Introduction

In April 2013, the loss of over 1,000 lives in Bangladesh due to the collapse of the Rana Plaza garment manufacturing facility, which supplied a number of leading international retailers, brought into sharp focus the challenges of sustaining an efficient global supply chain and the grave implications when corporate social responsibility (CSR) concerns are overlooked (The Economist 2013). The expanding geography of sourcing and manufacture of apparel over the past 20 years has resulted in an increasingly globalised retail industry (Gereffi et al. 2005) with garment sourcing representing a balancing act between manufacturing cost and spatial proximity/lead time considerations (Abernathy et al. 2006; Masson et al. 2007). The trend towards vertical disintegration of retailers’ supply chains increases the degree of global dispersal (Gereffi 1999) and sees a shift of the garment manufacturing function to lower labour cost countries (Pickles 2006; Taplin 2006; Pickles and Smith 2011). Concomitantly, debate increasingly focuses on the socio-economic impact of global business operations on developing countries with respect to worker safety, exploitation, sweatshops and child labour (New 1997; Smestad 2009). Retail firms increasingly face pressure to deliver CSR alongside shareholder profits (Palpacuer 2008; Rivoli 2003). The implementation of CSR has proved to be challenging across complex global production networks (GPNs), as evidenced by recent scandalous allegations of supply chain misconduct against leading fashion retailers (Smestad 2009; Chamberlain 2010). Such controversies illustrate the potential for fundamental conflict between sustainability objectives and overriding commercial imperatives of price competitiveness and responsiveness in the fashion retail sector (de Brito et al. 2008).

Previous research literature exploring ethics in garment sourcing networks has focused on both retail buyer and worker perspectives (Hughes 2005; 2012; Pretious and Love 2006), and detailed subsequent development and policy implications (Ruwanpura 2012; Ruwanpura and Wrigley 2011; Hancock et al. 2010). This qualitative research paper adopts a strategic perspective by accessing the situated knowledge at one nodal point of the production network: the managers of apparel production facilities in a developing country context. Our objective is not to establish whether CSR initiatives are effective in our case study country. Instead, the value of this study lies in its conceptualisation of how factory managers make sense of CSR implementation in their role as intermediaries between the powerless (garment workers) and the powerful (retail buyers). We explore how they rationalise the competing demands of social responsibility against the competitive challenges of apparel industry conditions—namely downward price pressure, responsiveness and the power imbalance between retailers and suppliers.

The instrumental role of management in espousing the organisation’s CSR policy is acknowledged by Hemingway and Maclagan (2004), who conceptualised CSR as an expression of individual managers’ values and as an outcome of the managerial decision-making process. Given calls for greater interdisciplinary research into retail globalisation (Coe and Wrigley 2009), our research attempts to access situated managerial insights into the complexities of managing ethics in fashion retailers’ international production networks. In doing so, we provide managerial insights for practitioners by identifying CSR activities and strategic CSR outcomes that are regarded as cementing a reputation for ethical practice with retail buyers, while also improving worker retention and productivity. Sri Lanka provides an interesting location for the study due to its established reputation for ethical garment manufacturing (Ruwanpura and Wrigley 2011; Loker 2011) and the importance of the apparel industry to the national economy as the country’s largest foreign exchange earner (Kelagama 2005). Hence, our insights from a country with a well-developed CSR reputation are likely to provide valuable knowledge for apparel producers in other regions currently facing high levels of cost-based competition.

The paper is structured as follows: we introduce the concept of CSR within the context of changing geographies of garment production networks and discuss how its implementation is affected by the commercial context of the fashion industry and retailer buying practices. We then set out the research methodology adopted to understand the nature of garment GPNs and their impact on CSR implementation. Our findings relating to suppliers’ perspectives on the dynamics of CSR in Sri Lankan export garment manufacturing are then examined. Our conclusion details factory managers’ perceptions of the positive drivers and negative inhibitors of ethical trading in the Sri Lankan context and considers the academic and managerial implications thereof.

Corporate Social Responsibility in Global Garment Production Networks

Due to the labour-intensive nature of the manufacturing process and its relative lack of technical innovation, garment sourcing decisions are significantly determined by labour costs and therefore developing countries hold greater competitive advantage in this respect (Abernathy et al. 2006; Taplin 2006; Begg et al. 2003). However, the global shift of production to lower labour cost countries has been linked with worker exploitation in the outer tiers of the network (New 1997; Smestad 2009). While retailers retain higher order functions of distribution, marketing and brand management, manufacturing activities are outsourced to a complex network of independent subcontractors. This shift may result in reduced visibility and control of the worker experience for the lead firm (Doorey 2011). Indeed, Miller (2004) notes how the daily experiences of female workers are often concealed within complex production networks, and consequently women are increasingly at the centre of NGO (non-governmental organisation) campaigns concerning worker rights (Hughes et al. 2010).

Barrientos (2013) explores how NGOs take up the cause of workers by adopting variegated strategies of engagement with lead retail firms in the production network—at times involving alliances and at others pursuing more adversarial campaigns which may threaten the hitherto dominance of the lead firm in the network (cf. Coe and Hess 2013). In the process, Barrientos argues such initiatives involve strategies that emphasise the commercial (risk) or alternatively the social (caring) dimensions of corporate engagement. Retailer responses are understandable given the vulnerability of brand equity to changes in public trust (Dickson and Eckman 2008; Kang and Hustvedt 2013) and potential disruption to the timely flow of product through the supply chain in the event of industrial action at subcontractor facilities. Retailers attempt to address working conditions in the supply chains that produce their merchandise, irrespective of whether or not they own the production facilities (Andersen and Skøtt-Larsen 2009; Hughes 2005; Hughes et al. 2007). Retailer codes of conduct and broader international multi-stakeholder schemes and accreditations seek to address the negative outcomes of globalisation in terms of labour standards, by imposing various conditions upon the lead firm and its suppliers (Hale and Wills 2007; Hughes et al. 2007; 2008; Ruwanpura and Wrigley 2011). CSR may thus be perceived as a form of privatised governance in the absence of necessarily tight national forms of regulation (Christopherson and Lillie 2005; Tallontire 2007; Mayer and Pickles 2010), though its uneven implementation remains a concern (Ruwanpura 2013). Despite this, retailers are keen to publicise their commitments to worker conditions and wider CSR in communications with stakeholders (Mann et al. 2013).

