2, June 2005 181–200

Globalization and Paradoxes of Ethical Transnational Production: Code of Conduct in a Chinese Workplace

Department of Politics and International Relations, Lancaster University, Lancaster, UK 2 School of Social Science, Hong Kong University of Science and Technology, Hong Kong
Neo-liberal economic globalization and acceleration in information and communication technologies have strengthened the competitiveness of transnational corporations by making it easier for them to stretch their global commodity chains to developing countries. Such control from a distance threatens national states’ capacity to regulate TNCs and their impact upon environment, labour and human rights in developing countries. This has prompted anti-TNC sentiments and campaigns that challenge brand-named TNCs on labour exploitation in the south. These changes have triggered discussions on setting labour standards in the international arena. With the WTO distancing itself from the ‘social clause’, this has prompted the TNCs, NGOs and other stakeholders to self-regulate by introducing codes of conduct. Concentrating on the clothing industry, this paper examines the emergence of company and multi-stakeholder codes of conduct that require the involvement of subcontractors and NGOs in developing countries. The moving of ethical codes to developing countries condenses and reproduces, under the twin pressures of competitiveness and social auditing, the paradoxes of ethical transnational production in the internal organizations of local firms. These paradoxes are demonstrated from a case study of a workplace in China. The paper ends by outlining three paradoxes and commenting on the development of a managerialist ‘audit culture’ in workplace practices as commodification of ethical code.

Audit culture, Civic activism, Codes of conduct, Commodity chains, Corporate social responsibility, Chinese workplace.

Introduction There is a growing trend for transnational corporations (TNCs) to introduce codes of conduct regarding labour conditions in their global activities. This paper discusses this phenomenon in general terms and then identifies some paradoxes in the implementation of such codes through a case study of one corporation in southern China. We first suggest that neo-liberal economic globalization and the acceleration in information and communication technologies have enhanced the potential competitiveness of TNCs by enabling them to stretch their commodity chains and to integrate developing countries more completely into their operations. This poses interesting questions about national states’ capacities to
E-mail address: © 2005 the Editors and W. S. Maney & Son Ltd DOI: 10.1179/102452905X45427


N.-L. SUM and P. NGAI

regulate TNC activities and their impact upon the environment, labour and human rights – especially as the TNCs often exercise their powers at a distance and international bodies such as the WTO are less committed to ‘social clauses’ (see the second section below). We then explore how the transnational nature of these commodity chains and the dominance of TNCs in their organization have prompted general disquiet and resistance over labour exploitation in the south and anti-TNC sentiments in particular. This has triggered discussions on setting labour standards through TNCs and/or through multi-stakeholder codes of conduct (see the third section below). Turning to the clothing industry, we then examine this ‘code rush’ and the emergence of an arena for global business governance that is mediated by the UN, the ILO, NGOs, TNCs and states. The emergence of this governance arena enables large corporations to adopt their own codes and/or multi-stakeholder ones. One possibility, which is the focus of our attention in this paper, is for TNCs to adopt in-house codes that contain the main provisions of a multi-stakeholder code (see the fourth section below). Next, we explore the implementation of this code in a case study in southern China. This case reveals how the transfer of this code from the transnational arena to a specific local site condenses and reproduces in the internal organizations of local firms some emerging paradoxes of ethical transnational production. This occurs under the twin pressures of competitiveness and social auditing (see the fifth section below). The paper concludes by identifying three such paradoxes and suggests that the managerialist audit culture that is being adopted in workplace practices in developing countries can be understood as a commodification of ethical codes (see the sixth section below).

Globalization and Transnational Production Economic globalization has accelerated transnational production through the acceleration of innovation in information and communication technologies, inexpensive transportation logistics and the often-asymmetric liberalization of world trade under the WTO and other multilateral arrangements. Together with the celebration of neo-liberalism as the highest principle of rationality and basis for action, this has strengthened the competitiveness of TNCs by making it easier for them to stretch their global commodity chains to developing countries and thereby exploit new cost advantages (Dicken, 2003; Gereffi 1999; Gereffi and Korseniewicz 1994). This involves these corporations collaborating flexibly with networks of suppliers in developing countries to produce goods mainly sold in the developed world. This new mode of transnational coordination rests on TNCs’ ability to control production (e.g. delivery dates, quality standards, design specifications) over long distances without exercising ownership. Various consumer goods, including clothing, footwear and toys, have commodity chains in which control resides with brand name producers such as LeviStrauss, The GAP, Reebok, Nike, Mattel and the like (Frenkel 2001; Jenkins 2001; Jenkins et al. 2002). Such control from a distance threatens national states’ capacity to regulate TNCs and their impact upon the environment, labour and human rights in developing countries. This has prompted anti-TNC sentiments that can be seen in anti-globalization movements, quasi-organized ‘corporate watch’ efforts and anti-sweatshop campaigns (Klein 2000) mainly initiated in developed countries. They challenge TNCs on labour-related issues such as low wage levels, deteriorating working conditions, forced overtime, child



