2. Literature review

(1)

(2)

2

Abstract

Partnering up with NGOs is an increasingly popular strategy for MNEs to develop and implement adequate CSR-policies in their global value chains. However, it remains unclear how these unconventional cross-sectoral collaborations can be effectively governed. This study explores what governance mechanisms are used in these MNE-NGO partnerships, and which are effective, by examining four similar partnerships aimed at implementing human and labour right policies in the MNE’s value chain. An embedded case study design is used to examine the following partnerships: Oxfam Novib – Unilever Vietnam, Oxfam Novib – Unilever Indonesia, Oxfam Novib – Sanorice and as an addition to the embedded case, Fair Labor Association – Nestlé. By adopting a qualitative approach through the use of interviews, both the perspectives of the MNE and the NGO are incorporated. The findings of this study show that formal contracts and relational governance are effective governance mechanisms and of complementary nature.

Additionally, the alternative governance mechanisms of reputational hostages and power dependence were introduced, of which reputational hostages proved to be an effective governance mechanism.

Keywords: MNE-NGO partnership, governance mechanisms, CSR-implementation, global value chains, formal contracts, relational governance, reputational hostages and power dependence.

(3)

3

1. Introduction

In the wake of globalization, multinational enterprises (MNEs) are becoming more and more aware of the social and environmental impacts of their global value chains (Pedersen &

Andersen, 2006). Nowadays it’s not just about being profitable anymore, but also about being a good corporate citizen (McWilliams & Siegel, 2001; Porter & Kramer, 2006). The potential backlash of irresponsible activities puts pressure on multinationals to protect their brands, even if that implies taking responsibility for their supplier’s activities as well (Amaeshi, Osuji &

Nnodim, 2008; Scherer & Palazzo, 2007). Being a good corporate citizen does no longer refer to the internal practices of the firm, but encompasses its whole value chain, including suppliers, third party logistic providers and intermediaries over which they have no ownership at all (Andersen & Skjoet-Larssen, 2009; Maloni & Brown, 2006). As a response to these increasing social and environmental pressures on their global value chains, many multinationals have incorporated the concept of Corporate Social Responsibility (CSR). However, it seems that the more multinationals have started to embrace CSR in recent years, the more they are being blamed for society’s failures (Porter & Kramer, 2011). The perceived legitimacy of MNEs is at an alltime low (Lodge & Wilson, 2016). A large part of this problem lays with the MNEs themselves, it appears to be hard to develop and implement the “correct” CSR-policies (Lund- Thomsen & Lindgreen, 2014). A contributing factor to this problem is that the MNE of today operates across a wide variety of countries, each with different cultural, political and economic norms (Aaltonen, Jaakko & Tuomas, 2008; Yang & Rivers, 2009). Therefore, the MNE faces conflicting environments in terms of CSR-standards, which complicates the process of choosing a CSR-strategy that is accepted by society (Matten & Moon, 2008; Marano &

Kostova, 2016).

An increasingly popular strategy of MNEs to deal with these different contexts and to implement CSR in their global value chains, is to partner up with NGOs (Seitanidi & Crane, 2009; Walters & Anagnostopoulos, 2012). Over the past years, large multinationals such as Starbucks, Mcdonalds, Nestlé and Unilever have started to collaborate with NGOs in order to address social and environmental challenges related to their value chains, aiming to improve overall CSR (Baur & Palazzo, 2011). Examples of such NGO-MNE collaborations to improve CSR are e.g. McDonald’s partnership with the Environmental Defense Fund (EDF) to reduce the environmental impact of its packaging and Unilever’s partnership with Oxfam Novib to reduce the poverty footprint of its supply chain in Indonesia. Participation of an NGO can provide valuable expertise and awareness of CSR-issues in local contexts where the MNE is

(5)

5 not as familiar (Boddewyn, 2003; Den Hond, De Bakker & Doh, 2015). Therefore, partnering up with an NGO that is embedded in the local context through contacts with communities, legislator and other stakeholders, can prove to be an effective tool to address CSR-issues in the value chains of MNEs (Oetzel & Doh, 2009; Poret, 2014). For instance, NGOs such as Human Right Watch, Oxfam Novib and Amnesty international have been cooperating with MNEs to help develop appropriate human rights policies in their value chains, mainly in developing countries where the local context is relatively unknown to the MNE (Dahan, Doh, Oetzel &

Yaziji, 2010). NGOs that have been working for years in these developing countries, addressing social issues such as human rights, are in a prime position to facilitate CSR-implementation in the value chains of MNEs within these countries (Baur & Palazzo, 2011; Poret, 2014).

Therefore, MNE-NGO partnerships have started to play an increasingly important role in overcoming the challenges that MNEs face in implementing CSR in their value chains (Walters

& Anagnostopoulos, 2012; Clarke & Fuller, 2010; Seitanidi & Crane, 2009).

Even though these MNE-NGO partnerships have proven to be a successful way of developing and implementing better CSR-policies in global value chains, the governance of these cross- sectoral collaborations has proven to be difficult (Berger, Cummingham & Dumwright, 2004;

Jamali & Keshishian, 2008). An MNE and a NGO are fundamentally different organizations, with different ways of working and worldviews, which makes collaboration problematic (Austin & Seitanidi, 2012; Selsky & Parker, 2005). Due to the relatively recent nature of these partnerships, it is unknown which governance mechanisms are most effective in facilitating the mutual development and implementation of CSR-policies. Traditional interfirm governance literature points towards the use of formal and informal governance mechanisms such as formal contracts and relational governance (Poppo & Zenger, 2002). However, it remains the question whether these mechanisms are also effective in cross-sectoral partnerships. Therefore, the purpose of this study is; (1) to explore governance mechanisms used in MNE-NGO partnerships and (2) how they facilitate the effective development and implementation of CSR-policies in global value chains.

Data was collected on four MNE-NGO partnerships aimed at developing and implementing human and labour right policies into the MNE’s value chain. Executives from both the MNEs and NGOs that were involved in these partnerships were interviewed and their perspectives on the governance mechanisms used recorded. Formal contracts and relational governance proved to be important in these partnerships and to be of complementary nature. Additionally, the

(6)

6 alternative governance mechanisms of reputational hostages and power dependence were introduced, of which reputational hostages proved to be an effective governance mechanism.

This study will proceed as followed: the next section provides a literature review on the nature, benefits and challenges of MNE-NGO partnerships and potential governance mechanisms that can facilitate the development and implementation of CSR into global value chains. Sections 3 and 4 cover the methodology, data and discussion of the findings respectively. Finally, section 5 provides conclusions, advice for MNEs and NGOs that want to engage in similar partnerships and, directions for future research.