In a broad sense, CSR concerns ‘the continuing commitment by business to behave ethically and contribute to economic development, while improving the quality of life of the workforce and their families as well as of the local community and society at large’ (WBCSD 1999, p. 3). Of course, the precise meaning of CSR continues to evolve over time and has slightly different interpretations across the social sciences (Carroll 1999); in-depth discussion of which lies beyond the immediate scope of this article. Put simply, the term conveys the idea that business should assume social responsibilities beyond any economic, technical and legal obligations which constitute its existence, with the objective of reconciling economic, social and environmental objectives (Davis 1973). Actions must go beyond the immediate interest of the firm and progress further than merely obeying the law or other regulations (McWilliams and Siegel 2001; Davis 1973).

In practice, the implementation of CSR within garment production networks commonly involves the application of guiding mechanisms and management tools such as codes of conduct and ethical audits to encourage socially responsible practices in the manufacturing process. Therefore, the rhetoric of CSR implementation may not always conform to the ideal of going beyond minimum legal or regulatory requirements: Welford and Frost (2006) note that CSR is often perceived as a compliance issue in Asian supply chains, while Hughes (2005) notes how the notion of ethical trade is commonly understood to mean the establishment of a minimum set of standards for suppliers—something that may be seen as conflicting with framing CSR as a voluntary activity (Dahlsrud 2008; McWilliams and Siegel 2001).

A rules-based approach to CSR governance does not necessarily lead to improvements on the factory floor or increased worker involvement in the governance process (Raj-Reichert 2013; Ruwanpura 2013). Indeed, active worker participation in corporate code implementation is seen as important in securing worker benefits (Yu 2009). At its worst, the adoption and implementation of codes of conduct may be seen as little more than a PR exercise to deflect further criticism of lead firms, with the monitoring of codes a mere box-ticking exercise that fails to fully address exploitative working conditions for the workers’ benefit. Plank et al. (2012) note the limitations of apparel retailer codes of conduct in affecting the social upgrading of supplier facilities, given their focus on measurable standards such as wages, working conditions, and health and safety issues, rather than enabling rights which seek to provide voice and empowerment to workers, such as freedom of association and collective bargaining. More widely, numerous researchers discuss widespread economic upgrading of the global apparel value network, while acknowledging that social upgrading remains somewhat more challenging to achieve (Barrientos et al. 2011; Bernhardt 2013).

Current thinking suggests that CSR initiatives may extend beyond good corporate citizenship to confer competitive advantage to the firm (Garriga and Melé 2004; Porter and Kramer 2006). This is known as strategic CSR (Lantos 2001). Increasingly, firms are using CSR benefits to serve their own commercial interests (Hanlon and Fleming 2009) to actively manage their image and reputation; manipulate stakeholders; and embed the organisation within its host community (Hemingway and Maclagan 2004). In doing so, the organisation enhances its long-term loyalty, legitimacy and brand equity (Godfrey and Hatch 2007). Conceptualising CSR as a strategic business initiative broadens our understanding of the reality of corporate drivers and inhibitors of better working conditions in GPNs. Of course, in practice, firms are focused on the pursuit of profit rather than developmental goals as such (Schrank 2004). For example, CSR initiatives in the agri-food sector evolved from technical and quality requirements to the inclusion of social and environmental criteria—largely to protect reputations and enhance quality control rather than any explicit goal to improve supplier welfare (Tallontire 2007). Therefore, our understanding of CSR in apparel production networks is the attempt by retailers to reconcile social and environmental objectives alongside economic ones. CSR is framed as a strategic business activity rather than a development mechanism as such. We now set out the key characteristics of global apparel production networks that may compromise successful implementation of CSR objectives.

Compromises to CSR in Global Apparel Production

The lead retailers that orchestrate global apparel production networks are conscious of their CSR vulnerabilities and keen to promote their CSR credentials. Achieving CSR brand leadership requires a shift from viewing the brand as tactical and reactive towards one that is strategic, multi-stakeholder, and visionary (Lindgreen et al. 2012). But, of course, CSR needs to be embedded throughout the entire value chain and within stakeholder relationships (cf. Maon et al. 2009). Even when efforts have been made in this direction, with lead firms co-operating with workers, NGOs and trade unions, results often remain disappointing (e.g. Lund-Thomsen and Coe 2013 in the case of Nike). Specifically in the highly competitive mid-market fashion retail sector, the realities of CSR demands may be compromised at the factory level by the need to balance the competing commercial realities of cost and lead time (Plank et al. 2012; Hearson 2009; Barrientos and Smith 2007). In particular, flexibility and responsiveness must be finely balanced against sourcing cost to achieve success (Hines and McGowan 2005; Masson et al. 2007).

Given a backdrop of demanding buying practices of powerful Western retailers that require more frequent shipments of cheaper garments, a supplier’s ability to fund and therefore adhere to ethical standards is challenged (Laudal 2010; New 1997). Tokatli et al. (2008) note how retailers have demanded greater product variety from manufacturers as well as adding mid-season purchasing to the traditional two-season calendar. Palpacuer et al. (2005) found that UK retailers made increasing demands of their suppliers for production flexibility and risk transfer, in addition to imposing continuous price reductions. In particular, increased payment terms and cost reduction practices affect the supplier’s ability to manage costs, especially for full package suppliers that purchase fabric upfront. Although open-book accounting allows buyers to facilitate cost management across the supply chain network by accessing suppliers’ costing data (Christopher 2000), this practice may also be used opportunistically for inducing competitive bids from rival suppliers (Free 2008).

Buying practices that have developed to address commercial pressures may jeopardise labour standards at the factory level and undermine CSR efforts, especially with the fast fashion business model (Hearson 2009). Industry evidence suggests that retail buyers are unwilling to increase prices paid to suppliers to reflect the increased cost of ethical production (Ruwanpura and Wrigley 2011; Yu 2008; FIDH 2008). Hughes’ (2012) study of retail buyers during the Western recession found that as ethical trading budgets succumbed to growing financial pressure, there was greater focus on strategic CSR to deliver benefits to the business in terms of efficiency gains and reputation protection, as opposed to improvements to labour standards being pursued as a fundamental human right. Ruwanpura and Wrigley (2011) concluded that the compatibility of retailer sourcing strategies with the implementation of CSR in production sites was further challenged by the global recession, as suppliers lost orders to lower labour cost countries and buyers prioritised price over compliance. Similarly, in the Caribbean basin, Schrank (2004) noted how full package suppliers sought to undercut each other in the race for greater market share during a period of economic recession, which resulted in downgraded conditions for workers.