labour, unsafe working conditions, and lack of the right to free association (organizing unions) in developing countries (Taplin and Winterton 1997; Waddington 1999). In particular, such criticisms have been directed against major brand-name producers by consumer associations, non-government organizations (NGOs), international trade unions, and the media concerning labour issues in these countries. Such public allegations of labour exploitation have triggered a discussion, especially from the early 1990s onwards, on setting labour standards in the international arena.1 The World Trade Organization (WTO) and the International Labour Organization (ILO) are important international economic institutional players in these discussions. In April 1994, during the final stages of GATT/WTO’s Uruguay Round, there was intense debate over a proposal to link labour standards with trade – the ‘social clause’ – as part of the international trade regime (Wilkinson 2002). Seeing the ‘social clause’ as ‘disguised protectionism’ by developed countries, negotiators from developing countries opposed its inclusion in the Uruguay Round. The issue of a social clause protecting workers was raised again in the WTO’s first Ministerial Meeting in Singapore in December 1996 and, as before, was deeply divisive. Nonetheless, the meeting did articulate the WTO’s and ILO’s positions on labour standards. The WTO renewed its commitment to observe labour standards and the ILO was accepted as the appropriate body to deal with these standards. Reflecting its neo-liberal orientation, however, the WTO only committed itself to continue its ‘existing collaboration’ with the ILO – whilst leaving unclear the nature of such ‘collaboration’ (Wilkinson 2002: 217). This ambiguity was reinforced in subsequent WTO summit meetings in Seattle and Doha in 1999 and 2001, respectively. This distancing from the ‘social clause’ and the resultant marginalization of labour compared with the golden years of the (international) postwar settlement has prompted alternative ways of introducing labour standards into the international agenda. There is a shift from harder methods, such as enforcing ILO conventions through national courts, to softer methods (e.g. codes of conduct) suggested by non-state actors such as TNCs, NGOs and other stakeholders. This occurs as global commodity chains in the consumption sector come under the increasing scrutiny of consumer activism from developed countries.

Global Commodity Chains and Civic Activism With the drive to cut costs and maintain competitiveness, brand-named TNCs such as LeviStrauss, Nike and The GAP have subcontracted their production processes to developing countries in search of cheaper labour and lower environmental standards. Gereffi’s work on global commodity chains is very helpful in understanding such developments in global capitalism (Gereffi 1999; Gereffi and Korseniewicz 1994). For him, each commodity chain has three main dimensions: (1) the structure of inputs and outputs that links products and services in a sequence so that each activity adds value in turn; (2) the territorial aspects of the networks among enterprises – networks may be spatially dispersed or concentrated; and (3) the structure of governance, with its vertical and horizontal relationships of power and authority, which determines the flow and allocation of resources in the chain. This paper is mostly concerned with the third dimension, especially in the case of buyer-driven chains (see Figure 1). The latter are chains in which brand-name retailers and trading companies are dominant buyers and collaborate flexibly with networks of suppliers in developing countries to produce consumer goods for sale in developed markets.


N.-L. SUM and P. NGAI

Fig. 1. An ideal-typical buyer-driven global commodity chain: directions of supply and control (context: neo-liberal globalization and debates on ‘corporate social responsibility’).

As dominant buyers, TNCs in buyer-driven chains are able to control production (e.g. delivery dates, quality standards, design specifications) over long distances without exercising ownership. Non-ownership of suppliers also enables them to collaborate flexibly on their terms with subcontractors (e.g. global sourcing allows them to order flexibly to suit changing consumer demand and flexible labour and overtime enables them to adapt to fluctuating demand and fashions). A variety of labour-intensive consumer goods, including clothing and footwear, are characterized by buyer-driven commodity chains in which control resides with brand-name producers such as Levi-Strauss, The GAP, Reebok and Nike (Barrientos 2002: 64–68; Jenkins 2001). These kinds of control from a distance and in-built flexibility are some of the many factors undermining the capacity of national states to regulate TNCs and to limit their adverse impact upon labour and human rights in developing countries.



