The Ineffectiveness of CSR: Understanding Garment Company Commitments to Living Wages in Global Supply Chains

ABSTRACT

In the face of pressure from civil society, unions and consumers to improve labour standards for the workers producing their goods, companies at the helm of global garment supply chains have made commitments to pay living wages within their supply chains. Harnessing insights from the critical political economy literature on corporate social responsibility (CSR), we investigate the actions of garment companies to meet these commitments. We do so through analysis of original data from a survey of 20 leading garment companies, which we co-developed in 2018–2019, as well as publicly available information for garment companies and relevant multi-stakeholder initiatives. Based on this data, we argue there is very little evidence to suggest companies have made meaningful progress towards achieving commitments to pay living wages in their supply chains, challenging widespread assumptions about CSR’s benefits to workers. We argue that in the face of mounting evidence of CSR ineffectiveness, including our own, there is a need for new political economy research into the benefits that companies derive from CSR commitments that deflect attention from their core business models and the uneven value distribution within global supply chains.

KEYWORDS:

• Corporate social responsibility
• global supply chains
• labour standards
• living wage
• business
• human rights

1. Introduction

Multi-national garment corporations have long faced criticism over labour rights violations in their supply chains. In recent decades, endemic worker rights violations have been documented including wage theft and violations (Anner et al. 2013, Anner 2019), gender-based violence (Evans 2017, Selwyn et al. 2019), unsafe working conditions and dangerous levels of productivity (Merk 2011b, Mezzadri 2017), and infringements on freedom of association (Egels-Zandén and Merk 2014, Anner 2017). For over two decades, civil society, unions, workers, and policymakers have exerted sustained pressure on the multi-national corporations (MNCs) leading fashion supply chains to address these infringements improve labour standards (cf. Saxena 2019).

One commitment companies have made in response to such pressures is to pay living wages within their supply chains. Living wages are important because they provide sufficient income to cover essentials for workers (e.g. food, housing, medical insurance), and because they can be a key protective factor in preventing more serious forms of labour abuse such as forced labour (LeBaron et al. 2018, LeBaron 2020). They can also enable workers in low-waged work to upgrade their position within supply chains (Phillips 2011, Phillips 2013). As such, corporate commitments to deliver living wages across global supply chains carry potentially important economic impacts for workers and their families, particularly in countries where production for export is a major component of industrial strategy. In a context in which MNCs are taking on functions and powers traditionally exercised by states, and given the links between global supply chains and inequality, MNC commitments to pay living wages merit investigation.

This is especially true in the garment industry, where nearly four decades of research document sub-standard wages (Weil and Mallo 2007, Chan and Siu 2010); furthermore, the industry employs an estimated 60m people worldwide (with 80% of the workforce estimated to be women) and is heavily concentrated in low and lower-middle income countries (ILO 2019), so living wages would reverberate through the economy. Several garment MNCs have made living wage commitments over the last decade. For instance, in 2013, H&M set public goals for paying a ‘fair living wage’ in its supply chain (DePillis 2013), while PVH (owner of brands including Calvin Klein and Tommy Hilfiger) stated ‘a goal of paying all workers no less than a living wage’ (PVH 2020).

To advance understandings of the potential for voluntary corporate commitments to living wages to address social problems in global supply chains, in this article, we investigate garment company MNCs commitments to pay living wages, actions taken to realise these, and whether these commitments are translating into living wages for workers on-the-ground. Empirically, we present original data from a survey of 20 leading garment companies (see Table 1), which we co-developed with the Clean Clothes Campaign (CCC), a global network of organisations that campaigns for improved rights for garment workers (see ‘Research Approach’). Additionally, we draw on publicly available information for garment companies and relevant multi-stakeholder initiatives (MSIs) operating in the garment industry focused on delivering higher wages.

Table 1. Responses to the CCC survey.

Completed the survey (14)	Did not complete the survey but provided some relevant information on the survey themes (5)	Did not complete the survey and did not provide any information (1)
Adidas	Amazon	Hugo Boss
C&A	Fruit of the Loom
Decathlon	GAP
Fast Retailing / Uniqlo	Levi Strauss
G-Star RAW	Zalando
Gucci
H&M
Inditex
Nike
Primark
Puma
PVH
Tchibo
Under Armour

Based on this data, we argue there is very little evidence to suggest companies have made meaningful progress towards achieving commitments to pay living wages in their supply chains. Amongst other problems, our data shows that companies have outsourced their living wage commitments to MSIs while failing to modify commercial practices that give rise to low wages in their supply chains. Corporate social responsibility (CSR) efforts are also diluting and distorting the definition of ‘living wage’ thus making it challenging for consumers and policy-makers to determine whether living wages are truly being paid.