Of course, it is not only individual suppliers that are competing for apparel manufacturing business; developing countries may relax national regulations to attract foreign buyers which may further threaten CSR initiatives and worker outcomes. Numerous studies have noted relaxed labour regulations in export processing zones which conflict with the ILO Core Convention for respecting workers’ rights to freedom of association and collective bargaining (Miller 2004; Arnold and Han Shih 2010; Hancock et al. 2010). It becomes apparent that the implementation of CSR initiatives at the factory level becomes highly problematic, given the backdrop of industry commercial pressures, labyrinthine GPNs and inconsistency in labour standards between codes of conduct and governmental FDI incentives (see Ruwanpura 2012 in the case of Sri Lanka). Therefore, consideration of the influence of industry characteristics is relevant for understanding the nuanced challenges of CSR implementation within the global garment industry. Guiding mechanisms and management tools to encourage socially responsible practices often have questionable effectiveness in the apparel supply chain, given the complexity and the inherent focus on sourcing cost and lead time (Mamic 2005; Welford and Frost 2006; Mares 2010).

Contextualising Our Focus: Management CSR Narratives at the Factory Level in Sri Lanka

It is clear that there is a growing body of research on the changing geographies of garment sourcing and an increasing focus on CSR, working conditions and wider ethical aspects of expanding buyer-driven apparel GPNs into the developing world. While previous literature on CSR in apparel GPNs has focused on the worker perspective (Ruwanpura 2012; Ruwanpura 2013, 2014a) and despite growing attention to the issue as evidenced by the international research project ‘Capturing the Gains’ (Miller et al. 2011; Mayer and Pickles 2010), there has been less focus on understanding production management’s role in espousing CSR within the GPN. To the best of our knowledge, this is a clear gap in the research literature and is the contribution of this paper. Importantly, factory managers represent a key stakeholder group situated at the ‘hard end’ of the CSR agenda, given their role in underpinning the supplier’s commercial competitiveness, while ensuring ethical labour force requirements are met.

In this paper, we explore CSR in the garment supply chain from the perspective of factory managers in a developing producer country. We seek to conceptualise how factory managers mediate between a CSR focused drive to provide socially responsible outcomes for workers, while accommodating ever present commercial imperatives. This exploration of manufacturers’ situated knowledge at one point of the production network is distinct from a study explicitly focusing on whether desirable CSR outcomes are actually being realised or beneficial.

Our study is focused on Sri Lanka. Sri Lanka’s government directives, labour union histories and societal norms provide a strong foundation on which to build CSR progress with retail buyers (Ruwanpura and Wrigley 2011; Knutsen 2004). Given the Sri Lankan tradition of protecting the virtue of the female in society (Lynch 2007) and the industry’s early understanding of the importance of human resources and promotion of gender equality (Chandrakumara and Budhwar 2005), the country already had the foundations in place for addressing working conditions in the female-dominated garment industry. The performance of Sri Lankan garment manufacturers in certain aspects of CSR is thus not only explained in terms of industry voluntariness (Dahlsrud 2008; McWilliams and Siegel 2001), but also by the supportive local conditions of strong labour laws and socio-development achievements (Ruwanpura 2012; Hancock et al. 2010). Ethical practices in Sri Lankan garment manufacturers are reinforced by cultural norms, specifically the moral teachings of the predominant religious persuasion of Sinhalese Buddhism, which demands fairness, social justice and equity (Perry 2012). Goger (2013) notes how Sinhalese Buddhism shaped local behaviours in Sri Lanka to promote the sharing of wealth in the local community; respectable and abstinent behaviour; and education, with an implied ban on child labour. Additionally, Sri Lanka’s (British Victorian) colonial history, with its notions of female sexual purity, cleanliness and domesticity, put pressure on garment factory owners to protect the decency of the largely female workforce (Goger 2013). These cultural underpinnings manifested themselves in the personal values of factory managers, and meant that some of the worker welfare practices which would later become known as CSR had been long embedded within Sri Lankan value systems (Loker 2011; Perry 2012).

In 2006, the ‘Garments Without Guilt’ campaign was launched by the Sri Lankan national garment and textile industry body to promote the country’s ethical credentials by assuring buyers that garments produced in Sri Lanka were made under ethical conditions. This aimed to leverage the reputation the country was already establishing in ethical production and provide some insulation from lower labour cost countries. Price competition heightened from 2005 following the phasing-out of the quota-based global sourcing system under the Multi-Fiber Arrangement (an international trade agreement on textiles and clothing that ran from 1974 to 2004) (cf. Goto et al. 2011). The aim was, therefore, to establish a differentiated form of competitive advantage founded on ethical manufacturing and to become recognised as a leader in terms of lack of child labour, high health and safety standards and the ratification of ILO Core Conventions (Goger 2013).

In common with the reasons for development of CSR initiatives in the agri-food sector (Tallontire 2007), the decision to promote ethical manufacture in Sri Lanka was based on strategic intentions rather than fundamental moral underpinnings or developmental objectives. It was geared to enhance the nation’s competitiveness following the shift to free competition in global markets (MacCarthy and Jayarathne 2012) and was driven by the industry, not directly by retail buyers or policymakers. However, despite increasing calls from retailers for ethical credentials, the chronic price deflation in the industry sector meant that suppliers found it increasingly difficult to capture significant added-value as a result of their efforts to rebrand as an ethical manufacturing location (Goger 2013). In addition, Ruwanpura (2014a) reports that while Sri Lanka has been positioned at the vanguard of ethical sourcing, ethical codes are not simply implemented, but inconsistently interpreted and applied, leading to tensions for labour. Sri Lanka thus provides a unique perspective to explore managerial discourses of CSR.