These developments have contributed in turn to the rise of anti-TNC sentiments that fuelled anti-globalization movements, quasi-organized ‘corporate watch’ efforts and antisweatshop campaigns (Klein, 2000). Civic activism along these lines and their related NGOs began to operate on the global level and, indeed, to cooperate on different scales, to check TNC activities or hold them more accountable. Based on guidelines formulated by ILO and OECD, the TNCs have responded by formulating their own codes of conduct. These corporate codes of conduct are written statements of ethical standards that TNCs claim to adopt and that apply to their suppliers and trade partners as well as themselves. They often contain a variety of clauses measured against ILO conventions, in particular those concerned with the respect of human rights at work. As documents initiated by corporations, codes of conduct are not legally binding and standardized in the same way as governments are bound by the international conventions that they sign and ratify. They amount to what is called ‘soft regulation’, which consists of ‘declarations, resolutions, guidelines, principles and other high-level statements by groups of states such as the UN, ILO and OECD that are neither strictly binding norms nor ephemeral political promises’ (ICHRP 2002: 73). The proliferation of multilateral and firm-level codes of conduct was a response to the growing multilateral challenge from a wide range of actors on different scales to the deteriorating conditions in the labour market and labour process in developing countries and to the limits on workers’ efforts to organize to improve these conditions (Taplin and Winterton 1997; Waddington 1999). As dominant buyers in the chain, these TNCs are few in number – this makes them susceptible to political pressure and charges of condoning sweatshop conditions. They are especially vulnerable to charges leveled against them by consumer associations, NGOs, international trade unions, religious organizations, student coalitions and the media concerning labour issues in industrial zones in Asia, Latin America and the United States. Civic activism of this kind can be seen in a number of anti-sweatshop campaigns worldwide. Well-known campaign and civic groups that target the garment industry include Clean Clothes Campaign (Europe), Maquila Solidarity Network (Canada), United Students Against Sweatshops (the USA), Oxfam’s Clothes Code Campaign, and the Interfaith Centre on Corporate Responsibility (the USA) (Lipschutz 2002; 2003). They participate in the new politics of consumption to end sweatshop abuses and urge TNCs to take on more responsibility for the workers in their employment; for consumers to look behind the label and think about the products before they buy; and in general, the concern for global and local inequality. For example, the Clean Clothes Campaign uses a mix of activities to promote political consumerism and civic activism. These include alternative fashion shows, emails to corporations, e-newsletters, on-line dissemination of research reports, fair trade evaluations in retail stores, in-house training of voluntary workers and demonstrations. These kind of campaigns, together with the media’s strategies of ‘naming and shaming’ highly visible companies and celebrities, has created a transnational resistance space against corporate power. For example, the Washington Post in 1992 exposed Levi-Strauss’s exploitation of Chinese indentured labour to make jeans on the Island of Saipan (Sajhau 1997). This was followed by a series of other cases, including the CBS reporting in 1996 on sweatshop conditions in a factory in Vietnam that supplied Nike products. This strategy of subverting high-profile symbols was also directed against celebrities. For example, Michael Jordan, who signed a $20 million multi-year endorsement deal with Nike in 1996, was condemned by the Daily News in New York as the ‘Pontius Pilate of American sports’ (Dwyer 1996).


N.-L. SUM and P. NGAI

The ‘Code Rush’ and the Emergence of Global Business Governance The emergence of this transnational resistance towards corporate excesses has triggered self-regulation based on the drawing up of in-house codes of conduct (Culter et al. 1998; Haufler 2001). Since the early 1990s, there has been a proliferation of company codes initiated by the brand-name traders and corporations. Levi-Strauss was among the first to draw up a code on labour standards, entitled Business Partner Terms of Engagement and Guideline for Country Selection, in 1991 (revised in 1994). It covered issues such as working conditions, minimum wages, environmental standards and child labour (see Table 1). It broke new ground in terms of the transnational scope of such codes by covering 600 suppliers and subcontractors in developing countries.2 Nike also unveiled a corporate code of conduct in the same year; but the external auditors were subsequently criticized, including PriceWaterhouseCoopers, for weak implementation (O’Rouke 2003). Seeing Nike’s problems, Reebok improved its practices by instituting a policy of not buying soccer balls from factories using child labour. In Europe, C&A has been the focus of many actions in the Netherlands and the UK since 1990, and in Belgium, Germany and France since 1995. Otto Versand was been the first action target of the German Clean Clothes Campaign since 1995, and developed codes to be sent to its suppliers. In 1999, the Italian newspaper Corriera della Sera revealed that Benetton was using child labour in Turkey and a code of conduct was adopted in the same year. By 2003, it was estimated there were about 10,000 codes that were accompanied by varying management systems for implementation (Global Unions 2003). For illustrative purposes, the following are three well-known codes of conduct, namely Levi-Strauss, Nike and The GAP (see Table 1). A detailed discussion of these codes is beyond the remit of this paper. However, outlining these codes does serve to illustrate the ‘code rush’ by TNCs since the 1990s. They contribute towards the making of the discourse on ‘ethical corporations’ that is being thickened by the emergence of narratives on Corporate Social Responsibility (CSR). Many TNCs have begun to pledge that they will assume social responsibilities in their transnational activities. Business schools and business consultancies started to construct ideas such as ‘corporate citizenship’ and tie ‘doing good’ to profit since the mid-1990s. Concepts such as capital are narrated in terms of companies’ reputation – reflected in the new notion of ‘reputational capital’ – that is essential for building corporate names and hence profit. This linkage between ‘capital’ and ‘reputation’ is also narrated as part of ethical corporate strategy in a global age (Bossone 1999; Fombrun 1996; Jackson 2004; King and MacKinnon 2001). This revival of talk on business ethics and its role in capital accumulation is narrated mainly in an instrumental manner, i.e. corporate culture and ethical obligations are seen as ‘good for business’. In spite (and perhaps because) of this company ‘code rush’ and discourses on CSR, trade unions, consumer groups, NGOs and international organizations continued to criticize these ‘soft’ self-regulating codes as public relations tools or window-dressing exercises to deflect consumer/community inquiries about workplace conditions and to project an image of a ‘caring company’ when in reality business carried on as usual. These complaints from trade unions, NGOs and UN-related organizations have pushed some TNCs to move the CSR agenda forward by entering into negotiation with international labour unions, NGOs and trade/industry organizations in formulating industry-wide or multi-stakeholder codes of conduct. This change in the balance of social forces has led to an emphasis on multi-stakeholder codes (Utting 2003) and the emergence of an arena for global business governance. Multi-stakeholder codes involve dialogues and negotiations between

GLOBALIZATION AND ETHICAL PARADOXES TABLE 1 Labour standards in three TNC codes of conduct Selected labour standards in each code Child labour Levi-Strauss (business partner terms of engagement) Use of child labour is not permissible Levi- Strauss encourages apprenticeship programmes. No use of prison or forced labour. Nike (Code of conduct) In accordance with local regulations.