Our study contributes to theory about the role of CSR, and the business interests served by it. We build on existing political economy theory about the limits of social compliance regime— which revolves around CSR tools like social auditing, ethical certification, and codes of conduct— in bolstering labour standards (cf. Soederberg 2007, Taylor 2011, Esbenshade 2012, Fransen 2012, Egels-Zandén and Lindholm 2015, LeBaron 2020) and we challenge theories that view CSR as a neutral and pragmatic response to governance gaps created amidst globalisation as states have either refused or been unable to institute and enforce labour standards, or as a slowly-working, incrementally effective form of governance (cf. Vogel 2005; for political economy critiques see Cutler et al. 1999; Dauvergne and LeBaron 2014, Lund-Thomsen 2020). We highlight the limits of dominant approaches to studying CSR that prioritise procedural questions such as about initiative design, stringency, and interactions rather than performance or outcomes (cf. Henson and Humphrey 2010, Klassen and Vereecke 2012, Ibanez and Blackman 2016, Glinksi 2017, Graafland and Smid 2019, Marshall et al. 2019) over investigations of CSR effectiveness, both on-the-ground and in terms of evolving company policies and practices.

We seek to move the literature forward by analysing the dark sides of CSR. We add further weight to claims that corporations derive legitimacy from CSR commitments that deflect attention from exploitative business models (Hanlon and Fleming 2009, Clarke and Boersma 2017) and that MSIs can obfuscate corporate inactivity towards meeting CSR commitments (Dauvergne and LeBaron 2014) by showing that MNCs strategically use CSR to fend off criticism of their business models and supply chain dynamics while refusing to redistribute value down the supply chain in the form of higher wages for workers. We challenge the conventional assumption that ‘something is better than nothing’ when it comes to CSR and stress the dangers that these governance dynamics pose to low-waged workers who face significant barriers to exercising their rights. In doing so, we seek to advance political economy scholarship on the negative role of CSR including its use to facilitate and consolidate the power of large corporations (Banerjee 2008), undermine sovereignty of countries in the global South (Clapp 1998, Fridell 2007, De Neve 2009), and construct superficial forms of governance that preserve status quo business models while failing to promote real change for workers (LeBaron 2020).

The article is organised as follows. We first outline our research approach. We then describe how the interdisciplinary literature on CSR has tended to understand and investigate social compliance efforts, including MSIs, and how we move theory about CSR’s role within supply chain governance forward. We proceed to describe our empirical analysis and findings. The article ends with a summative conclusion.

2. Research approach

2.1. Definition of living wage

There are competing and contested definitions for what precisely constitutes a living wage. Most converge around the idea that a living wage should cover ‘basic’ needs of both a worker and their family beyond ‘mere subsistence’ (Parker et al. 2016) in a way that is suitable for their geographical situation (Glasmeier 2004, Anker and Anker 2017). This is enshrined in wage standards set by international organisations. For instance, the International Labour Organization (ILO) Constitution preamble notes the need for ‘an adequate living wage’ (ILO 1919), while its Minimum Wage Convention 131 refers to ‘the needs of workers and their families’ and a ‘satisfactory standard of living’ (ILO 1970a) and its Minimum Wage Fixing Recommendation 135 specifies this must cover ‘more than subsistence needs’ (ILO 1970b). The United Nations (UN) Declaration on Human Rights states that ‘Everyone who works has the right to just and favourable remuneration ensuring for himself and his family an existence worthy of human dignity’ (UN 1948). Meanwhile, the OECD calls on MNCs to pay a wage that satisfies ‘basic needs of the workers and their families’ (OECD 2011).

Key concepts relevant to living wages are interpreted differently. For instance, while ‘basic needs’ is usually taken to comprise food, shelter and clothing, definitions used by NGOs vary with respect to whether these needs include education, healthcare, childcare and transportation (Anker 2011, p. 21). The flexibility of component definitions reflects the ‘essential relativism’ and contextual specificity of the living wage concept and related living standards (Parker et al. 2016, p. 1, Anker 2011), since norms around living wages also differ across geography and sector. For instance, while definitions centred on living wages in the global North often account for taxation in calculating the necessary ‘take home’ wage (Glasmeier 2004, Luce 2017), definitions for the global South often include a requirement that a living wage is earned in a standard workweek (48 h) (Anker 2011, p. 23). Additionally, NGO definitions focused on global South countries frequently include the need for discretionary income or savings (Anker 2011, p. 21), which is vital in contexts where workers without savings are vulnerable to exploitative labour practices (LeBaron 2021). International organisation standards around living wages fall short in establishing and harmonising definitions for key concepts like a ‘standard’ workweek (beyond such general assertions as having a right to ‘reasonable limitation of working hours’ (UN 1948)) and appropriate levels for savings/discretionary income.

In this study, we use the CCC definition of a living wage, which defines a living wage as one ‘earned in a standard working week (no more than 48 h) [that allows] a garment worker to be able to buy food for herself and her family, pay the rent, pay for healthcare, clothing, transportation and education and have a small amount of savings for when something unexpected happens’ (CCC 2014).¹ The CCC’s definition is suitable for our study given that it: a) clearly demarcates what is included with respect to ‘basic needs’; b) is appropriate for the garment industry, in which low-wages are widespread, extensive and forced overtime is common, and workers’ lack of savings can render them vulnerable to forced labour and overlapping forms of exploitation (Anner 2017, 2019, 2020, Mezzadri 2017, LeBaron 2020).