Research Methodology

Our empirical data collection is characterised by an approach reminiscent of what Clark (1998) referred to as close dialogue with key elites—a methodology subsequently employed across a wide range of recent research in the social sciences (e.g. Hughes 2005, 2012; Hughes et al. 2008; Wood and Reynolds 2011). Data collection involved semi-structured interviews with key managers within Sri Lanka’s export garment manufacturing industry and permitted access to various actors at different levels of seniority and responsibility within the firms, including managing directors, operations managers and HR managers.

Aware of the challenges associated with ‘seduction and co-option’ by experienced managerial elites (Clark 1998 p. 80), a rich process of data collection was adopted. The lead researcher maintained an in situ presence in Sri Lanka over a 7 month period which facilitated repeat visits, factory tours and informal discussions with a wide range of actors positioned at various levels of seniority within the constituent firms, alongside more formal interviews with key managers. Interviewees were assured of both personal and company confidentiality in order to build rapport and facilitate an honest dialogue. This approach contributed to the on-going system of triangulation and sense-checking of information collected (Yeung 2003). Triangulation was achieved by conducting interviews within the ‘context’ of the cases such as consultants, government advisers and industry associations (cf. Perry 1998) and included a Senior Manager charged with investment appraisal for the garment industry at the Board of Investment (a Sri Lankan governmental body charged with promoting and regulating the export garment industry) as well as the Head of Sustainable Business at a large UK mid-market retailer on the basis that this was a common customer for most companies. Managerial interviewing triangulated with key informant ‘context’ interviewing is a valid approach that has previously been applied to the study of CSR (Hughes 2012; Ruwanpura and Wrigley 2011).

Our research focused on seven purposefully sampled companies that represented a variety of business sizes and business models in the industry, and formed part of several EU, US or Australian retailers’ supply chain networks (see Table 1). The sample included three of the top nine garment exporters in Sri Lanka (Invest Sri Lanka 2013). In terms of its best practice and innovation in CSR, as evidenced by numerous industry awards (Goger 2013), Sri Lanka may be conceptualised as an ‘extreme case’ location. Indeed, the country markets itself as the world’s #1 ethical garment sourcing destination (Sri Lanka Apparel 2013). The justification for this type of case study, notwithstanding its limited scope for generalisation, is its potential for greater understanding of the context in which CSR is embedded than in the study of an ‘average case’ location (Yin 2003). Although previous research has identified a disconnect between the rhetoric of CSR and garment workers’ cultural politics and lived experience in Sri Lanka (Ruwanpura 2012, 2013, 2014a, b; Hewamanne 2008; Lynch 2007), our focus within this paper is not to assess the extent to which CSR exists in the Sri Lankan garment industry, but rather to take the factory management perspective concerning the challenges of CSR more broadly. As such, we do not consider the success or otherwise of CSR initiatives, but rather explore how CSR is framed by leading manufacturing managers in the context of commercial pressures and relations with wider stakeholders. Thematic analysis of interview transcripts was conducted manually, using Eisenhardt’s (1989) method of within-case and cross-case analysis, and triangulated with observational data and documentary evidence (Yin 2008). Findings are presented alongside verbatim quotes to ensure transparency and rigour (Baxter and Eyles 1997).

Table 1 Breakdown of export garment manufacturer companies
Full size table

Analysis: Sri Lankan Export Manufacturer Perspectives on CSR Implementation

Across all cases, manufacturers offered narratives of engagement with the CSR agenda, though this was not always formalised as such in the smallest businesses. Larger suppliers had formulated their own codes of conduct and CSR initiatives that addressed the ethical implications of the firm. Programmes included raising machine operators’ salary and minimum working age beyond that required by national law, rehabilitating ex-LTTE (Tamil Tiger) soldiers to work in newly opened plants in the Northern Territories, female empowerment programmes, and undertaking philanthropic donations to local communities and monthly (rather than annual) payment of production bonuses. Entrenched cultural norms underpinned the factory managers’ social responsibility for the workforce and explained the sunken nature of many CSR practices that predated retailer codes of conduct:

The mindset is that you can’t exploit people who are under you…to get large money out of your business… The Buddhist culture demands fairness and social justice and equity between all the people and so it is a part of the culture…[Even if] no such demands, no rules and regulations, no standards, still I would say 70-80 % of the things which we are practising would [still exist]. (Company A)

Similarly, the education levels and social upbringing in Sri Lanka facilitated the ability of the workforce to grasp the concept and significance of CSR:

The upbringing and education levels of people in Sri Lanka are very good; even today the rural girls are very educated. They don’t have to go to university to understand CSR, they know…they get the benefits of CSR. So I think it plays a major role. I would say the people understand what this is all about. (Company E)

Framing CSR as a Strategic Productivity Imperative

Consistent with the findings of Hemingway and Maclagan (2004), our data showed how the managerial discourse of CSR was focused on the strategic commercial interest of the supplying organisation, which was itself partly the product of the demands of the lead retail buyer. CSR was conceptualised by managers as a means of keeping workers happy and productive, as well as providing an ethical point of differentiation to retail buyers.

Many factories had practised what they considered to be CSR activities for a number of years that demonstrated their caring for the largely female workforce, such as providing free breakfast, transport and healthcare. The provision of employee benefits, such as free breakfasts for workers, it was argued, had the potential to increase productivity beyond that possible when working on an empty stomach. Furthermore, some of these employee benefits over time had become enshrined in Board of Investment (BOI) regulations for garment exporters and therefore established as a minimum standard for labour conditions. This point echoes Welford and Frost’s (2006) observation that CSR is often perceived as a compliance issue in Asian supply chains, more focused on technical outcome standards than enabling rights (Barrientos and Smith 2007). Philanthropic CSR activities such as donating to local orphanages or building schools also conferred strategic benefits on the supplier by fostering goodwill and visibly contributing to the licence to operate within the local community. Similarly, Goger’s (2013) study highlights how factory owners initially implemented CSR as a means of garnering community respect, to gain a licence to operate and reduce labour issues, rather than for the benefit of the workers as such. CSR in Sri Lanka appeared to be characterised by the presence of measurable standards, such as wages and benefits, rather than enabling rights such as freedom of association or collective bargaining.