The GAP (sourcing guidelines for vendor selection) Subject to all applicable laws, rules and regulations.

Forced labour

Working conditions

Health and safety

Forced labour – prison or otherwise – must not be used at any stage of production. • Working hours in In accordance with accordance with local trade legislation local standards. as regards minimum • Maximum: 60 hours wages, overtime, leave, a week. public holidays. Business partners must In accordance with provide good health local regulations. and safety conditions.


Prevailing industry standards in the surrounding community.

Implementation of codes and evaluation

Employees must not be exposed to conditions dangerous to their health. Internal methods: Internal methods: Internal methods: • Evaluation carried • For each plant, a • The plants of out by a Levirepresentative of subcontractors are Strauss team. Nike is responsible selected and • A 20-page evaluation for the day-to-day inspected by document reviews application of the representative of the practices in Memorandum. management of The production and • Obligations on GAP and/or by a working conditions. suppliers/ commercial • Workers are also subcontractors to representative directly concerned retain and to make responsible for and unannounced available to Nike all purchasing. visits to enterprises documentation External methods: are sometimes concerning ethical • An independent carried out. criteria contained monitoring External methods: in the Memorandum. working group set • The assessment is • The Memorandum up in cooperation also based on is posted up in all the with NGOs and opinions about the plants of Nike’s unions subcontractors.



Selected labour standards in each code Levi-Strauss (business partner terms of engagement) enterprise by trade unions officials and local associations. Nike (Code of conduct) The GAP (sourcing guidelines for vendor selection) • Work towards the improvement of codes and propose solutions.


Basic ethical principles of the corporation

• Nike reserves the right at all time to have an independent monitoring carried out of the plants of suppliers/subcontractors. Termination of trade In the event of nonrelations with respect of the partners who do not Memorandum, Nike respect guidelines. may insist that the situation is redressed or it may cancel its trade commitments. Business partners Principles that govern should respect a code the conduct of of ethics that includes: business • Respect for include trust, employees, suppliers, teamwork, honesty clients, consumers and mutual respect. and shareholders; Nike expects • Respect for basic compliance with the human rights. same principles by its trading partners

Termination of trade relations with suppliers/ subcontractors who do not respect the code of conduct.

Learning from NGOs, human rights leaders, development experts, factory personnel and from garment workers. Open dialogue generates fresh ideas and stakeholders learn about The GAP’s commitment to creating sustainable progress in the workplace.

Source: Adapted from information derived from Sajhau (1997).

governments, TNCs, trade unions and NGOs. They work together to improve the quality of standards, reporting and auditing procedures of corporate subcontracting activities. Altogether, there are 11 major codes. Here, we will concentrate on four prominent ones related to the garment/sportswear industry (see Table 2). Let us start with the Fair Labor Association code in the USA. In 1996, the Clinton Administration, in response to the sweatshop movement, launched the Apparel Industry Partnership (AIP), which consisted of 18 garment makers, unions and NGOs. Controversy arose when several unions and NGOs withdrew because they disagreed with the design of the monitoring system, unionization and the concept of the living wage. After their withdrawal, the AIP created a code of conduct with the Fair Labor Association as a permanent organization to implement it. The code stipulates a minimum wage, worker’s age, working hours and so on (see Table 2). This led to a split among the organizations. Two apparel firms withdrew, feeling the provisions were too strict. The unions

GLOBALIZATION AND ETHICAL PARADOXES TABLE 2 Main features of selected multi-stakeholder codes of conduct Main features Year of foundation Fair Labour Association (FLA)


Workers Rights Social Accountability Clean Clothes Consortium (WRC) International (SAI) Campaign (CCC) (SA8000) US-American Students’ Organization 2000. Members: 110 universities, USAS. US-American consumer organization 1997. Board: 2 NGOs, 1 lawyer, 3 companies. Dutch Research Institute SOMO 1990. More than 250 NGOs and unions in 14 CCCs in 12 European countries. Freedom of association Collective bargaining No forced labour No child labour No discrimination Living wage Occupational health and safety Hours of work Establishment of Employment Relationship (reference to ILO)

Clinton Administration 1996. Membership/ Members: governance 13 companies, 6 NGOs, 3 universities, 1 chair. Labour Freedom of standards association Collective bargaining No forced labour No child labour (except 14 years if legal) No discrimination Legal minimum wage including benefits Occupational health and safety Hours of work (exceptions from ILO – 48 hours plus 12 hours overtime/ week possible) Overtime pay possible at normal rate No harassment or abuse Methods/ Independent instruments verification, certification by third-party auditing firms that cover the whole supply chain.