(7)

7

2. Literature review

2.1 The challenges of implementing CSR in global value chains

In order to facilitate socially responsible business practices throughout their value chain, from the supplier of raw materials to the firm’s customers, multinationals have started to develop codes of conduct (Pedersen & Andersen, 2006; Schrerer & Palazzo, 2007). Codes of conduct are documents that state the CSR-standards of a multinational, of which it expects all involved parties it engages with, to comply to (Acquier, Valiorgue & Daudigeo, 2017; Scherer &

Palazzo, 2007). These CSR-standards in codes of conduct often refer to the social impacts of the global value chain (Gereffi & Lee, 2016; Kogg & Mont, 2012). Social impacts in global value chains can e.g. refer to the impacts that human and labour right policies have on the workers employed in an MNEs value chain, for which MNEs have received a large amount of negative publicity recently (Porter & Kramer, 2011). MNEs have been criticized for maintaining policies that might fit to the standards of the local environment, but are in no way acceptable anymore in the modern Western world, such as the acceptance of child and forced labour (Kolk & Van Tulder, 2002; Zutshi, Creed & Sohal, 2009). To address CSR-issues in this area, this research will focus on the development and implementation of human and labour right policies in global value chains (Andersen & Skjoett-Larsen, 2009; Gereffi & Lee, 2016). MNEs establish these policies in their codes of conduct to ensure compliance to these policies by the other members in its chain as well; if a member doesn’t comply to these human and labour right policies, access to the value chains and the MNEs business will be denied (Acquier et al., 2017).

Even though developing codes of conduct has improved CSR in global value chains to some extent, MNEs still struggle to be socially responsible throughout their global value chain and regularly receive negative attention (Acquier et al., 2017; Lund-Thomsen & Lindgreen, 2013;

Porter & Kramer, 2006). Codes of conduct have not been sufficient in terms of compliance to stakeholders, both locally and globally (Porter & Kramer, 2011). MNEs are having difficulties to bridge the gap between what they perceive as legitimate and what stakeholders perceive as legitimate in terms of CSR (Matten & Moon, 2008). Additionally, it proves to be a challenge to make local suppliers understand why these codes of conduct, which are often in contrast with local culture (e.g. the acceptance of child labor), have to be complied to (Campbell, 2006;

Gillman, 2005).

(8)

8 Previous literature has accounted these difficulties to the fact that MNEs are subject to different and conflicting institutional environments (Aaltonen, Jaakko & Tuomas, 2008; Matten &

Moon, 2008; Yang & Rivers, 2009). CSR practices are embedded in the institutional environments of which the MNE operates in (Marano & Kostova, 2016). Because institutional environments are for the most part country-specific, CSR-standards are different in every country (Matten & Moon, 2008). These differences serve as challenges for MNEs to implement CSR in their value chain in an appropriate way, since they operate in multiple of such environments at the same time (Aaltonen, Jaakko & Tuomas, 2008; Yang & Rivers, 2009).

MNEs have to be aware of the diverse behavioral assumptions, norms and regulations of stakeholders in every country that they are operating in. Therefore it is hard for MNEs to judge what is perceived as a legitimate social or environmental practice in host-country environments (Kostova & Roth, 2002; Meyer, Mudambi & Narula, 2011). Managers of MNEs have to assess which claims of host-country stakeholders are legitimate and need to be paid attention to in terms of CSR-implementation into their value chain (Lund-Thomsen & Lindgreen, 2013). It is important to know what is perceived as legitimate in the local context, because stakeholder support from e.g. the local government and population depends on it, this support is an important condition to be able to conduct business efficiently in a host-country environment (Lodge & Wilson, 2016).

For MNEs operating in different institutional and cultural environments, it can be difficult to assess which stakeholder claims are more important in those particular contexts (Teegen, Doh

& Vachani, 2004). These host-country environmental differences combined with the limited amount of resources and time that the MNE can dedicate to all these different contexts, makes it hard to adequately implement CSR in every value chain, according to that particular country’s social and environmental standards (Matten & Moon, 2008; Marano & Kostova, 2016).

Therefore, MNEs have started to engage and collaborate with NGOs to fill these knowledge gaps and to be able to develop and implement adequate CSR-policies in all of their value chains.

NGOs that already have the previously mentioned stakeholder support from local governments and populations can assist MNEs in bridging the gap between what the MNE perceives as legitimate and what the local stakeholders perceive as legitimate. Additionally, NGOs that have local ties can also assist in ensuring that local suppliers will comply to the MNEs codes of conduct, which has proven to be difficult (Campbell, 2006; Gillman, 2005).

(9)

9 2.2 MNE-NGO partnerships, benefits and challenges

Cross sectoral partnerships between MNEs and NGOs aimed at implementing CSR into global value chains can be characterized as social partnerships (Berger et al., 2004; Seitanidi & Crane, 2009). Social partnerships differ from traditional interfirm partnerships in that there is always one nonprofit actor involved and, objectives are focused on increasing social and environmental welfare as opposed to economic objectives (Berger et al., 2004). Therefore MNE-NGO partnerships represent the alignment of business interests of the MNE with societal expectations through the NGO (Austin, 2000; Seitanidi & Crane, 2009). Previous research on MNE-NGO partnerships has mainly focused on the benefits and challenges that these collaborations involve. For instance, MNEs are usually motivated by increased legitimacy and positive reputational effects in host-country environments through being associated with an NGO (Easterly & Miesing, 2009; Marano & Tashman, 2012). Furthermore, engaging in partnerships with NGOs offers the opportunity to co-create a better CSR-policy as part of the corporate strategy (Arya & Salk, 2006; Dahan et al., 2010), being aware of local social issues as part of that strategy (Boddewyn, 2003; Den Hond et al., 2015) and to access distinct networks of communities, regulators and public influencers to be able to leverage those strategies (Oetzel

& Doh, 2009; Poret, 2014). Therefore, it offers the MNE access to resources and knowledge in local contexts that were otherwise hard to obtain (Dahan et al., 2010). Additionally, these partnerships offer opportunities for the MNE to learn in the field of CSR and how to implement it in their value chains (Arya & Salk, 2006; Jamali & Keshishian , 2008).

The NGO on the other hand, is motivated by increased competition for funding (Robinson, 2012; Weerawardena, McDonald & Mort, 2010), due to many new entrants in the sector in recent years (Chew & Osborne, 2009). Additionally, shrinkage and uncertainty about government funding lead to more problems in terms of the funding of NGOs (Weerawardena et al., 2010). These conditions have led to NGOs exploring new sources of funding, to be able to continue their operations, while at the same time addressing social and environmental issues (MacDonald, 2010; Sagawa & Segal, 2000). Partnering with MNEs provides the NGO with financial resources they normally do not have access to (Austin, 2000; Wymer & Samu, 2003).