In adopting this definition, we intervene in the academic literature as follows: the paper focuses on how living wages are implemented, moving beyond conceptual discussion regarding what constitutes a living wage (Anker 2011) and whether living wages should be implemented (Adams 2017, Brennan 2019). As such this paper aligns with scholarship on the current use by companies of living wages in their business strategy and their attempts at integrating the payment of a living wage in their practices (Egels-Zandén 2017, pp. 100–2) and makes a contribution regarding the question of how (or not) living wages practices can be integrated in the fight against labour exploitation (Miller and Williams 2009, Merk 2011a, Marshall 2019).

2.2. Data collection and analysis

We undertook this research to investigate how and to what extent MNCs had realised their commitments to living wages. We selected the garment industry in light of sustained activist pressure on MNCs to address low-waged work, which we thought might translate into progress towards living wages compared to other sectors. We were aware that CCC planned to undertake research around living wages and developed a partnership to co-design a survey of garment companies that would give us traction to evaluate MNC progress towards living wages in the industry. As well, we analysed publicly available company documentation relevant to living wages. By combining survey data and public documentation, we can explore: a) how companies present their living wage commitments; b) how company living wage commitments and living wage MSIs interact with business models; c) the actions that companies claim they have taken to realise their living wage commitments; and d) the outcomes associated with these actions.

There is very limited public information about the outcomes, impacts, and overall effectiveness of MNC efforts to deliver living wages. Moreover, MNCs are challenging to research, particularly in relation to labour standards. To understand the progress MNCs have made towards living wages, we sought to gather information from them directly. Working with CCC, an established civil society organisation with established relationships with garment MNCs around the world, facilitated efficient access and communication with company representatives. CCC disseminated our survey to a sample of 20 MNCs in December 2018 as part of their 2018/19 Wages Survey (Table 1). The sample was selected by CCC to represent global garment MNCs with significant influence across sourcing and retail markets, and ensuring variation across key garment sectors: sportswear, fast fashion, luxury fashion, and online-only retail.

14 of the 20 companies provided detailed answers to all or most survey questions, with some providing additional supporting documentation and evidence. Six companies did not provide direct answers to the questions; however, all but one provided relevant information via email including links to publicly available company documents around living wages. CCC informed companies about the nature of the academic research collaboration and published their own report of the survey data (CCC 2019).

Data analysis involved multiple steps. First, we read all responses to identify areas of convergence and difference across companies. This informed the creation of ‘coding frame A’ comprised of six indicators to enable direct comparison: 1) MSI participation; 2) living wage definitions; 3) living wage strategies and benchmarks; 4) monitoring and enforcement of living wages; 5) transparency and 6) freedom of association protections (which ensure workers can report wage violations and underpayment).

Using the same sample of 20 companies, we next compiled data on each company’s policies, business practices, and activities relevant to living wage payment using public information, including supplier codes of conduct, sustainability reports, modern slavery reports, and relevant webpages. The purpose was to understand and compare the degrees of alignment between living wage commitments and overall company practice, such as whether supplier codes of conduct included wage requirements. We also wanted to source missing information about company living wage commitments where survey questions remained unanswered. We then coded the survey responses and publicly available information from all 20 companies based on the above indicators, and cross-checked survey responses against the information within our compiled dataset.

A key strategy reported by companies to realise their living wage commitments is participation in MSIs; 18 of the 20 companies are associated with more than one MSI (see Table 2). To understand their effectiveness in delivering living wages, we researched MSIs operating in the garment industry that seek to increase wages for workers using publicly available online materials. This primarily consisted of MSI websites and reports, with some secondary academic and news sources. Following Fransen (2012), we use the term MSI broadly to cover initiatives with the following characteristics: both for-profit and non-profit stakeholders (including unions, civil society groups, states and global governance organisations) are involved; the involvement of for-profit stakeholders is voluntary, and the initiative aims to upgrade and enforce labour standards within the supply chains of the for-profit stakeholders involved.

Table 2. External initiatives that companies are members of or associate themselves with.

		Action, Collaboration, Transformation (ACT)	German Textile Initiative and Dutch Textile Covenant	Fair Labor Association	Fair Wage Network’s Fair Wage Method	Ethical Trading Initiative (ETI)	ILO Better Work	IndustriALL Global Framework Agreements	Sustainable Apparel Coalition	Social Accountability International	Social and Labour Convergence Project	Total
Survey respondents	Adidas		x	x	x		x		x		x	6
	C&A	x	x		x				x			4
	Decathlon									x		1
	G-Star RAW^a		x		x		x		x		x	5
	Gucci				x				x	x	x	4
	H&M	x	x		x	x	x	x	x		x	8
	Inditex	x				x	x	x	x		x	6
	Nike			x			x		x		x	4
	Primark	x	x			x	x		x			5
	Puma		x	x	x		x		x		x	6
	PVH	x		x			x		x	x	x	6
	Tchibo	x	x			x		x		x		5
	Under Armour			x					x			2
	Uniqlo			x			x		x		x	4
Non-respondents	Amazon											0
	Fruit of the Loom			x			x					2
	GAP					x	x		x		x	4
	Hugo Boss		x	x								2
	Levi Strauss						x		x		x	3
	Zalando	x					x		x		x	4

^a G-Star RAW became a member of ACT in September 2019 but at the time of the company completing the CCC survey and during our period of analysis they were not a member.