Enabling rights are based on intrinsic principles of social justice and facilitate progressive change in labour relations. Although freedom of association and collective bargaining are enshrined in the Core Conventions of the International Labour Organisation (ILO), the upholding of these rights in developing country production locations is often problematic and uneven (Barrientos and Smith 2007). Sri Lanka’s suppliers have been criticised for their lack of trade union voice for garment workers (Ruwanpura 2014b), despite the fact that freedom of association and collective bargaining are enshrined in Sri Lanka’s own constitution as well as being recognised in the Garments Without Guilt Charter. In practice, however, factory managers advocated quasi-union initiatives for workers, such as employee representation councils and dedicated HR representatives to deal with any issues the workers had, whether in the workplace or at home. Although employee councils have been condemned for not being independent of the employer and essentially undermining the formation of unions (Busser 2005), as well as being a contravention of the ILO Core Convention for Freedom of Association (Ruwanpura and Wrigley 2011; Ruwanpura 2012), factory managers argued that they promoted a harmonious workplace as best they could under the circumstances:

We have a joint consultative committee in the factory representative of the workers and we have a monthly meeting with them to ensure that we have a good relationship and also that their grievances are brought up at the meetings. (Company D)

We have three people in the Personnel Department: two ladies and one man. Almost every day [they] go to the lines and individually they are talking with the employees … because sometimes in the factory they have some problems, sometimes in their home… Every day they are asking and write down and come and tell me the girl have a problem. Then we can sort out the problem. (Company B)

The overriding motivation for good labour relations appeared to be the strategic commercial interest of the firm: happy workers were more reliable, loyal and productive (cf. Ruwanpura and Wrigley 2011), which resulted in fewer factory scheduling issues for management as a result of absenteeism or labour turnover. Managing the employer–employee relationship was seen to promote harmony in the workplace, which ultimately served the company’s commercial interests, rather than supporting workers’ efforts to negotiate better labour conditions. This finding is consistent with Barrientos and Smith’s (2007) concerns regarding the failure of codes of conduct to enhance the process through which workers claim their rights.

Outcomes of good working conditions were often captured through metrics which underlined the benefit to the business—for example, the HR Manager of Company E reflected proudly on the low turnover rate and the longevity of service for a large proportion of workers in the factory. In this company, the HR office was located on the factory floor rather than in the management suite, with full length glass panels for walls so that workers felt they could approach HR officers more easily (but also provided HR officers with a clear view of the workers). To encourage workers to return to their jobs after marriage and children, Company E instigated a career path for workers to progress to supervisor level. In a country where women traditionally work in the garment industry only until they marry and are able to claim state benefits (Lynch 2007), this was a striking innovation. It remains to be seen whether the presence of female supervisors on the factory floor would result in a less harrowing experience for machinists due to harassment or abuse from male supervisors—a concern of recent research studies (Lynch 2007; Hancock et al. 2010; Ruwanpura 2012). In Company D, managers contended that machinist training was geared towards building a multi-skilled workforce which contrasted with the traditional Taylorist division of labour in garment factories, where workers specialised in one operation only and were consequently unable to sew a whole garment at home, even after several years’ service (Lynch, 2007):

Although initially they specialise in one operation, if they retain with us, we ensure by end of second year they are fully skilled, multi-skilled operators and we bring up to the Grade A level where they get the highest salary by internal job rotation. We train them, we encourage them even at the interview, we say this is your basic but after 6 months Grade B you get this much, and up to Grade A you can go and don’t wait here, try to get this higher salary. Because we don’t need anyone to retain as a trainee forever, we also want them to be skilled. (Company D)

For workers, this resulted in a less monotonous job and the opportunity to develop higher order skills and pay. However, the underlying motivation from the point of view of factory managers was to improve loyalty and motivation, retain the skills of experienced workers, and hence reduce the cost of recruitment and training of new employees. Additionally, a multi-skilled workforce reduced issues of line scheduling due to absence. Factory managers were aware of the role of CSR in supporting sustainable business development in a labour-intensive industry sector, in terms of retaining workers, enhancing productivity, and building employee loyalty and commitment. CSR was thus framed as a strategic choice and a means of achieving tangible benefits for the firm as well as for society:

Most of the businesses especially in apparel are looking for a sustainable business. They do understand that unless you pay enough and look after your employees enough then ultimately you will not have enough people to work in your business (Company A)

Framing CSR as a Strategic Differentiator

Factory management recognised that Sri Lanka’s global reputation for CSR and eco-manufacturing provided them with a unique point of differentiation against other lower labour cost countries. CSR was thus perceived by senior management as a form of national competitive advantage:

We are using (CSR) as a weapon. We have clean factories; we have superb manufacturing compared to other neighbours in this part of the world. (Company A)

You are competing with a lot of companies not only from Sri Lanka but in China, India, Bangladesh, Vietnam. There is something unique for us to tell our buyers that we are doing our business ethically, we are having these types of characteristics in our business which are not followed by other companies in China or India. Our company is so much better than those other places. Thereby we can get even more business, good business. (Company C)

CSR was also characterised by larger manufacturers as a means of differentiation against rival companies within Sri Lanka in terms of attracting and retaining high quality machine operators which were crucial to success, given the labour intense nature of the sector:

One challenge we have at the moment in this industry is scarcity of labour. You don’t find enough good people. So it’s a battle to attract good people to your business. We have a large workforce, it’s over 25,000 people. If I take this zone, there are about 30 entities similar to us. So there is a lot of competition for good people… So for us to benchmark ourselves and highlight the differences between another company and us, our ethical practices have helped us a lot … So ordinary employees will understand the difference between a [Company A] plant and another average plant. So it’s very easy for us to compare and show them that we are a different employer. (Company A)

Our interviews suggested that CSR implementation was supported by the notion of adding value to the business, in terms of meeting buyer requirements as well as addressing labour shortages by attracting and retaining good quality workers. Despite restrictions on workers’ rights to organise and bargain collectively (Ruwanpura 2012; Dent 2002), Sri Lankan export garment manufacturers have won several international awards for social and environmental responsibility as well as being the focus of a number of best practice case studies in recent years (Goger 2013). This illustrates the disjuncture between relative and absolute performance—Sri Lanka’s accomplishments in ethical garment manufacturing are lauded in some aspects (MacCarthy and Jayarathne 2012; Loker 2011), yet challenged by others (Ruwanpura 2013, 2014a, b; Hancock et al. 2010). Ruwanpura (2012) reports how investments in improved factory working conditions were perceived to have been made primarily for the benefit of the business. She notes how, according to labour rights activists, this was to ‘show off’ to buyers and auditors and thereby present an image of caring for the labour force. Ultimately, factory managers contended there was a point at which CSR outcomes designed to benefit workers conflicted with commercial imperatives and presented a restriction on the ability of the business to service its customers. For example, a limit on working hours of female workers presented workflow difficulties in an environment of fluctuating production demands of retailers:

There are restrictions when we cannot do what we want … The negative side is if we get stuck for a style when we cannot do extra overtime and night-work and things like that. (Company D)

Such views echoed Yatawara’s (2004) observation that over-protective labour laws represent an impediment to productivity and competitiveness, which highlights the tensions at play between characterising CSR initiatives as a strategic business imperative and also as a development mechanism.