Also, these partnerships can greatly increase the visibility of the NGO, by being associated with the MNE and all its stakeholders (Baur & Schmitz, 2012; Herlin, 2015)

However, MNE-NGO partnerships have been subject to a steep learning curve; few know how to bridge the gap between corporate and nonprofit sectors (Berger et al., 2004). When fundamentally different organizations cooperate, it sets the stage for disputes, conflicting

(10)

10 management styles, different strategies and misalignment of objectives (Babiak & Thibault, 2009; Jamali & Keshishian, 2008; Regan & Oster, 2000). Differences in values, norms and motivations (Lefroy & Tsarenko, 2013; Seitanidi, 2012), goals and motivations (McFarlan, 1999) and organizational characteristics and structures (Berger et al., 2004; Selsky & Parker, 2005) serve as barriers to cooperation and large gaps in institutional logics (Friedland & Alford 1991; Thornton and Ocasio 1999). Therefore, to be able to reap the benefits of these partnerships, it is important for both the MNE and NGO to understand how these collaborations can be effectively coordinated (Baron, 2003; De Lange, Armanios, Delgado-Ceballos, &

Sandhu, 2016; Le Ber & Branzei, 2010). It remains unclear how the collaboration process between MNEs and NGOs is governed; “While existing literature emphasises the importance for MNEs to collaborate with various stakeholders, including non-governmental organisations (NGOs), research on the scope, peculiarities and impact of such cooperation has been limited”

(Kolk & Lefant, 2013, p. 47). Therefore, the nature of how these partnerships are implemented and managed is not well understood (Seitanidi & Crane, 2009). To gain insight on these problems in MNE-NGO partnerships, it is important to understand what kind of governance mechanisms are used in order to manage these unconventional cross-sectoral partnerships and for the interest of this research, how these mechanisms can facilitate the implementation of CSR into global value chains.

2.3 Non-equity governance mechanisms in MNE-NGO partnerships aimed at CSR- implementation in global value chains

In traditional interfirm partnerships, the use of equity is a common arrangement in order to deal with uncertainty (Das & Teng, 1996; Gulati, 1995). Equity creates a mutual hostage situation, which means both partners invest in each other (Das & Teng, 1996). This aligns the interests and objectives of partners, when the risk of opportunism is significant (Gulati, 1995; Kogut, 1988). This mutual hostage situation, usually leads to the creation of a separate entity, aimed at the formal coordination and control of the partnership (Casciaro, 2003). However, in MNE- NGO partnerships the use of equity is not realistic due to the nature of the NGO (Rivera-Santos

& Rufín, 2010). A NGO will typically not invest in a profit-oriented organization, and in most countries this is also illegal (O’Conner & Shumate, 2014). Additionally, it would be assumable that the followers of the NGO would also consider this form of governance as unacceptable, as it is one of the core values of NGO’s not to be involved in profit-oriented activities (O’Conner

& Shumate, 2014). Therefore, equity cannot serve as a governance mechanism in MNE-NGO partnerships in order to facilitate the implementation of CSR-performance in global value chains. This has important implications for governance in these kind of partnerships. Due to the

(11)

11 restriction of equity, these partnerships can supposedly only be governed through the use of formal contracts and relational governance, mechanisms identified by traditional interfirm governance literature (Mellewigt, Madhok & Weibel, 2007; Poppo & Zenger, 2002; Ryall &

Sampson, 2009. Besides the traditional formal and informal governance mechanisms in interfirm partnerships, previous literature points towards additional mechanisms that could potentially play an important role in cross-sectoral collaborations, such as reputation (Dellacoras, 2006; Mayer, 2006; Reuer & Devarakonda, 2016) and power dependence (Jones, Edwards, Bocarro, Bunds, Smith, 2017; Stoker, 1998). How can these mechanisms of governance facilitate CSR-implementation in global value chains?

2.3.1 Formal contracts

Formal contracts stand for the obligation to perform certain actions in the future (McNeil, 1978). From a transaction cost economics (TCT) perspective, contracts serve as defense mechanisms in interfirm relationships, to discourage parties to engage in opportunistic behavior (Das & Teng, 1998). Contracts are considered to be the primary non-equity form of interfirm partnership governance in terms of formal control and coordination (De Jong & Klein Woolthuis, 2009). Formal contracts can help to specify task division in interfirm partnerships, specify procedures for monitoring, dispute settlement and most importantly, determine what has to be delivered by whom in terms of outcomes (Poppo & Zenger, 2002). This is expected to be no different in MNE-NGO partnerships aimed at mutually implementing CSR in global value chains. For instance, Rabobank (a large Dutch bank) and the World Wide Fund (WWF) engaged in a partnership to mutually address climate change, utilizing the financial resources and large value chain of the Rabobank (chain of bank transactions), and the brand equity of the WWF (Gray & Purdy, 2018). In this partnership, the objective was to develop a climate-neutral credit card, with inbuilt CO2-emission compensation in order to eliminate the environmental impact of banking transactions within Rabobank’s value chain. However, in the first stages of the partnership a formal contract was missing and both parties failed to agree on issues such as the climate compensation method and the distribution of the costs and benefits associated with the card (Van Huijstee & Glasbergen, 2010). As previously mentioned, these are key governance problems that can be solved by formulating a formal contract, in order to specify task division and determine objectives.

Only after the first few difficult months of the partnership, Rabobank and the WWF decided to issue a formal contract, which solved these governance problems and eventually led to the

(12)

12 successful launch of their climate-neutral credit card (Van Huijstee & Glasbergen, 2010; Gray

& Purdy, 2018).

Accordingly, Walters & Anagnostopoulos (2012) in their study on the partnerships of UEFA (Union of European Football Associations, a profit-driven organization) with six different NGOs in order to improve CSR on multiple aspects, such as discrimination and emotional health in football, found that in all of these partnerships a formal contract was issued. These contracts defined the roles and responsibilities of the partners, the scope of the partnership and also, the amount of funding UEFA will provide to the NGO. According to all partners, the issue of a formal contract was crucial to define roles and responsibilities during the partnership design phase, especially since UEFA expected specific outcomes relative to the amount of funds the NGOs received (Walters & Anagnostopoulos, 2012). The latter is especially relevant to cross- sectoral partnerships aimed at implementing CSR in global value chains. It is common in these cross-sectoral partnerships that the NGO brings its expertise and reputation in addressing CSR to the table, in exchange for funding and resources from the MNE that can assist the NGO in doing so (Liu & Ko, 2011; MacDonald, 2010; Sagawa & Segal, 2000; Seitanidi & Crane, 2009).

Accordingly, Jamali & Keshishian (2008) found in their case study on five business-NGO partnerships that a formal agreement (contract) was a necessary partnership condition in order to determine what was expected of the NGO in return for the funds provided by the MNE.

Additionally, in the profit-driven sector in which the MNE is operating, it is standard practice to issue formal contracts when relation-specific investments take place in exchange for certain outcomes (Crawford, 1990; Lyons, 1994; Woolthuis, Hillebrand & Nooteboom, 2005).