Freedom of association Collective bargaining No forced labour No child labour No discrimination Living wage Occupational health and safety Hours of work Establishment of employment relation No harassment or abuse Women’s rights

Freedom of association Collective bargaining No forced labour No child labour No discrimination Living wage Occupational health and safety Hours of work Management systems (reference to ILO and to UN Human Rights Conventions)

• More complaint-driven. • No certification; but spot-check inspections in whole supply chain. • Networking North-South campaigns.

SA 8000 Standard, • Code of Labour independent Practice. verification, • Independent certification of verification and producers/ suppliers. certification. • North-South . solidarity. • Information of consumers (sic)



Main features Scope of application Fair Labour Association (FLA) The whole supply chain of garment products. Workers Rights Social Accountability Clean Clothes Consortium (WRC) International (SAI) Campaign (CCC) (SA8000) The whole supply chain of garment products. The whole supply chain of garment products. • Brand certification. • Verification institution with board of 50% company and 50% NGO/ union representatives, contracts with audit company possible.

Monitoring/ • verification • •

• Reporting/ disclosure •

Complaints/ • appeals/ corrective actions •

Emphasis on factory/farm, but also recommendations for supply chains. Brand • Brand orientation. • Factory/farm certification. • Investigations certification. Annual rather than • 9 SAI accredited monitoring monitoring or audit companies and certification. certification verify producers As from second • Spot-check according to year onwards, inspections. SA8000 guidance internal • Local NGOs in document monitoring of all joint investigative • SA8000 certificate facilities in teams. is valid for 3 years. supply chain, independent verification of 5% of all facilities. 13 accredited audit companies. Internal and • WRC verification • Certified factory independent and remediation listed on website. monitoring reports go to all • Audit reports go reports go to the parties involved. to SAI and to the FLA staff. • Full disclosure of companies. The FLA all reports from • Other parties can evaluates audits, investigations. get audit reports jointly develops after having signed remediation a confidentiality plan, and then agreement. publishes summary reports of remediation results. Third party • The WRC Agency Complaints/appeals complaints go examines the can go to: to FLA, which complaints. • the management then to informs • Appeals, initiates of the factory, those who filed and corrective • the certifying body the complaints. action in • the accreditation appeals about cooperation with agency, the outcome of labour or corrective action. ganizations.

• After each inspection audit reports go to all parties. • The verification body informs the public.

Complaints/ appeals go to the foundation, which then initiates corrective action.



Main features Fair Labour Association (FLA) • Also possible complaints direct to brand companies. Administration costs are paid by annual company fees and contributions by universities. Workers Rights Social Accountability Clean Clothes Consortium (WRC) International (SAI) Campaign (CCC) (SA8000) each of them being able to initiate corrective action. • Administrative costs are paid by the fees of member universities: 1% of annual licensing revenues. • Funds from foundation. Annual fees for member companies vary between US$ 1,500–15,000 according to revenues. • Participating companies and organizations contribute to the administrative costs of the independent verification body. • Pilot projects mainly paid for by companies.

Costs/ financing

Source: Adapted from Wick (2003: 68–79).

involved in AIP, including UNITE (a major US garment and textile labour federation), also withdrew, charging that some provisions were too lenient. The NGOs in the AIP also split, with the Interfaith Center on Corporate Responsibility leaving the original Clinton initiative. Nonetheless, the FLA advances a monitoring and certification system (see Table 2). Current members of the FLA include some major branded apparel and sports shoe companies (e.g. Levi-Strauss, Nike, Adidas, Reebok, Liz Claiborne and Polo Ralph Lauren), as well as 131 universities whose licensed apparel is produced by US manufacturers. FLA members are moving toward implementing a monitoring and certification system based on a monitoring sample of 30% of the company’s suppliers. It also announced a US non-profit monitoring organization (Verité) and three commercial compliance certification firms (Intertek Testing Services, Merchandise Testing Labs and Global Standards) as its ‘external’ monitors to carry out code compliance verification for members. Monitors’ reports will be made available to FLA staff and summary reports of remedies published. As the FLA were approving companies and accrediting auditors for its external monitoring programme, two US-based initiatives were also moving ahead in certifying factories. First, the Worker Rights Consortium (WRC) is an initiative by the United Students Against Sweatshops. When several trade unions and NGOs withdrew from the AIP, claiming that the provisions were too weak (no standard for living wage), some student activists from the United Students Against Sweatshops sided with them and established the WRC in 2000. With support from the AFL-CIO and the Union of Needleworkers, Industrial and Textile Employees (UNITE), the WRC draws on its social logic and advocates a living wage, independent unions, women’s rights, and unannounced factory investigations and full disclosure of factory conditions (see Table 2) (Gereffi et al. 2001: 4; Jenkins 2002: 22–25). It