After collecting information about MSIs we created ‘coding frame B’ to provide a framework for assessing the rigour and stringency of the initiative, including: a) whether the MSI employs robust living wage benchmarks; b) whether the MSI’s enforcement strategy involves independent assessments of suppliers; c) the extent of meaningful accountability and enforcement; and d) whether structural features of the business model that affect the payment of living wages are addressed. We compared this to information given in survey responses and in public information from the companies to analyse the role of MSIs in delivering living wage commitments.

The findings we present are drawn from our analysis of the information provided in company survey responses and from public information from companies and MSIs. Cross-interrogating this body of data allows us to explore degrees of alignment between the data sources, to consider company commitments to living wages in their own words and to evaluate the steps taken to implement and enforce living wage commitments.

3. Theorising the effectiveness of CSR

Critical political economy scholars have stressed the dangers and dark sides of CSR for workers, consumers, and the planet (Bartley 2005, Burchell and Cook 2013, Lund-Thomsen 2013, Dauvergne and LeBaron 2014, Jung and Kim 2016). When it comes to labour standards, this body of research stresses that CSR initiatives: outsource governance beyond the state and can facilitate the erosion of state-based regulation (Mayer et al. 2016, LeBaron and Rühmkorf 2017), consistently fail to deliver promised labour standards on the ground (Fridell 2007, Esbenshade 2012, Lund-Thomsen et al. 2012, Bartley 2018), weaken unions and worker-led campaigns for labour rights (Esbenshade 2004, Anner 2012, Appelbaum and Lichtenstein 2016), and create an illusion of governance where significant regulatory voids exist. More broadly, critical political economists have argued that CSR: ‘normalizes and thereby recreates the dominant neoliberal-led development paradigm’ (Soederberg 2007); comprises a form of ‘stakeholder colonialism’ which elevates capitalist and colonial interests above those of vulnerable stakeholders (Banerjee 2008, 2018); constitutes economic and cultural imperialism in the buyer-supplier relation (Khan and Lund-Thomsen 2011); and perpetrates ‘slow violence’ against local environments and dispossessed populations (Gamu and Dauvergne 2018). Critical veins of political economy scholarship see CSR as reflecting industry interest, rather than those of workers or the public (Cutler et al. 1999).

Yet, political economy scholarship extolling the normative dangers of CSR has done little to alter the optimism of scholars who see CSR as a promising means of ensuring and raising standards in global supply chains (cf. Gifford et al. 2010). A large interdisciplinary literature split across management and business studies, law, political science and other disciplines focuses on CSR’s positive potential, seeing corporate engagement with stakeholders as a key site for ‘deliberative democracy’ (Scherer and Palazzo 2007) and a means of reorienting corporations beyond their shareholders and narrow financial interests (Schoeneborn et al. 2020). This literature tends to conceptualise CSR initiatives as a positive or neutral force within global governance, explaining any lacking progress as relating to procedure, design, or technical considerations that can be rectified (Short et al. 2020).

Debates about CSR’s role in the global economy would benefit from further evidence about its on-the-ground effectiveness, and in particular, analysis of how far and in what ways CSR aims and standards are realised for workers within supply chains. We see the question of whether CSR programmes will be effective as empirical, not one to be determined solely through theory. While critical political economy scholarship often assumes a priori that CSR will be ineffective given that it reflects corporate interests, other veins of literature reflect a prevailing yet unproven optimism about its role and value in supply chains. Further empirical investigation of CSR effectiveness can deepen our understanding. We undertake such an investigation here in relation to a relatively new and understudied feature of CSR related to living wages, including company commitments, codes of conduct, and MSIs.

In this study, by effectiveness we are interested in whether CSR commitments and initiatives achieve their stated aims and outcomes; in our case, whether CSR has led to living wages for garment workers, or at least, meaningful progress towards them. Given widespread adoption of CSR as a means of governing global production standards, it is important to investigate whether, to what extent, and under what conditions company CSR commitments are being met on-the-ground.

After all, studies that have investigated CSR effectiveness in relation to labour standards have unearthed considerable problems finding that initiatives: fail to implement standards around wages (LeBaron 2020); mislead consumers and policymakers about the extent to which agricultural production relies on forced labour (LeBaron and Gore 2019); do little to uphold workers’ freedom of association rights (Barrientos and Smith 2007, Anner 2012, Egels-Zandén and Merk 2014); and generate falsified or unreliable information about labour standards in supply chains (Judd and Kuruvilla 2020). In relation to MSIs, scholars have found the effectiveness of MSIs on improving labour standards is constrained by power asymmetries between stakeholders and workers, and MSIs carry potentially harmful effects of crowding out unions (O’Rourke 2006); ‘are seldom able to identify process rights violations’ (Egels-Zanden and Lindholm 2015, p. 38); and contain trade-offs that shift initiatives ‘away from more radical … social goals … towards the need to create business value’ (Riisgaard, Lund-Thomsen and Coe 2020, p. 231). Informed by these insights, we examine the effectiveness of CSR living wage initiatives.