Therefore, in MNE-NGO partnerships aimed at implementing CSR in global value chains, we assume that the same applies and that the MNE wants to have specific returns and outcomes outlined in a contract in exchange for the financial assets they provide. However, since MNEs and NGOs are so different in many perspectives, it would be assumable that formal contracts as only mechanism of governance wouldn’t be sufficient (Selsky & Parker). Namely, if key partnership conditions that are not listed and controlled by a contract, such as mutual trust (Carson, Madhok, Varman & John, 2003), are not present at all, it would be questionable whether MNEs and NGOs would even engage in a partnership (Rivera-Santos & Rufín, 2010), let alone to develop and implement CSR-policies together.

2.3.2 Relational governance

Many researchers, including transaction costs economists have speculated about that relation governance, through trust-building mechanisms, can mitigate against the same exchange

(13)

13 hazards as formal contracts can (Carson et al., 2003). According to Poppo & Zenger (2002), relational governance entails that the obligations normally formulated in contract clauses, will now be enforced through social processes that endorse norms of flexibility, solidarity and information sharing. Flexibility facilitates adaptation to unforeseeable events or future contingencies as previously mentioned in formal contracts (Brown et al., 2015; Eggleston et al., 2000). Solidarity promotes the commitment to take action together, e.g. the mutual implementation of CSR in MNE-NGO partnerships (Poppo & Zenger, 2002). Information sharing supports mutual problem solving and mutual adaptation when partners are willing to openly share information about each other’s goals and organizational processes (De lange et al., 2016). Relational governance therefore, is a trust-based mechanism that can mitigate exchange hazards without the expenses of continuously monitoring the relationship (Li, Xie, Teo, & Peng, 2010). Therefore, it is a self-enforcing mechanism (Li et al., 2010). However, in cross-sectoral partnerships it seems difficult to build a level of trust that is sufficient to serve as self-enforcing governance mechanism. Namely, due to the large organizational differences between organizations, initial trust is generally low and is hard to build throughout the partnership (Jonker & Nijhof, 2006; Selsky & Parker). Trust however, does play a key role in building capacities and complementary resources through these partnerships (Abramov, 2009).

Relational governance within partnerships is strongly dependent on inter-partner trust;

relational benefits can only be derived when there is a mutual amount of trust between partners (Ariño, de la Torre & Ring, 2001; Poppo & Zenger, 2002). Therefore, for relational governance to act as a self-enforcing governance mechanism that can complement the use of formal contract in MNE-NGO partnerships, partners have to build a sufficient amount of inter-partner trust (Rivera-Santos & Rufin, 2010). Inter-partner trust within MNE-NGO partnerships and its relevance as governance mechanism to facilitate CSR-implementation in global value chains can be described using the framework of inter-partner trust as described by Ariño et al. (2001).

According to this framework, inter-partner trust is heavily dependent on initial conditions and can be developed through frequent partner interaction.

Initial conditions

The initial conditions refer to the amount of trust between parties prior to engaging in a partnership (Arino et al., 2001). This amount of trust can be dependent on the common

(14)

14 characteristics, reputation and the prior partnership experience of both organizations. In cross- sectoral partnerships, these initial conditions prove to be vital (Austin & Seitanidi, 2012; Jamali

& Keshishian, 2009; Seitanidi & Crane, 2009). Common characteristics of organizations are usually fundamentally different and therefore it is hard to build trust (Selsky & Parker, 2005;

Simo & Bies, 2007). Where in traditional firm collaborations, partners have some kind of mutual understanding through their profit-oriented goals (Lucea, 2007), partners in MNE-NGO partnerships don’t have such common ground (Simo & Bies, 2007). NGOs are for instance not likely to build deep relationships with MNEs that have a bad reputation in terms of CSR, partly because it can damage their own reputation and also because, CSR-implementation will potentially be difficult with such a partner (Rivera-Santos, 2010; Simo & Bies, 2007). Similarly, MNEs might not want to engage in partnerships with NGOs that have a reputation to push their ideals very aggressively on firms (Rivera-Santos, 2010; Simo & Bies, 2007). Additionally, the chance that the MNE and NGO already have former partnership experience with these kind of partnerships is small, compared to partnerships between two firms or NGOs (Rivera-Santos &

Rufín, 2010). Therefore, it seems that it is heavily dependent on initial conditions in order to build interpartner-trust in these partnerships and additionally, for it to facilitate relational governance as an effective governance mechanism.

Partner communication and collaboration

Due to the fundamental differences between MNEs and NGOs, it can sometimes be hard to establish relational governance through building trust, in these partnerships (Berger et al., 2004;

Den Hond et al., 2015). Communication can play a facilitating role in supporting the informal governance mechanism of relational governance between parties (Arino et al., 2001; Austin, 2000; Rivera-Santos & Rufin, 2010). Kolk, Van Dolen, and Vock’s (2010) study on cross- sectoral partnerships found that “by means of frequent, open efficient communication between the partners trust can be created.” (Kolk et al., 2010, p. 127). Therefore indicating that frequent and especially open communication is important as an aspect of relational governance in these collaborations (Selsky & Parker, 2005). By examining the partnership between Unilever (consumer goods multinational) and Oxfam Novib (human rights organization), which is aimed at implementing CSR in Unilever’s value chain in Indonesia in regard to human rights and women empowerment, Rivera-Santos & Rufin (2010) conclude that “partners in Business-to- Nonprofit alliances will use any possible mechanism to signal commitment to the alliance and to reduce miscommunication problems within the alliance” (Rivera-Santos & Rufin, p. 11). As the relationship progressed and both sides improved communication through being more open

(15)

15 in terms of granting access to internal management documents, frequent face-to-face meetings and increased email-traffic between parties, a level of mutual trust was established that increased the effectiveness of the partnership in terms of governance (Oxfam GB, Novib Oxfam Netherlands, Unilever & Unilever Indonesia, 2005; Rivera-Santos & Rufin, 2010). As a result CSR was successful implemented in Unilever’s value chain in Indonesia and human rights and women empowerment improved (Oxfam GB et al., 2005).

Having a certain level of inter-partner trust is not only important for relational governance, but also for information-sharing (De Lange et al., 2016). If parties do not trust each other, communication will be limited and there will be careful selection of which information will be shared, reducing the effectiveness of the partnership (Berger et al., 2004). Information-sharing is key to govern the relationship, build trust and to keep partnership objectives aligned (Di Domenico, Tracey & Haugh, 2009; Jamali & Kenishian, 2008; Kolk et al., 2010). Austin (2000) found in their study about 15 business-NGO partnerships, that communication of knowledge, or information-sharing between partners was key to managing the relationship. By establishing information-sharing channels specifically meant for the partnership aimed at implementing CSR, it supported the alignment of objectives throughout both organizations, by engaging all employees. Therefore, a need for established communication channels in these cross-sectoral partnerships seems to exist, in order to facilitate information-sharing and the development of relational governance (Omar, Leach & March, 2014). Amnesty International for instance, a human rights organizations, setup a separate entity to communicate with the businesses it is cooperating with, called the Amnesty Business Group (ABG) (Ählström & Sjöström, 2005).