N.-L. SUM and P. NGAI

does not certify factories or brands; instead, it carries out investigation (both pro-active and in response to worker and third-party complaints) of factories. Investigations are carried out by joint teams made up of WRC members and local NGO and labour organizations. It makes all its investigation reports public and develops an on-line database for public dissemination. Second, Social Accountability 8000 (SA8000) was created in 1997 by the Council on Economic Priorities, a US NGO which specializes in corporate responsibility. It outlines the workplace standard for producers and suppliers on the commodity chain. It is administered by Social Accountability International (SAI), with an advisory board (see Table 2). Brand name companies become ‘SA800 Signatories’ and they are audited by one of its nine accredited Certification Auditors. Apart from accrediting auditing firms, the SAI also provides training for SAI auditors. The audit reports go to companies and other parties after signing a confidentiality agreement. The SA8000 system differs from the FLA in a number of regards, most notably on issues of wages and certification. SA8000 requires factories to pay workers a ‘basic needs’ wage and not a prevailing wage, which is a weaker requirement. It also certifies manufacturing facilities and not brands or retailers. Concurrent with these developments in the USA, there were several initiatives in Europe. The Dutch Clean Clothes Campaign (CCC) exists in 12 European countries. The resulting activities are coordinated by an international secretariat in Amsterdam that organizes a network of over 250 unions and NGOs. The campaign aims to improve labour conditions in the garment and sportswear industry by informing consumers, influencing companies and supporting workers’ organizations. Its Code of Labour Practices for the Apparel Industry including Sportswear was adopted in February 1998 (see Table 2). Some TNCs’ in-house codes contain the main provisions of the CCC code and the relevant verifying body can be composed of companies, industry associations and employers’ associations, on the one hand (50%), and trade union organizations and NGOs, on the other (50%). After each inspection, audit reports go to all parties. The CCC also lobbied the European Parliament and the EU Commission concerning their recent initiatives on codes of conduct and corporate social responsibility (Wick 2003: 27–30). The same emphasis on codes of conduct and corporate social responsibility can be found in the UN Global Compact project. In Summer 2000, UN Secretary General Kofi Annan officially launched the UN Global Compact, which, in addition to environment protection and human rights principles, invites companies, large and small, to promote the principles in the ILO Declaration. The Compact is not a regulatory instrument but it does seek to promote institutional learning and social dialogue between business, unions and NGOs. Within the Global Compact, corporations are called upon to support the nine basic principles derived from the universal declaration of human rights, the ILO declaration on fundamental principles and rights at work and the 1995 Copenhagen social summit and the Rio Declaration of the UN conference on environment and development. It became the UN’s attempt to deal with CSR and entered into partnership with TNCs. By October 2001, approximately 400 companies from 30 countries had associated themselves with the Global Compact.3 The co-existence of two types of code – TNC and multi-stakeholder – has two important effects. First, it has created an arena for global business governance in which actors as varied as the UN, the ILO, TNCs, industry associations, employers’ organizations, NGOs, labour unions and states all participate (albeit with unequal impact) in the shaping of the international political economy of production. Second, it enables some TNCs to create room for manoeuvre in this ethical minefield, with its double potential for them to gain



prestige (and protection) by adopting their own codes as well as using their involvement in multi-stakeholder codes to integrate themselves into the state–TNC–NGO circuit. There are two distinct types of TNC–NGO links. In the direct-link type, TNCs, even if they have their own in-house code, subscribe to one of the multi-stakeholder codes (e.g. Nike and Levi-Strauss have both adopted the FLA code and its monitoring system). The indirect-link type involves the TNCs’ adoption of in-house codes that contain the main provisions of a particular multi-stakeholder code backed up by the commissioning of particular NGO project teams to monitor their compliance with the code (e.g. the case study in the fifth section, below, illustrates this type). Whether direct or indirect, these codes are implemented in specific sites through the setting up of monitoring and verification systems at the local level. This typically involves the participation of local NGOs in pilot schemes that monitor selected supply factories in the commodity chains. For example, the FLA in March 2001 designated the Guatemalan Commission for the Monitoring of Code of Conduct (COVERCO) and the Bangladesh-based NGO, Phulki, for monitoring projects. They were the first southern NGOs to carry out external monitoring for companies affiliated with the FLA. Similarly, the WRC contacted NGOs in a number of countries (e.g. India and Mexico) concerning their participation in joint investigations. The various national Clean Clothes Campaign groups also involved southern NGOs in piloting similar projects (SOMO 2001). In 2003, the Swedish Clean Clothes Campaign in conjunction with a local NGO – the Hong Kong Christian Association – finished a pilot study on monitoring labour standards in China. In general, the implementation of these codes has prompted various concerns in developing countries. There is growing recognition that these codes have been driven primarily by actors and interests in developed countries that have paid insufficient attention to the culture and concerns of workers in developing countries (e.g. female workers having to work alongside men and the lack of a place to pray). These workers are often portrayed as ‘victims’ to be studied by outside organizations rather than as potential agents of change. It has also been questioned whether monitoring constitutes a public relations gimmick that may usurp the regulatory role of national governments and unions by privatizing the enforcement of labour standards through codes of conduct. Furthermore, some worry that smaller firms in developing countries will be displaced because they are less able to comply with the codes than larger ones (Utting 2001: 20–21). These are important questions that reflect the struggles and worries that are involved when codes are transferred to developing countries. Paradoxes of Ethical Transnational Production in a Chinese Workplace Most studies in this area concentrate on how these codes affect workers’ wages and conditions, labour organization, and the forms and extent of unionization (Ascoly and Zeldenrust 2003; Lum 2003; Scharage 2004). This paper does not aim to cover this familiar ground but, instead, concentrates on how codes create and reproduce the paradoxes of transnational production in the internal organization of subcontracting firms in developing countries. This case study enables us to identify at factory level three paradoxes in the adoption and implementation of a code of conduct, which we will discuss in our concluding remarks. Bur first we present our fieldwork in China.4 This case was based on empirical research conducted between 2002 and 2003 in a large Hong Kong-invested company located in an export-processing zone of Dongguan