At the same time, we advance debates about CSR (in)effectiveness by asking what else CSR might be accomplishing through its ineffectiveness, where it is indeed found to be falling short. Are companies weaponizing CSR commitments to deflect criticism? Are CSR commitments misleading consumers and policymakers and leading to inaccurate understandings of labour practices within global supply chains? How can we understand the relationship between CSR commitments and business models?

As a first step towards exploring these questions, in the remainder of this article, we present our evidence about CSR effectiveness related to living wages, noting that companies are failing to deliver meaningful progress. This evidence, we contend, adds further weight to claims about CSR ineffectiveness. But, we argue, the problem isn’t a straightforward one of CSR simply not working. Rather, there is a need to explore why corporations are making CSR commitments they cannot achieve in their current form, and the impact this is having on vulnerable workers in global supply chains.

4. Findings

In this section we set out key findings from our analysis of the survey answers given by companies, information presented in public company documents, and public information from MSIs. We find that: 1) Companies are making only very modest changes to their relationships with suppliers and workers and are insufficiently addressing the features of their business models that lead to low wages and pose obstacles to living wages in supply chains; 2) Instead, companies have largely outsourced their living wage commitments to MSIs which lack robust enforcement mechanisms and usually do not address purchasing practices; 3) Companies have diluted and distorted the definition of ‘living wage’ thus enabling suppliers to continue to pay non-living wages; 4) Companies have ultimately made limited progress towards achieving living wages. While we illustrate each finding in turn, they are not discrete but intrinsically connected and mutually reinforcing.

4.1. Masking exploitative business models

Previous studies demonstrate that labour exploitation in the garment industry is rooted in prevailing business models (cf. Anner 2019, LeBaron 2020), and especially in sourcing practices that concentrate value and profit at the top of supply chains (Selwyn et al. 2019). These studies highlight the need for fundamental overhaul of garment company business models to facilitate the necessary value distribution along the supply chain to achieve living wages, supported through measures to benchmark wages and reform sourcing practices.

Indeed, previous research shows how global garment companies exert significant pressure upon a wide base of suppliers through their purchasing practices (Hearson 2009, Raworth and Kidder 2009, Barrientos et al. 2011). This pressure, characterised by Anner (2020) as a ‘pricing squeeze’ and a ‘sourcing squeeze’, can force suppliers to produce high volumes of goods for the low prices offered by lead firms, which often fall below the costs of production, making it impossible to meet wage standards. Facing cost pressure from the top of the supply chain, suppliers cut corners on wages and use cheaper subcontractors to keep costs low to secure and maintain orders. Research also demonstrates that long-term relationships between buyers and suppliers increases compliance with labour standards (Locke et al. 2009, Distelhorst and Locke 2018).

To explore whether and to what extent companies have adapted purchasing practices to achieve living wage commitments, our survey asked companies if they make long-term sourcing commitments at specific volumes to suppliers; whether they seek to limit the use of subcontractors in the supply chain and consolidate their supplier base, and if they seek to limit the use of labour contractors.

We found very limited evidence that living wage commitments have spurred change in the low-cost and high-volume sourcing model through which companies profit. Indeed, purchasing practices often exist in tension with companies’ living wage commitments and can undermine them. Companies were asked whether ‘freight on board’ prices paid per item are sufficient to allow for compliance with the company’s wage commitments through ring-fencing labour costs in negotiations with suppliers. While nine companies stated they do ring-fence labour costs when determining prices per item, only three currently use these cost calculations in purchasing.² Of these three companies, none employs a living wage benchmark which would ensure that all key components of a living wage are covered. It thus appears that living wage provision has not yet been built into pricing negotiations, and therefore, that CSR commitments to living wages have not translated into meaningful improvements in suppliers’ ability to pay living wages.³

4.2. The ‘outsourcing’ of living wage commitments to MSIs

Whilst we find little evidence to suggest that the companies in our sample are modifying their purchasing practices to realise living wage commitments, we find that they strongly seek to demonstrate their commitment to living wages through their involvement with multiple MSIs (as Table 2 demonstrates). Across our sample – and across all company types – the survey answers of companies and company documents place considerable emphasis upon involvement with MSIs to corroborate how they are meeting and seeking to meet their commitments; a pattern we characterise as the outsourcing of living wage commitments to MSIs. Companies are already able to absolve some responsibility for low wages and wage violations through outsourcing production to suppliers. The additional outsourcing of their living wage commitment further muddies the waters of accountability about who is ultimately responsible for garment workers receiving living wages, and raises questions about the effectiveness of MSIs as there is little evidence they are making meaningful progress towards workers receiving living wages.