Separate entities that are setup specifically for information-sharing between nonprofits and businesses can greatly enhance information-sharing between partners, since there will be no interference with other activities of the business and the NGO, which are occupying the existing communication channels (Austin & Seitanidi, 2012). Accordingly, Seitanidi & Crane (2009), illustrated that a successful partnership between the Bank of Scotland and Prince’s Trust, a youth charity organization, was successfully maintained by the establishment of virtual teams and that trust between executives of both organizations was established. Executives remarked that they now had more confidence in the commitment of both parties to the objectives of the collaboration. Therefore, it seems that trust can be built through frequent partner communication and ultimately serve as governance mechanism in cross-sectoral partnerships.

We predicted that formal contracts are a requirement in MNE-NGO partnerships in order to effectively facilitate CSR-implementation in global value chains and to align the interests of

(16)

16 two organizations originating from fundamentally different sectors (Jonker & Nijhof, 2006;

Selsky & Parker, 2005). Could relational governance occupy a similar role in these partnerships and bridge or mitigate the differences between sectors and would that result in a reduced need for formal contracts?

2.3.3 Reputational hostages

A mutual hostage situation between two partners is usually characterized by mutual investment, which involves equity (Bierly & Coombs, 2004; Contractor & Woodley, 2009; Globerman &

Nielsen, 2007). Since the use of equity is not an option within MNE-NGO partnerships (Rivera- Santos & Rufín, 2010), a hostage situation can also be created through the use of reputation, a non-equity governance mechanism (Reuer & Devarakonda, 2016). Partners can create a so- called reputational hostage to ensure commitment and performance in the partnership (Dellacoras, 2006; Mayer, 2006), by promoting the partnership to the outside world (Rivera- Santos & Rufín, 2010). For instance, Danone’s (dairy multinational) partnership with Grameen (global human rights NGO) is a prime example of a MNE-NGO partnership that was characterized by a reputational hostage. Namely, the partnership was called Grameen Danone and became a separate entity, with both parties names in it and essentially, putting both reputations on the line (Yunus, Moingeon & Lehmann-Ortega, 2010). The idea behind this is that when the social enterprise of Grameen Danone fails, both parties will experience negative reputational effects and are therefore motivated to commit to the partnership and not engage in opportunistic behavior (Hahn & Gold, 2014). “Reputation encourages players not to defect and therefore reduces the threat of opportunistic behavior. As a consequence, a firm will be more confident with a reputed partner and, reciprocally, it will be trustworthy if it also has the reputation of being reliable” (Beuve & Saussier, 2008). Another example is how Unilever heavily promoted its cooperation with Oxfam Novib about improving the living standards of Indonesia, a country where Unilever has a large corporate footprint with its activities (Oxfam GB et al., 2005). Unilever promoted the partnership in the media, on their website and even released joint reports with Oxfam Novib on the process of the collaboration. These promotional activities implied significant costs for Unilever, but created a valuable reputational hostage in terms of raising stakeholders attention for the partnership between the two (Rivera-Santos &

Rufin, 2010). Through promoting the partnership, Unilever wanted to make sure that both partners committed to the partnership and that the objectives of CSR-implementation would be fulfilled. The goal was to improve human and labour rights in its supply chain, and through promoting the partnership, Unilever wanted to make sure that Oxfam would make a difference

(17)

17 in terms of implementation and improvement of these rights in Indonesia (Oxfam GB et al., 2005; Rivera-Santos & Rufín, 2010).

From the perspective of the NGO, relationship promoting is also a common activity (Ryall &

Sampson, 2003). NGOs can leverage the importance of corporate reputation, by associating itself with the firm and promote its relationship to the firm’s operations (Ryall & Sampson, 2003). The partnership between Greenpeace and Foron, a household appliances firm, is an example of a partnership that was heavily promoted by the NGO. Together, Greenpeace and Foron aimed to launch an environmental-friendly refrigerator in 2005 (Stafford, Polonsky &

Stafford, 2000; Stafford & Hartman, 2001). While Foron more so kept quiet about the partnership, Greenpeace heavily marketed the partnership through press conferences, to their own followers and to the general public (Stafford et al., 2000). Foron was also the only firm that Greenpeace approached that was willing to test the application of the environmental- friendly ‘hydocarbon’ technology, which was considered as highly flammable by the industry in general (Stafford & Hartman, 2001). Executives of Foron later described that they felt that Greenpeace promoted the project so intensively because Greenpeace valued the success of this project extensively, and wanted to prove to the rest of the industry that climate-friendly fridges were actually possible (Cohen; 2017). Therefore, Greenpeace wanted to make sure this climate- neutral fridge would be a success, in order to encourage the rest of the industry to implement CSR-standards in terms of more climate-friendly products. By taking the previous cases in consideration, it seems that as a result of both partners being extremely afraid to damaging their reputation and therefore lose resources such as funding (NGO) and sales (MNE), reputational hostages could be an additional governance mechanism in MNE-NGO partnerships.

2.3.4 Power dependence

The ideal type of partnership would be characterized by an equal power-balance in decision- making, in most partnerships however, equality in decision-making power is not realistic (Brinkerhoff & Brinkerhoff, 2011). According to Pfeffer and Salancik (1978), power imbalance and dependence in relationships can be explained by resource dependency theory (RDT). RDT has been used by many scholars to explain power imbalances and dependencies in collaborations in the private sector such as strategic alliances (Gulati, 1998; Saxton, 1997), joint ventures (Casciaro & Piskorski, 2005; Inkpen & Beamish, 1997) and mergers & acquisitions (Nienhüser, 2008). More recently, some scholars have started to apply RDT to cross-sectoral partnerships as well (Austin, 2000; Gras & Mendoza-Abarca, 2014; Suarez and Hwang, 2008).

The concepts of power imbalance and power dependency are particularly characteristic to the

(18)

18 structure of cross-sectoral partnerships (Lefroy & Tsarenko, 2013). As previously mentioned, one of the benefits of these partnerships for the NGO, is the access to resources through the MNE, to which the NGO usually does not have access. However, this also implies their reliance on the MNE for resources, placing the MNE in a favorable bargaining position (Berger et al., 2004; Selsky & Parker, 2005). But at the same time, the NGO’s social resources such as reputation, legitimacy and contacts with local communities, legislators and other stakeholders can be just as valuable to the MNE (Berger et al., 2004; Selsky & Parker, 2005). “Governing from the governance perspective is always an interactive process because no single actor, public or private, has the knowledge or resource capacity to tackle problems unilaterally” (Kooiman, 1993). Therefore, the dependence on each other’s resources is an important aspect in explaining how power dependency can serve as a governance mechanisms in these partnerships (Lefroy &

Tsarenko, 2013). In essence, power dependence also creates a hostage situation, but based on resources instead of reputation. It is assumable that the party that is least dependent on the other’s resources, has the most power in the partnership and therefore the most decision-making authority (Stoker, 1998). Therefore, decision-making will be faster and there will less bargaining processes, since one partner is dominant (Bacharach & Lawler, 1986; Casciaro &

Piskorski, 2005).

