chain partners is unclear. A major question for MNC procurement managers is the degree to which their systems for social evaluations ensure compliance with government regulation. In parallel, assessing the socioeconomic impacts of enhanced government regulation of supply chain practices is an important task for future research. Notwithstanding these limitations, we hope that this case study sheds light on this vital area and will inspire future scholars to embark on empirical investigations, leading to a positive change.