This willingness of companies to outsource their living wage commitments to MSIs is problematic in light of well-documented shortcomings about MSIs (as outlined in Section 3). While all of the MSIs in our study have different strategies for how to improve labour standards ranging from audit toolkits to better data collection, they are all voluntary and reliant on self-monitoring and self-reporting by companies to show compliance with the MSI’s standards, and there is a severe lack of accountability when external monitoring is applied. Crucially, none of the MSIs require companies to make changes to purchasing practices. For example, the Ethical Trade Initiative offers advice on responsible purchasing practices to its members but implementing that advice is at the discretion of companies. The Action, Collaboration, Transformation (ACT) initiative is more ambitious and seeks to provide living wages for garment workers by striking industry-wide national collective bargaining agreements (CBAs) that ACT members commit to align their purchasing practices with (ACT 2020). But in keeping with other MSIs the future effectiveness of ACT is an open question because companies will self-report on their implementation of CBAs. In short, MSIs in our study offer little guarantee of progress towards living wages given their reliance on self-reporting and lacking enforcement.

Our analysis of survey answers shows a clear pattern of companies extensively citing their involvement with MSIs – and often multiple MSIs – to evidence their commitment to living wages. Two companies refer to MSIs in 22 of their answers to the survey’s 26 questions. When asked to supply information about their purchasing practices that relate to meeting their wage commitments, many companies answered with direct reference to their MSI associations suggesting that companies see responsibility resting with the MSI and not the company.

Throughout the survey dataset, outsourcing responsibility for living wage commitments can be observed particularly strongly in answers given by companies that are members of ACT. Despite ACT being unable to identify any workers actually receiving higher wages as a result of their initiative (ACT 2020) ACT members in our sample – comprising fast fashion, luxury and online-only companies – frequently refer to ACT in very positive terms and cite their membership as evidence of their commitment to living wages. When asked how they define a living wage and to provide details of where their living wage commitment is publicly made, one company stated ‘Our central approach is the ACT Initiative on Living Wages’ before continuing in the same answer to provide a detailed overview of the aims of ACT and to refer to the company’s founding membership of ACT. In a response to a follow-up question which asked how the company monitors supplier compliance with this living wage commitment, the company states ‘We do not measure the payment of living wages as of yet although this will come through ACT as soon as the first industry-wide binding CBA is negotiated and signed by the social partners in the ACT countries.’ The company’s answers are illustrative of how ACT members pull the potential future CBA into the present as evidence of their commitment to living wages for workers. Yet, several years after living wage commitments were made, no ACT members are required to modify their purchasing practices because no CBAs have been agreed.

We also reviewed annual reports, sustainability reports and webpages of all 20 companies to assess how MSI involvement is used to evidence living wage commitments. 12 of the 14 survey respondents had webpages where MSI membership was displayed, and 10 make multiple mentions of MSIs in company reports. Five of the six companies that did not respond to the survey refer to MSIs in public reports and webpages. The majority of companies in our sample (17 out of 20) clearly publicise their MSI membership to evidence action around wage improvement, despite the lack of evidence provided to suggest that MSI involvement is leading to reform of purchasing practices and crucially that this has led to any workers actually receiving living wages.

4.3. The dilution and distortion of living wage

The MSIs that companies are involved with have their own definitions for living wages and wage requirements. And yet, whilst companies often draw on MSI definitions to inform their own living wage definition, the two are frequently misaligned, which is surprising given the close alignment presented by companies. Companies do not use standardised definitions for living wages and instead use their own definitions which are often limited in scope and do not meet the criteria for wages that sufficiently provide for workers and their families (see also Anker 2011). Our analysis also shows significant discrepancy between how companies define a living wage and their codes of conduct for suppliers. We find that the wage requirements placed upon suppliers in codes of conduct are soft, allow room for interpretation and often aren’t mandatory. The significant variance in how companies present their commitment to living wages whilst not providing evidence that they are acting to reshape their purchasing practices accordingly has the effect of diluting what is understood by a living wage. Furthermore, the lack of a common definition across companies and across MSIs distorts the very meaning of the term ‘living wage’. This not only casts serious doubt about the extent to which companies are significantly committed to make changes needed to realise their living wage commitments, but also gives license to suppliers to continue to pay non-living wages.

Our survey asked companies to provide their definition of a living wage. Column 1 in Table 3⁴ presents this information. It shows companies use multiple definitions and that across the different company types in our sample there is no common understanding about what a living wage for workers is. The company definitions are analysed against CCC’s living wage definition that a living wage covers the worker’s basic needs; provides some discretionary income; covers the needs of the worker's family, and is earned in a standard working week (i.e. without overtime being needed to top-up wages to a sufficient level). In their survey responses just three companies provided a definition of a living wage that included all four necessary components of the CCC definition.