Long-Term Collaborative Relationships Mitigating Price Pressure and Risk to CSR

Our data collection was undertaken soon after the onset of the economic recession which led to obvious concerns regarding order volumes, retailer opportunism and the consequent implications for supplier ability to engage in CSR initiatives. The effects of the Western recession were referred to during interviews in terms of reduced order volumes from EU and US customers as well as an increase in payment terms, for example from 30 to 45 days. Factory managers described a typical situation of the larger the customer, the less likelihood of the supplier having sufficient power in the trading relationship to negotiate favourable terms. However, they argued that the adoption of strategic supply chain management strategies could provide a means of negotiating the threat of downward price pressure. For full package suppliers that had long-term, collaborative and technologically integrated trading relationships with their retail customers, downward price pressure had less of a negative effect as both parties were able to work together to achieve a mutually acceptable solution—while maintaining a CSR focus. One interviewee revealed that open-book accounting was used to collaboratively make improvements to reduce production cost. For example, Company A and one of its key buyers, a large US casualwear retailer, worked together to reduce cost by innovating in the assembly process or the fabric choice:

When they are asking for a discount, we give them our cost sheets: tell us where you think that we can reduce further. And they are suggesting. So when they are asking for a discount what we’ll do, we do some changes, design modifications and we say to them if we do these things we can come down the price by this much. So it’s a win–win solution, ultimately we agree. (Company A)

Open-book accounting was present in Company A only and was perceived by factory managers as a positive development that enabled buyer and supplier to work together to achieve target costs, thereby supporting relational continuity despite downward price pressure. In larger full package suppliers, a long-term vision and a partnership approach to relationship management was evidenced by long trading relationships with retailer customers (8–20 years), long-term sourcing agreements at board level, and supplier development programmes provided by key buyers that reinforced the push for continuous improvement in ethical manufacturing. Provision of technical training and collaboration on cost reduction measures was seen as having supported the long-term sustainability of the trading relationship:

They help us technically… They help us [lower the cost] … They organise training for us. It’s a big relationship, you know, we get a lot of support from the customers. (Company C)

Through industrial upgrading and the development of valuable skills, some factory managers argued that they were effectively able to pick and choose their retail customers, which facilitated some redistribution of power in the supply chain. For example, Company D, a medium-sized exporter with seven production sites in Sri Lanka, chose to discontinue supplying a large Spanish fast fashion retailer due to the latter’s shift to 90 day payment terms. Its expertise in manufacturing high quality garments in difficult-to-handle and expensive fabrics was seemingly valued by other customers and enabled it to earn a premium and become a preferred supplier for certain garments, thus partly addressing the power imbalance characteristic of buyer-driven production networks.

In contrast, interviewees observed that downward price pressure had greater impact for contract manufacturers that operated on smaller margins than full package manufacturers that provided added-value services. This is consistent with a widespread concern regarding the lack of bargaining power of the weakest members of a GPN that produce easily substitutable commodity items (see Coe et al. 2008). Although upgrading to full package production provided the opportunity for higher margins than contract manufacture, it appeared not to eliminate the presence of downward price pressure (cf. Tokatli 2013).

The Nature of Fashion Garments and Associated Retail Buying Practices in Strengthening Sustainable CSR

Sri Lanka typically competes in higher-value apparel segments rather than low-cost fast fashion (Ruwanpura and Wrigley 2011), leading to a reputation for producing quality fashion basics for mid-market retailers, such as men’s casual woven chinos and technically complex items such as bras. For core basics, product life cycles were longer than for high fashion and garment construction was relatively straightforward. The long-running nature of certain styles improved forecasting of retail demand and reduced pressure for short lead times. Factory managers contended they were better able to plan ahead and workers could achieve a higher rate of efficiency and earn production bonuses. For example, the long-running production of Gap men’s trousers enabled the operators in Company A to earn a good production bonus on top of their basic salary, overtime and attendance incentive. Key to this was the fact that long-running orders provided workers with sufficient time to reach the top of their learning curve.

Importantly, such findings contrast with Ruwanpura’s (2012) study of ethical compliance from the perspective of Sri Lankan garment workers which suggested that new designs took longer as the workers needed time to learn the style and reach optimum productivity levels. Therefore, in Ruwanpura’s findings, although production bonuses were available, the likelihood that workers would reach the level of productivity necessary to earn these was less certain for shorter style runs. Manufacturers also argued that the nature of the product partly alleviated the effect of downward price pressure, since buyers and suppliers had time to collaborate on improving efficiency for long-running orders. Finally, there was less likelihood of last minute changes to orders: with core basics, buyers were able to forecast more accurately and the consequences of under- or over-buying were less severe than for fast fashion products that may have a product life cycle of only a few weeks.