According to Stoker (1998), governance as in interactive process involves various forms of partnerships. Namely, principal-agent relations and inter-organizational negotiations. These forms of partnership can be used to illustrate how power dependency in terms of resources can serve as a governance mechanism in MNE-NGO partnerships. In principal-agent relations, one organization (the principal) contracts the other organization (the agent) to fulfil a specific task (Broadbent, Dietrich & Laughlin, 1996; Petersen, 1993). In terms of MNE-NGO partnerships aimed at implementing CSR in global value chains, this can refer to an MNE hiring an NGO to examine their supply chain in a foreign country, utilize their local knowledge on the particular CSR-issues and help the MNE to improve upon these issues. For instance in 2011, Unilever hired Oxfam to conduct a study on their supply chain in Vietnam in order to assess labour rights and how Unilever could improve the situation of the workers in that particular chain (Wilshaw, Do Quynh, Fowler & Pham Thu, 2016). Unilever provided full transparency into their Vietnam operations and supply chain and provided Oxfam with the resources necessary to conduct the study (Wilshaw et al., 2016). In exchange, Oxfam utilized their knowledge about labour rights and provided recommendations to Unilever on what they could change to improve (Maitland, A., & Ciencia, S. (2018). In these kind of principal-agent partnerships, most of the power is in

(19)

19 the hands of the MNE, since they are the contracting party and provide the NGO with the resources and access to the supply chains that it requires. Therefore, the MNE has most of the decision-making authority in the process of implementing CSR into their value chain.

In inter-organizational partnerships, partners combine their resources and mutually negotiate projects in order to reach their own organization’s objectives (Stoker, 1998). For example, Nestlé’s (food and drinks MNE) partnership with Verité (NGO specialized in fair labour) to address labour right issues in their seafood supply chain. In this partnership the objective of Nestlé was to eliminate any labour right violations from their seafood supply chain, as they received negative publicity in the media and from stakeholders, which led to decreases in share value; the objective of Verité was to improve the labour rights of workers in global supply chains in general (Agarwal, Gneiting & Mhlanga, 2017). Nestlé and Verité respectively , combined their resources in terms of expertise in seafood supply chains and expertise on labour right issues in order to mutually reach objectives (Lindley, Techera & Webster, 2019). In these kind of partnerships, the need of the MNE and NGO for specific resources determines which party has the most decision-making authority (Lefroy & Tsarenko, 2013). The dominant party in terms of governance of the partnership, will be the one that is least dependent on the other’s resources (Lefroy & Tsarenko, 2013).

These different forms of partnerships and relevant MNE-NGO collaborations that were outlined, shed light on the role that power dependence in terms of resources can play in the governance of these partnerships. If one partner is less dependent on the other in terms of resources, that partner will most likely be the dominant party in terms of decision-making and governance of the relationship. Therefore, power dependence could be an additional governance mechanism in MNE-NGO partnerships.

(20)

20

3. Methodology

3.1 Research design

The goal of this research is to describe how MNE-NGO partnerships aimed at implementing CSR in global value chains are governed. Due to the lack of literature on this topic and the exploratory nature of the subject, a case study design is most suitable (Robson, 2002).

Governance is a complicated mechanism in which context is particularly important (Börzel &

Risse, 2005; Glasbergen, Biermann & Mol, 2007). Therefore, case study research seems most appropriate to be able to describe this context and the governance process within these kind of partnerships (Morris & Wood, 1991). The case study design offers great potential to answer

‘how’ questions (Saunder et al., 2009; Yin, 2003), which is suitable to answer the question of how MNE-NGO partnerships are governed. In order to be able to compare different governance relationships and generalize upon the results (Saunders et al., 2009), we use a multiple case- design instead of a single case-design (Yin, 2003). A second dimension of case-study design, as described by Yin (2003) is holistic vs embedded, which refers to the unit of analysis. In this study an embedded case-design will be used, meaning that this study will describe multiple MNE-NGO partnerships, consisting of the same NGO with different MNEs. An embedded case-study design is chosen to also facilitate the comparison of different governance relationships, from the perspective of one focal organization, which assists in drawing generalized conclusions across cases (Saunders et al., 2009; Yin, 2003). A single (holistic) case on one MNE-NGO partnership would not be sufficient to draw conclusions about governance, a concept that is strongly dependent on context (Börzel & Risse, 2005; Glasbergen et al., 2007).

Therefore, a embedded multiple-case design is appropriate to generalize about the governance in these partnerships in the most effective way (Yin, 2003). The reason for the NGO being the focal organization and the ‘embedded case’ in this research is that NGOs are known for willingness to promote their cause and cooperate with stakeholders, even students (Powers, 2014). Therefore we suspect that NGOs won’t shy away from sharing governance information, which is information that MNEs are less willing to share (Kaymak & Bektas, 2017).

Additionally, NGOs have in comparison to MNEs significantly more partnerships and experience in the nature of the subject (Greenberg, J., Knight, G., & Westersund, 2011; Hansen

& Spitzeck, 2011), which makes the NGO a logical choice to be the focal organization in this research. Therefore, we aim to get access to one high-profile NGO that has partnerships with several multinationals and through that NGO, establish contact with the MNEs it has partnerships with. Establishing contact with executives of MNEs should be easier, through

(21)

21 engaging the already established connection between them and the NGO, which will serve as a focal point for the researcher.

3.2 Sampling strategy

The sampling criteria for selecting the focal NGO were: (1) CSR-involvement, specifically a focus on the development and implementation of human and labour right policies for the global value chains of MNEs. (2) NGO sector, specially a focus on human and labour rights enforcement. (3) Size, specifically a focus on large NGOs with offices in multiple countries and a large base of full-time and salaried staff, in order to be able to conduct projects with MNEs on a global scale. (4) Private sector involvement, specifically a focus on an open attitude to private sector collaboration and a minimum of two known partnerships with MNEs. (5) Country, specifically a focus on NGOs that are headquartered, or have an office in the Netherlands, which grants access to more means of contact in comparison to NGOs that are located across the border.