Table 3. Company and MSI wage requirements analysed against the CCC definition of a living wage.^b

Key:
CCC's living wage definition is comprised of four necessary components. The wage must: 1 – cover the basic needs of the worker; 2 – provide discretionary income; 3 – cover the needs of the worker's family; 4 – be earned in a standard working week.
MSIs require company supplier codes of conduct to pay wages that include the following components of the CCC definition:
Action, Collaboration, Transformation (ACT): 1, 2, 3 & 4	Fair Labor Association (FLA): 1 & 2	Fair Wage Network (FWN): 1	Ethical Trade Initiative (ETI): 1 & 2
	Living Wage definition provided in their survey response	Wage criteria included in their Supplier Code of Conduct (SCoC)	MSI associations	Does the SCoC align with the living wage definition of the MSI?
Adidas	1,2	1,2,4*	FLA, FWN	yes, yes
C&A	1,2,3,4	1,2,3,4	ACT, ETI	yes, yes
Decathlon	1	1	-	-
G-Star RAW	1,2,3,4	1,2	FWN	yes
Gucci	1,2,4	1,2,4	FWN	yes
H&M	1,2,3,4	1,2,3	ACT, FWN, ETI	no, yes, yes
Inditex	1,2,3	1,2,3,4	ACT, ETI	no, yes
Nike	1,2	1,2,4*	FLA	yes
Primark	1,2,3	1,2,4	ACT, ETI	no, yes
Puma	1,2,4	1,2,4*	FWN, FLA	yes, yes
PVH	1,2,4	1,2,4*	ACT, FLA	no, yes
Tchibo	1,2,3	1,2,3,4	ACT, ETI	no, yes
Under Armour	1,2,4	1,2,4*	FLA	yes
Uniqlo	1,2	1	FLA	no
Amazon	Did not respond	–	–	–
Fruit of the Loom	Did not respond	1,2,4	FLA	yes
GAP	Did not respond	1,2*	ETI	yes
Hugo Boss	Did not respond	1,2,4*	FLA	yes
Levi Strauss	Did not respond	–	–	–
Zalando	Did not respond	4	ACT	no

^b An earlier version of Table 4 appeared in Edwards, Hunt & LeBaron (2019).

Companies provided scant evidence about their use of wage benchmarking methodologies. In the absence of requiring suppliers to pay wages in line with a calculated benchmark, suppliers are expected to pay wages that meet criteria set out in codes of conduct. Table 3 presents the discrepancies between the living wage definitions provided in company survey responses (Column 1) and the expectations included in supplier codes of conducts for wages that workers must receive (Column 2). 8 of the 14 survey respondents provided a living wage definition misaligned with their own code of conduct. Just one company has a code of conduct that expects suppliers to pay wages in accordance with the four components of a living wage outlined by CCC.⁵

The wage requirements in codes of conduct are frequently vague and allow room for divergence by suppliers, thus making an assessment of any noncompliance hard to discern and compliance hard to enforce. The codes of conduct for eight companies in our sample contain an acknowledgement that suppliers may not be able to pay wages that meet the requirements in the code. This effectively licenses suppliers to pay non-living wages.⁶

Two companies provide suppliers with no or minimal guidance about the wages workers must receive. The first, an online retailer, has a code of conduct that includes no details about wages that suppliers must pay. The code of the second, a luxury garment company, states they will only do business with suppliers that ‘provide wages and benefits that comply with any applicable law and match the prevailing local manufacturing or finishing industry practices’; this grants significant license to suppliers to pay sub-standard wages given that production often occurs in jurisdictions where laws on worker compensation are lacking and unenforced, and where ‘prevailing … industry practices’ routinely means low wages.

The analysis in Table 3 (Columns 3 and 4) also shows the frequent misalignment between company supplier codes of conducts and the living wage definitions of MSIs that companies associate with and refer frequently to. We draw particular attention to the misalignment between the codes of conduct of six of the seven companies in our study that are members of ACT and ACT’s definition of a living wage. The misalignment is most pronounced with one company whose code makes no reference to wages needing to meet the basic needs of a worker and their family, nor provide discretionary income, thus falling far short of ACT’s living wage definition. When companies place great emphasis upon their MSI affiliations as evidence of their commitment to living wages yet do not align their supplier codes of conduct with those same MSIs it prompts further scepticism about the strength of those commitments.

4.4. Lacking progress

In assessing whether companies are modifying their purchasing practices in ways that could better achieve the payment of living wages to workers, we also considered the evidence they presented to directly substantiate their progress in meeting their commitments. We found limited signs of progress. Companies were asked if they have a public roadmap for how, and by when, they will achieve living wages for all workers in their supplier network, but none provided a clear timeline for how and by when this goal would be met. Across all companies in our sample, the lack of public roadmaps, published indicators of progress and transparent information about suppliers means it is extremely difficult to make meaningful assessments about any progress companies may be making.

Transparent publication of information about suppliers is a key data source for external observers – whether NGOs, workers or unions – to monitor wages and labour standards. Companies were asked if they publish information about suppliers including addresses, parent company, products supplied, the number of workers employed and crucially, wage data. Whilst 15 companies publish a supplier list with addresses, only seven publish supplier information below Tier 1. No company publishes fine-grained wage data at the supplier-level and just four companies publish any form of wage data.⁷

Whilst most companies in our sample do not publish detailed information about suppliers, companies do monitor supplier compliance with their wage requirements through social auditing; a tool well-documented to be sub-optimal for detecting and remediating non-compliance issues in supply chains (Anner 2017, LeBaron et al. 2017). Just two companies in our sample solely use unannounced audits and only nine stated that audits may involve worker interviews. Only five companies confirmed they audit suppliers below Tier 1. This raises questions about the accuracy of claims about labour standards compliance because previous research documents that workers are more vulnerable to exploitation below Tier 1.