N.-L. SUM and P. NGAI

in southern China. This company, which we shall call SSG, produced sportswear for a European brand-name corporation that adopted an in-house code of conduct in 1995.5 This contained the main provisions of labour standards in the Conduct for the Apparel Industry including Sportswear formulated by the Clean Clothes Campaign (see Table 2 and the fourth section, above). The management of SSG had some knowledge of company codes and international labour standards such as SA8000 (see Table 2). Five years later, in 2000, SSG was also supposed to introduce a new European Corporation Code to its production workers. Of some 800 employees in SSG, about 700 were production workers, and most of them were rural migrant workers mainly from Hunan, Hubei and Sichuan provinces. Almost all the workers had received at least primary education, 80% had junior secondary schooling, and a few had senior secondary schooling and technical tertiary education. They should have had no problem in understanding the provisions of the Code. During the three visits to the firm in 2002 and 2003, a two-page Chinese summary of the Code was displayed on the wall of each shop floor. The font was so small that workers could hardly read the content. Even if it was legible, the clauses lacked details concerning child labour, discrimination and overtime pay/work. No full Code was posted on the noticeboard in the workplace, canteen or dormitories. The use of a two-page summary indicated that SSG was not serious in promoting the Code. However, the management insisted that every worker had been provided with a full written copy (in Chinese) of the European Corporation Code when they signed or renewed their contracts. They also claimed they had introduced the Code to the workers during morning briefing meetings. Yet all workers interviewed denied that they had received a written copy of the Code. They also said that they were coached to provide standard answers to the code inspectors or auditors, especially on the issues of wages, working hours and rest days. Thus, when the issue of overtime work was raised, the workers replied as follows:
We work eight hours each day. No compulsory overtime work. We have Sunday off. We get 1.5 times pay for evening work, double for weekends and triple for public holidays.

The ‘willingness’ to repeat this standardized answer was secured from the workers through a tactical quid pro quo. The company told the workers that giving the ‘wrong’ answers about the Code would undermine the company’s reputation and, hence, the firm’s prospects for production orders and the workers’ chances of work. A social instrument that aims to inform and empower workers about their rights has been turned into an economic narrative that interpellates ‘workers as (loyal) employees’ who should not upset the managerial apple cart both for their own sake and that of the firm. In this regard, workers’ consent to management hegemony was traded for a little tale of profit and opportunity. The tactical alliance between the workers and the management in the face of the Code inspectors reflects how this little tale suppresses (without resolving) the contradiction between the social and the economic in the workplace. In most cases, the economic seemed to dominate the everyday practices of the factory. Further interviews with some workers revealed that they had to work up to 12 to 14 hours a day, and they were only paid an overtime bonus for evening and Sunday work. The workers said that, for normal work, the hourly rate was RMB 2.5, and for overtime work, only RMB 3.2 rather than the 3.75 RMB (1.5 times the normal rate) specified in the Code. Regarding the practice of payment, the workers were provided with wage slips, the format and details of which they did not understand. Despite this, they were still expected to show the slips to the inspectors for the purpose of auditing.



In our interviews with management representatives, the production manager directly criticized the Code of Conduct. She described it as a ‘paradoxical’ and ‘hypocritical’ act by the TNCs to assuage the ‘inner sins’ of the ‘rich western countries’. She stressed that, when pressed to meet production orders by the TNCs, she had to give up ‘labour rights’ and violate the Code in favour of pushing workers to work faster and longer. Severe global competition for low-cost, just-in-time and fashion-conscious products imposed structural limits on the implementation of the Code of Conduct, revealing its ‘hypocritical’ nature.6 The ability to comply was undermined by TNC sourcing practices such as cutting order times, reducing prices and changing fashion constantly. Likewise, the company director summed up this paradox by pointing out that production order prices continuously dropped but the requirement for labour codes had become more stringent. He also remarked that the code required by the European Corporation involved a change in its ‘market behavior’ – it demanded huge investment to meet the labour standard in order to secure production orders from European and American retailers. He also complained that TNCs were unwilling to share this investment cost so that the whole burden fell on local producers. ‘To survive in the world market, we have no choice but to do as we are instructed,’ said the director. In this regard, from the company’s viewpoint, the Code of Conduct had less to do with protecting labour rights than with its own and the buyer’s business strategies. For the subcontractor and the TNC alike, the Code has become a marketing device to enhance their respective positions in a global commodity chain. This is a far cry from the original intention behind ‘corporate social responsibility’, at least as seen by consumer groups and NGOs, namely, to entrench and protect labour rights. In July 2002, an inspector from the European Corporation produced a report that directly and forcefully identified SSG’s non-conformities with the Code. In order to protect its ‘reputational capital’, the Corporation formulated a corrective action plan for SSG to address its non-conformities within a specified timeframe that included child or juvenile workers as well as wages and contractual arrangements. In September 2002, a Social Compliance Officer (SCO) who would be solely responsible for implementing the Code in the workplace was employed. The responsibility of this SCO was ambiguous because she had to report her activities to the European Corporation as well as to SSG’s management. The major role of this SCO, a local graduate, was to set up a task force to implement the Code standards in SSG. Under the instruction of the task force, she devoted all her efforts to building up filing and documentation systems. During the visits, she was keen to show the inspectors freshly constructed files and records in order to prove the institutionalization of labour standards. Thus there were files on Code of Conduct Training Procedures, Code of Conduct Test Records, New Workers’ Assessment Forms, Workers’ Complaint and Handling Records, Guidelines on Admission and Dismissal, and the like. Being pro-management and proud of this documentation system, she said:
We now learn how to build up a filing and reporting system. We learn this from other companies. The code monitors came for a few hours only for every inspection. They like tables, figures and data, don’t they?