Factory managers viewed that the main issue for full package suppliers was the increase in retailer payment terms which affected business cash flow, as they had to pay for raw materials in advance. It is known that a shift to full package production may be risky for suppliers, especially during economic downturns as manufacturers become vulnerable to the whims of retail buyers keen to recoup any losses (Schrank 2004). Perhaps understandably, the Managing Director of Company G, a small domestic subcontractor, preferred to subcontract rather than export directly due to the need to maintain regular cash flow, avoid lengthy payment terms and fund upfront the purchase of fabric and trims:

We could do direct exports but since we are small here we cannot handle a customer like those customers …also we need banking facilities to do that … [It] is safer for us to do work for local companies … for these contracts for subcontractors [we get] payment 14 days after delivery. (Company G)

Written contracts of advance commitment were rare, however, long-term trading relationships of 10–20 years existed and many factory managers seemed secure in the knowledge that the relationship would continue based on the current level of full package service provided (cf. Towers and Burnes 2008):

With (retailer Z) we have a very close relationship … it will take them four or five years at least to build a relationship like this. So they are not going to move away in one month; that will not happen. (Company A)

By creating a reputation for higher-value products for brands and retailers that put a premium on supply chain compliance (Montlake 2011), Sri Lankan manufacturers argued that they naturally lean towards long-term trading relationships rather than short-term multiple sourcing ones primarily based on price. For some of the leading global brands that have suffered from ethical supply chain scandals in the past, such as Nike and Gap, sourcing from Sri Lanka seemingly manages risk, protects brand equity, and supports the development of a rationalised supply base of long-term trusted trading partners.

Supplier Development and CSR in Garment Production Networks

Sri Lankan manufacturers contended that industrial upgrading and supplier vertical integration increased visibility through the production network and hence offered greater management control of CSR. Factory managers felt that full package suppliers could offer greater assurance of CSR since sourcing and production were mostly conducted within the boundary of the firm—in our cases, all within Sri Lanka. The nature of buyer–supplier relationships in Sri Lanka is defined by the nature of the product: predominantly mid-market fashion basic, and this, in turn, was seen as supporting CSR implementation, especially for large global brands such as Gap, Nike and Marks & Spencer. Meanwhile, contract manufacturers had greater supply chain complexity than full package suppliers, as fabric sourcing was undertaken outside the firm boundaries. For the smallest companies, supply chain complexity was increased by downward price pressure and lack of scale economies: materials had to be sourced from the cheapest location which put ethical standards at risk. The Managing Director of one of the smaller contract manufacturers explained the company’s policy towards trim (i.e. thread, buttons, zips) sourcing as being primarily influenced by cost:

I can buy them in Australia imported from China cheaper than I can buy them here, and air freight them here and it’s still cheaper … It’s sad you can’t buy trims and things from the country you’re in because they’re too expensive – they’re ridiculously expensive. (Company F)

In this case, there were no CSR audit requirements beyond the first tier garment supplier, so there was no pressure on the supplier to source raw materials ethically. Similarly, unless the customer specified a nominated supplier, Company D was free to use its own sources of raw material supply and in such cases there were no demands made on the manufacturer to ensure social compliance further upstream in the supply chain. When questioned, the Director admitted:

We have not gone into that too much – checking about [upstream] compliance. Of course, I think we might have to also do it because [these] things are becoming more and more important (Company D)

As full package suppliers expand their operations into other lower labour cost countries in order to satisfy their buyers’ demands for lower unit prices, it remains to be seen how CSR implementation will be managed across national boundaries, given the inconsistency in standards.

The strategic benefits of long-term orientation and shared goals between buyers and suppliers in Sri Lanka were also confirmed by the Head of Sustainable Business for a leading UK apparel retailer: a co-defined approach to CSR that recognised the opportunity for shared learning and mutual benefit which seemingly produced better results than a coercive, compliance-based approach that stifled innovation and creativity. This is consistent with previous literature which questions the value of monitoring as a means of progressing CSR compliance, given the issues of audit methodology weaknesses, deception and cost ineffectiveness (e.g. Miller et al. 2011; ETI 2007, Usher and Newitt 2009; Barrientos and Smith 2007). The vision of some Sri Lankan factory owners and managers was perceived by the retailer as significant in progressing the CSR agenda—something consistent with Hemingway and Maclagan’s (2004) recognition of the role of managers’ personal values in implementing CSR.

To summarise, the negative influences on CSR implementation in the Sri Lankan export garment industry were identified by manufacturers as the sector characteristics of chronic price deflation and shorter product life cycles, which resulted in the development of commercial buying practices geared towards reducing cost and lead time. These were seen as potentially jeopardising suppliers’ ability to implement and uphold social compliance. The uneven power balance in the trading relationship meant that suppliers felt less able to negotiate terms with retailers; however, larger full package manufacturers that had developed long-term trading relationships with their customers appeared to benefit from greater support from retailers in terms of training provision and collaboration on cost reduction measures. The complexity of the supply network reduced visibility and sometimes resulted in inconsistent implementation of CSR where third-party suppliers of raw materials and trims further upstream from the garment manufacturer were not obliged to uphold the same ethical standards as the manufacturers. However, positive drivers were also identified by interviewees: complexity was reduced in vertically integrated manufacturers that provided a full package service, with ethical visibility improved and standards more likely to be applied consistently throughout the outer supplier layers. Larger manufacturers with superior management resources viewed themselves as better placed to implement and uphold ethical initiatives.

Amid this complex backdrop, factory managers remained conscious of the need to maintain CSR initiatives. However, such CSR activities were underpinned by an enduring awareness of the strategic CSR outcomes that would inevitably support improved productivity, output and enhanced relations with the lead retail clients (see summary in Table 2).

Table 2 Factory management narratives on the linking of CSR activities and outcomes
Full size table

Such initiatives remain persistent given the manner in which factory managers linked CSR with positive worker outcomes in terms of staff retention and productivity; but also in terms of reinforcing relationships with powerful Western retail customers that required responsiveness and product quality yet remained aware of the potential vulnerability of their retail brands to non-ethical practices at the factory level (see Table 3).

Table 3 Factory management narratives on the linking of CSR goals and outcomes for workers and retailers
Full size table

The narratives of Sri Lankan factory managers in reconciling the seemingly contradictory pressures of CSR and price competitiveness in apparel production provide important insights for practitioners as well as perhaps a more obvious academic contribution. Both Tables 2 and 3 demonstrate how CSR is not viewed simply as a cost by manufacturers, but instead acts as a means of securing continued orders from retailers concerned with ethical production and provenance. In addition, CSR is seen as contributing to developing and retaining workers with positive implications for enhancing productivity.