Furthermore, whilst all 20 companies use auditing, only 12 use compliance performance to rank suppliers which may determine the frequency and volume of future orders and whether the supplier remains within their network. The fact that not all companies in our sample use compliance performance to determine whether suppliers receive orders raises further serious questions about the willingness in the industry for fundamental reform of purchasing practices to meet living wage commitments and increase labour standards.

5. Conclusion

An influential definition of CSR comes from David Vogel, who defines it as ‘practices that improve the workplace and benefit society in ways that go above and beyond what companies are legally required to do’ (Vogel 2005, p. 2). His definition assumes that CSR leads to improved conditions for workers. But how should we understand CSR commitments and initiatives that don’t lead to improvements?

Returning to our key focus of how we can understand garment company actions to deliver recent commitments to living wages in their supply chains, we found very little evidence that companies followed their commitments with meaningful action. Not only are companies largely failing to modify the practices directly within their control that lead to sub-living wages being paid, but across the documentation from companies and MSIs, there was almost no evidence that commitments to living wages are actually being realised on-the-ground, or even that the commercial preconditions to making this possible were being established. This underscores the need for scholars concerned with CSR to look beyond procedural questions, such as around the design and uptake of MSIs and codes of conduct, and to investigate effectiveness. In our case, any initial optimism about garment company living wage commitments and the launching of MSIs is outweighed by the seriously lacking evidence that they are delivering living wages for workers.

The ineffectiveness of these commitments and failure of companies to make structural changes in their business models cannot be viewed in isolation. Companies yield immense power over supply chains and over working conditions in the global economy more broadly. Further research is required to assess whether and how CSR may help to entrench the business model that leads to persistently low wages; not just passively failing to improve labour standards but actively stabilising and facilitating garment industry supply chains that are widely documented to lead to labour exploitation. We have argued that fine-grained analysis of the actions taken by companies to realise their commitments to pay living wages to workers within their supply chains provides new insights into how corporations are exerting power in the global economy. Features of CSR often taken to be benign—such as incrementalism, density and volume of information and activity, and adopting multiple standards— are in fact worrisome insofar as they give the impression that problems like illegally low wages are being addressed when they are not. We have argued that garment companies are using these features of CSR as a strategy to shield themselves from pressures to significantly reform their highly profitable business model. Incrementalism reflects not a slow but steady process of changing complex systems, as is often assumed, but rather, a deliberate strategy to draw out commitments and create the impression that meaningful change is occurring while corporations continue to profit.

Acknowledgements

Funding to support the research was provided through the Economic and Gender Justice in the Global Apparel Industry Project at the Worker-Driven Social Responsibility Network, which is supported by the NoVo Foundation.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This work was supported by The Novo Foundation: [Grant Number Economic and Gender Justice in the Global Apparel].

Notes on contributors

Genevieve LeBaron

Genevieve LeBaron is Professor of Politics within the Department of Politics and International Relations at the University of Sheffield.

Remi Edwards

Remi Edwards is an ESRC-funded doctoral student within the Department of Politics and International Relations at the University of Sheffield.

Tom Hunt

Tom Hunt is Deputy Director of Sheffield Political Economy Research Institute (SPERI), University of Sheffield.

Charline Sempéré

Charline Sempéré is an ESRC-funded doctoral student within the Department of Politics and International Relations at the University of Sheffield.

Penelope Kyritsis

Penelope Kyritsis is Director of Strategic Research at Worker Rights Consortium.

Notes

1 Since our analysis was conducted CCC have adopted the Asia Floor Wage Alliance’s definition for what a living wage in the garment industry should comprise (see https://cleanclothes.org/poverty-wages). The version used in our study is available at https://archive.cleanclothes.org/livingwage.

2 Six of the nine companies are members of ACT and so are waiting for CBA negotiations to conclude before factoring any agreed wage level into FOB prices.

3 Information in this section refers to the 14 survey respondents. Information about purchasing practices was not available publicly for non-respondents.

4 An earlier version of Table 3 appeared in Edwards et al. (2019). The MSIs included were the most prominent in company survey responses.

5 C&A meets the CCC definition because they state that a standard week must not exceed ‘48 h per week excluding overtime’. The codes of conduct of H&M, Inditex and Tchibo refer to all four components of CCC’s definition but are more imprecise about what a standard working work entails. For example, Inditex state ‘Manufacturers and suppliers shall not require their employees to work, as a rule of thumb, in excess of 48 h a week (emphasis added)’.

6 The companies are indicated in Column 2 of Table 3 with an asterisk.

7 One company publishes aggregated regional data on the percentage of its suppliers that comply with the wages section of its code of conduct. Two companies publish country-level data comparing workers’ wages with average minimum wages. A fourth company publishes information about wage violations by suppliers but not about actual wages paid.