After an initial round of reaching out to NGOs that fulfilled these criteria, Oxfam Novib was selected to serve as the focal organization and the embedded case in this research. Oxfam Novib is a poverty, human and labour right organization that has a strong focus on developing and implementing appropriate CSR-policies for MNEs and their global value chains (Oxfam Novib, 2019). Oxfam Novib strongly promotes their private sector collaboration and is engaged in many high-caliber MNE-NGO partnerships with e.g. Unilever, Nestlé and Nutreco. They have a well-known campaign, Behind The Brands (BTB), which is based around cooperation with MNEs to improve CSR in their value chains, focusing specifically on human and labour right policies (Oxfam Novib, 2019). Therefore, Oxfam Novib’s expertise and experience with MNE- NGO partnerships aimed at developing and implementing labour right policies in the MNE’s value chain is of particular interest for this research. This makes Oxfam Novib a suitable candidate to be the embedded case, to be able to gain insight on their experience with multiple MNE-NGO governance relationships for comparison purposes. Additionally, Oxfam Novib operates offices in 34 countries with approximately 350 full-time and salaried workers and are headquartered in the Netherlands (Devex, 2019).

The initial participant in this case, was Maarten de Vuyst, private sector leader at Oxfam Novib.

This particular participant shared a lot of information about different partnerships with multinationals, including Oxfam Novib’s perspectives of governance within these partnerships.

Therefore, the initial participant helped to identify potential future participants of interest. The

(22)

22 participants were recommended based on the conditions of (1) The mutual development and implementation of human and labour rights in global value chains. (2) The degree of collaboration between Oxfam Novib and the MNE, so that the partnership would be suitable to describe governance mechanisms. And (3), the importance of the partnership to the NGO, which refers to the value Oxfam Novib derives from the partnership in terms of improving human and labour right policies and making a difference for the workers that are impacted by the MNE’s value chain. Because these conditions are hard to judge upon from an external researcher-perspective (Baxter & Jack, 2008), being able to gain insight to these perspectives through the embedded case was helpful in identifying the relevant MNE’s for this research.

Therefore, the embedded-case study design proved to particularly helpful in identifying the partnerships of interest (Baxter & Jack, 2008). Because the recruitment of participants was based on the recommendations of the initial participant, the corresponding sampling method of this research is snowball sampling (Noy, 2008).

3.3 Data collection

Access to the MNEs identified in the initial interview was acquired through contacting the manager or senior executive in charge of CSR by email or telephone. First a formal email containing the scope of the research project and the interview was sent, explaining why it would be valuable to incorporate the perspective of the MNE in the governance of these partnerships aimed at implementing CSR in global value chains, alongside the perspective of the NGO (in this case Oxfam Novib). The initial participant, the private sector leader of Oxfam Novib Netherlands, gave permission to incorporate his details and support in these emails as well, in order to provide more leverage to get into contact with the relevant executives. Additionally, contact was established with the ethical trade manager of Oxfam Novib UK, to strengthen the perspectives on the partnerships with Unilever in Vietnam and Indonesia, in which she was personally involved in the decision-making process. Multiple perspectives within the same organization can provide a more complete picture about the phenomena being observed (Dooley, 2002; Houghton, Casey, Shaw & Murphy, 2013). Sanorice, the largest private label rice cake producer in the world was also willing to cooperate and share information about their partnership with Oxfam Novib.

Additionally, contact was established with Nestlé. However, the executive that was reached (Creating Shared Value Specialist) had not been involved directly or indirectly in a partnership with Oxfam Novib and also shared that this partnership had been of short existence. Therefore, from the perspective of this executive, this was not considered as an MNE-NGO partnership

(23)

23 aimed at implementing CSR in global value chains and could not be used to extent the embedded case (Baxter & Jack, 2008). However, the partnership Nestlé was currently involved in fulfilled all the sampling conditions that were previously stated for the NGO and MNE of interest. Namely, Nestlé maintains a partnership with the Fair Labour Association (FLA) in order to address labour right violations in their cacao value chain and to implement policies to prevent these violations from happening. Even though this partnership did not involve Oxfam Novib, it meets the scope of this research and involves a cross-sectoral partnership involving an MNE and NGO that cooperate to implement CSR into the MNE’s global value chain.

Therefore, this case could be used to enrich the embedded case, by comparing one additional partnership containing a different NGO to be able to determine whether generalized conclusions also hold up in this partnership (Saunders et al., 2009). Considering this a study about governance in MNE-NGO partnerships, it would be valuable to be able to draw conclusions not limited to the embedded case, which is Oxfam Novib. This approach is supported by Bansal, Smith & Vaara (2018), who’s study on new ways of seeing through qualitative research states that embedded case studies are limited to the dynamics of one focal organization and that providing additional external cases can enrich the data. Therefore, this study takes this opportunity to explore MNE-NGO partnership governance by extending the embedded base with one additional comparative case between Nestlé and the FLA, in order to strengthen generalized conclusion about governance mechanisms (Bansal et al., 2018; Greckhamer, Furnari, Fiss & Aguilera, 2018). In order to gather the relevant data on both perspectives in the additional Nestlé case, the FLA was also contacted.

In total, 5 executives were interviewed in regard to 4 partnerships. As only exceptions, executives of Nestlé and the FLA answered questions about their partnerships by email, because of schedule constraints. These emailed questions and answers can be found in appendix 5 & 6, and were similar to the questions in the interview guide. The key topics that comprised the interview guide were: the implementation of CSR in global value chains, the importance of these projects for both the MNE and NGO, the governance dynamics in these partnerships, formal contracts, relational governance, reputational hostages and power dependence as governance mechanism. To ensure that the perspectives of the executives of both the MNE and NGO on the governance of these partnerships can be fully captured, open interview questions were used (Ritchie, Lewis, Nicholls & Ormston, 2013). All though the interview guide was followed to keep a strong focus in the interview and to be able to collect data relevant to the propositions of this study, this did not restrict from using probing questions (Ritchie et al., 2013)

(24)

24 to gather more detailed information from the executives interviewed. Therefore, the interviews were semi-structured but allowed for flexibility (Saunders & Lewis, 2012). The interview guide as well as the interview transcripts can be found in appendix 1 and 2-6 respectively.

Interview Organization Sector Function of

participant Interview

purpose Interview

setting Interview

duration Date

1 Oxfam Novib

NL Poverty,

human and labour rights

Private sector leader

Governance of partnerships with Unilever, Nestlé &

Sanorice

Skype-call 80

minutes 05-04- 2019

2 Sanorice Rice, corn

and multigrain -cakes

Corporate QA and R&D manager

Governance of partnership with Oxfam Novib

Skype-call 63

minutes 18-04- 2019

3 Oxfam Novib

GB Poverty,

human and labour rights

Ethical trade manager

Governance of partnerships with Unilever

Skype-call 55

minutes 16-05- 2019

4 Nestlé Food and

beverages Creating shared value specialist

Governance of partnership with the FLA

Questionn

aire - 28-05-

2019

5 FLA Labour

rights Associate director of the agriculture program

Governance of partnership with Nestlé

Questionn

aire - 2019

Table 1. Interview details

The details of each interview can be found in Table 1. All of the executives interviewed, were either directly in charge of the implementation of CSR in the value chain of the MNE or indirectly involved in the decision-making process. All of these partnerships were focused on the mutual development and implementation of human and labour rights policies in the value chain of the MNE, which resulted in a narrow research focus. All of these partnerships are of the same nature and therefore, generalized conclusions could be drawn. Data was collected on the following partnerships, for which the details are provided in table 2.