Given our identification of Sri Lanka as an extreme case location with an established reputation for CSR, these linkages of CSR activities to outcomes should aid apparel manufacturers located in other regions to identify strategic options for CSR repositioning—not least if they are suffering from strong price competition. However, the role of the national context—both in cultural and regulatory terms—should also be recognised in supporting the distinctive ability of Sri Lankan suppliers to successfully manufacture mid-price, mid-market apparel and effectively resist price-led competition from elsewhere. As such, many of the drivers and outcomes identified in Table 2 are likely to be underpinned by unique circumstances. From both a manufacturer and a lead retailer perspective, alternative approaches to implementing CSR at the factory level may be necessary in different cultural and institutional contexts. This requires that retail buyers, in particular, are sensitive to the idiosyncrasies of the region when considering any relocation or expansion of the geography of apparel production.

Our findings also imply further managerial implications for lead retail firms within the global apparel production network. Our own observations and interviews suggest suppliers are focused on CSR at the factory level, but that this is driven in large part by productivity, employee retention, maintaining product quality and an awareness of the lead retail firm’s brand vulnerability to any ethical shortfall. While this underlines the power of the lead retailer in the relationship, it also suggests that there is value in regularly assessing worker conditions on the factory floor. In addition, our interviews suggested where suppliers sourced from further upstream suppliers, ethical checks were not always observed. Lead retailer appraisal of these issues ‘on the ground’ would assess how embedded CSR proclamations by supply chain partners are in practice, beyond a systematic private standards ‘tick box’ exercise, and serve to further protect retail brand equity from unethical practices.

Conclusion

The power yielded by lead firms within global garment production networks potentially leads to undesirable outcomes for vulnerable workers in less developed countries (Coe and Hess 2013). While a wide range of research has explored self-imposed responsible sourcing strategies within the supply network (Hughes et al. 2007), along with some positive examples of functional and social upgrading (Tokatli 2013), there has been little understanding of how these narratives of ethical trade and CSR have been constructed and framed at the factory management level. This is the distinctive focus of our research paper. Such insights inform our understanding of ethical discourse at one nodal point of the production network, where factory managers negotiate the industry characteristics surrounding the CSR agenda alongside the ever present commercial imperatives.

Factory managers conceptualised CSR in strategic terms, underlining its capability to deliver staff retention, reputation enhancement and efficiency gains. Factory managers were clear that improving working conditions through training and employee councils attracted better workers, reduced turnover and improved productivity, which also catered to the concerns of its retail buyers who demanded product quality and some degree of brand assurance given the risks of shortfalls in ethical standards. This had implications for the labour outcomes. Measurable outcome standards by their very nature serve a purpose of presenting a positive image to outsiders and thereby lead to greater strategic benefits for the firm, as opposed to necessarily enabling worker rights. Factory managers understood CSR initiatives to be measurable standards geared towards keeping the labour force content. Respondents perceived that they were doing the best they could for workers, given the industry imperatives of cost and lead time pressure.

In conceptualising the nature and operationalization of strategic CSR, our research provides insights that are relevant both for practitioners as well as academics. For example, Tables 2 and 3 reveal how factory managers drive forward CSR initiatives which they see as responding to lead retail buyer concerns about ethical sourcing and brand reputation yet also offer a commercial logic that delivers outcomes which positively affect production efficiency and labour relations. Given that Sri Lankan apparel manufacturers have been at the forefront of developing a CSR reputation alongside commercial competitiveness, these insights should be of value to apparel producers from other regions that are keen to avoid continued descent into wholly price-based competition. As we have seen, a CSR focus provides garment manufacturers with some degree of insulation from shifting post-MFA geographies of apparel production.

A growing reputation for CSR appears to support Sri Lanka’s competitive advantage in an increasingly global production landscape. Sri Lankan manufacturers regard their ability to balance the complexity of buyer demands, the management of the workforce and CSR implementation as underpinned by sophisticated supply chain management systems, long-term non-adversarial relationships with buyers and some redistribution of power towards the supplier (based on garment specialisation and ethical credentials). However, the influence of the industry sector’s competitive context appears—to some degree—to constrain the supplier’s ability to provide positive labour outcomes for workers. Indeed, our research supports some findings of other studies (e.g. Ruwanpura 2013) that suggest a continued tension between commercial imperatives and ethical outcomes for workers.

Suppliers must balance the sometimes conflicting demands of a number of stakeholders in navigating the challenging landscape between ethical outcomes and commercial sustainability, ensuring a balance between the short and long-term (cf. Goger 2013). While there is evidence of a changing trading relationship between retail buyers and suppliers towards a partnership model of mutual gain, our findings suggest some degree of discord. Factory managers see ethical initiatives as subject to pressure from commercial buying practices which do not directly reward ethical performance. For example, although there is increasing interest from mid-market and fast fashion retailers in projects that seek to address factory productivity issues as a means of improving labour standards (ETI 2009; Primark 2009; Marks & Spencer 2010; Impactt 2012), manufacturers acknowledged this is not always followed through. Recent evidence from retailers such as Marks & Spencer and C&A, suggest that despite the best efforts of the ethical trading function to implement factory efficiency improvement programmes as a means of enhancing labour outcomes, worker benefits may be eroded as buyers recalculate the target price based on the new factory efficiency rate and demand price reductions from the supplier (Sadler and Chamberlain 2012). Admittedly beyond our immediate study, in the wake of the Rana Plaza disaster, it was alleged that retail buyers requested price reductions from Bangladeshi suppliers on the basis that the disaster had brought the country into disrepute (Smallman, 2014). Hence, there remains a tension between the retailer buying function and the CSR function. Since cost reduction and increases in productivity were seen as key dimensions in production upgrading, it becomes unreasonable to expect that production upgrading will necessarily lead to better labour outcomes (cf. Bernhardt 2013; Tokatli 2013).

This study provides important insights for academia and practice, but has certain limitations. Our data provides in-depth understanding at only one nodal point within the apparel GPN—the factory management level. Future studies could extend our research to form a larger programme of work to trace these narratives throughout the global apparel production network—to include consumers, NGOs, regulators, retail buyers and workers. Such insights would develop our understanding of the expectation, interpretation, implementation and practical realisation of CSR holistically within the network. In doing so, it would identify key differences and inconsistencies between actors in the production and consumption process that would aid us in further exploring the degree to which the global and retailer-driven nature of apparel production facilitates or restricts ethically and socially responsible practices.