(25)

25

Partnership Purpose CSR-implementation in the MNE’S value chain

Oxfam Novib – Unilever (Vietnam)

Partnership to address human and labour rights violations and implement better policies into Unilever’s Vietnamese value chain, by building on the UN guiding principles on Business and Human Rights (UNGDPs).

Vietnam is one of the largest emerging economies in the world, but the costs of living have outgrown the economic growth, causing the necessity for higher minimum wages, living wages and better labour rights (Wilshaw, Unger, Quynh & Thu, 2013).

The partnership consists out of different stages: (1) Oxfam Novib conducted an initial study on Unilever’s supply chain and made recommendations on how it could be improved (2011- 2013). (2) Follow-up study in order to determine whether the implemented policies have had their desirable effect and what could still be improved (2015-2016).

(1) Mutually developed a Responsible Sourcing Policy (RSP), which improved human and labour rights across

Unilever’s value chain.

(2) Implemented a change in management style in the Vietnamese factories which improved mechanisms for dialogue and raising grievances about human and labour rights.

(3) Implemented improved worker policies in Unilever’s factories that facilitated an increase in direct employment, wages and benefits. Development of KPIs used to monitor human and labour rights.

Oxfam Novib – Unilever (Indonesia)

Partnership to address poverty, human and labour rights of the workers in Unilever’s Indonesian value chain. Indonesia experienced high levels of poverty, despite its abundance in natural and human resources, more than half of the population was living on less than 2$ a day (Oxfam GB et al., 2005).

Combining their knowledge in business, public policy and sustainable development, Unilever and Oxfam Novib cooperate to find the ideal balance between wealth creation and poverty reduction.

Jointly conducted a poverty footprint study focused on: (1) Mapping the impact of Unilever Indonesia at the macro-

economic level. (2) Exploring Unilever Indonesia’s policies and practices on human and labour rights. (3) Analyzing supply chain relationships and issues.

(1) The mutual development of value chain policies that facilitate sustainable poverty reduction.

(2) The implementation of policies that improved the working terms and conditions in Unilever’s Fast Moving Consumer Good (FMCG) value chains. And at the same time, turned them into opportunities, by creating a learning environment that facilitates their workers in developing skills that increase their independence, value and ultimately, their living standards.

(3) Implemented a contract-review process, which enables Unilever to analyze all the contractors in their value chain and check whether they are meeting their commitments on labour rights and policies and particularly on gender issues.

(26)

26

Partnership Purpose CSR-implementation in the MNE’S value chain

Oxfam Novib – Sanorice

Partnership to transform the Cambodian rice farmers in

Sanorice’s value chain from traditional agriculture to biological farming, in order to improve their living standards, while addressing human and labour rights at the same time.

Living standards in Cambodia are generally low and poverty is high. Sanorice has a large impact on the living standards of the population there, since rice is one of Cambodia’s main exports and Sanorice sources most of its rice there.

(1) Mutually developed and implemented human and labour rights policies that improved gender equality and minimum wages.

(2) Mutually developed policies that facilitated the establishment of worker unions.

(3) Improved the living standards of the Cambodian rice farmers in Sanorice’s value chain by transforming them from traditional to biological agriculture, for which they received increased revenues and benefits.

Fair Labor Association - Nestlé

Partnership to monitor and improve Nestlé’s cacao value chain in Ivory Coast on code awareness, eliminating child and forced labour and, to develop better health and safety policies for workers that are impacted by Nestlé’s operations.

This is an annual process, the FLA monitors Nestlé cacao chain every year and provides recommendations based on the current issues that could be improved upon.

(1) Mutually developed better codes of conduct, which have decreased child and forced labour in Nestlé’s value chain to almost none.

(2) Developed a Child Labour Monitoring and Remediation System (CLMRS) to effectively detect when child labour is used in Nestlé’s value chain.

Table 2. MNE-NGO partnerships in data collection

(27)

27 3.4 Data analysis

Thematic analysis (Braun & Clarke, 2006; Gioia, Corley & Hamilton, 2013) was used to analyze and evaluate the interviews. Thematic analysis is a qualitative method of analysis for uncovering themes, levels of patterned responses and meanings (Braun & Clarke, 2006).

Thematic analysis was chosen for this study because it can be used to systematically identify and organize similar themes and meanings across multiple interviews (Braun, Clarke, Hayfield

& Terry, 2019). Additionally, since thematic analysis can be applied to most qualitative research and is a neutral way of analysis (Bryman & Bell, 2011), as well as being an accessible and flexible tool (Braun et al., 2019), thematic analysis was chosen to analyze the collected data. In the initial phase all interviews were listened to/read and transcribed into text. During this process, the most interesting statements/findings were already written down and kept in mind. In the second phase, the coding process started. Coding is organizing and sorting parts of the interviews into meaningful groups (Tuckett, 2005). These codes were linked to words that were of special interest in our study and that frequently occurred. The purpose of this process is to shed light on the features of the data that are relevant for our research, for answering the research question and proving information for our propositions (Boyatzis, 1998). This kind of coding is called descriptive coding, which is used to summarize the central theme of the data (Holton, 2007). For instance, all interview sections that covered words such as ‘agreement’,

‘memorandum of understanding’ and ‘contract’ were used to summarize the concept of interest:

formal contracts as governance mechanism in MNE-NGO partnerships. To keep the analysis focused and make sure there is was not too much overlap between codes, a final analysis to reduce overlap and redundancy was conducted in order to ensure validity and reliability (Golafshani, 2003). After the coding process, the different codes were placed into possible themes by following Gioia et al. (2013) thematic analysis approach. Codes were bundles based on the analysis of similarities and differences and afterwards, first-order themes were compiled (Gioia et al., 2013). Braun & Clarke (2006) advice that visual presentations can be helpful to sort the codes into themes, so therefore we designed table 3 that consisted of the overarching themes in our interviews. These themes are; Formal contracts and relational governance as governance mechanism, reputational hostages as governance mechanism and power dependence as governance mechanism. Under these themes, second-order themes were then compiled (table 3) such as the ‘value of reputation’ under ‘reputational hostages’ in order to extent the understanding of the underlying concepts (Gioia et al., 2013). In the last phase of the analysis, full understanding was gained of the data and the relevant themes could be used for discussion

2. Literature review

Abstract

Table of contents

1. Introduction

2. Literature review

3. Methodology