This way of implementing a code of conduct was further applauded and confirmed by professional auditors, who played an indispensable part in advising the Code’s implementation in SSG as well as other companies. The auditors’ role was particularly important in institutionalisation, insofar as almost all their recommendations reiterated the importance of setting-up registers and filing systems. They taught the company how to revise the Code provisions, and to design recording systems on code training, wages and working hours,


N.-L. SUM and P. NGAI

contracts, and non-discrimination employment policies, hoping to satisfy the corrective action plan. A two-page summary of the Code in Chinese was then distributed to all workers who were asked to sign their new contract and the copies of their signature were properly filed. Training sessions on the Code provisions were organized for new workers who had to pass a test covering, most importantly, the Code provisions, basic information on the company and their contract provisions. Sample test papers were shown to the team and all had a full mark. However, interviews with workers continued to indicate that the workers did not actually understand the reasons for having such a Code and its benefits for themselves, although some did mention ‘human rights’ and ‘labour rights’ such as nondiscrimination and no compulsory work. In fact, except for the new workers, training sessions were only provided to the upper and middle managerial staff, such as supervisors and line leaders, who in turn disseminated the information on the Code to their work units or production lines. This process of institutionalization was further crystallized in the redrafting of a new contract and a new wage register by the task force. In March 2003, the whole workforce was provided with a new contract that was re-written according to the China Labour Law, and the workers again had to sign to confirm receipt of the copy of their contract. A new wage register system had also been set up. The wage calculation sheets distinguished normal workdays, rest days and holidays. The register could now show that the net monthly wages of the workers for a 40-hour week corresponded to the legal monthly minimum wage. The workers could also receive a wage slip each month. The slip, however, did not explicitly list normal and overtime hours worked in a month. Nor did it detail the piece-rate and the payment of statutory holidays. Thus the workers were still unable to work out by themselves whether they had received the correct wages. Recruitment policy was also reformulated by implementing a written nondiscrimination policy in the summary of the Code, in particular with respect to age and gender. The workers, however, did not fully understand the issue of discrimination but noted that it was common knowledge that new jobs were only open to the 18–25 age group. The deposit system had been discontinued: all the deposits were refunded to the workers who had signed the receipt and new workers hired were no longer required to leave a deposit. The system of salary deductions for disciplinary reasons had also been discontinued. However, on closer inspection, it had been re-categorized as ‘performance reward’, which meant that disciplinary fines would still be deducted from it. Thus, the contradictions between the formal requirements of the Code and actual practices in the Chinese workplace seemed to have been resolved through this rapid process of ‘soft’ institutionalization. Documentation, registers and filing were seen as a must in order to implement the Code, and at least as having records to demonstrate their compliance with it and Chinese Labour Law. The trouble of unannounced inspections could be easily dealt with if piles of files were available for show. In this regard, the implementation of the Code of Conduct, if it really happened, was actually a process of institutionalization of paperwork with little improvement in working conditions, as shown by SSG. It is, after all, a windowdressing practice without real commitment from either local or transnational companies.

Concluding Remarks The increasing inability of nation-states to control TNCs’ activities has promoted the rise of ‘corporate watch’ efforts and anti-sweatshop campaigns. Transnational civic activism of



this kind began to operate on the global level in order to hold these brand-named corporations more accountable through corporate and multi-stakeholder codes of conduct. At first sight, these codes of conduct as initiated by TNCs and NGOs seem to open up an arena that holds out the promise of rebalancing the social obligations of dominant market players in transnational production. There are numerous attempts by TNCs and NGOs to move these codes to developing countries. Our case study in China adds to the critical literature on implementation of these codes in two ways. First, it reveals three paradoxes regarding business ethics and just-in-time competition in transnational production. Second, it highlights the difficulties in introducing ethics in a new-liberal environment and this leads to a commodification of the codes of conduct. The first main conclusion from our case study is the three paradoxes regarding transnational ethical production. First, the pressures from just-in-time, low-cost and fashion-conscious production for the world market structurally constrain the forms and extent of compliance with the Code. The firm faces the paradox of lowering production order prices at the same time as adopting more elaborate labour codes. This paradoxical combination of pressures to engage in business ethics and maintain just-in-time competitiveness has led the subcontractor to prioritize the market logic over ethical principles in the workplace. Thus, the subcontractor reacts in terms of its market logic and sees the Code as an opportunity to enhance ‘reputational capital’ and thereby promote its competitiveness in the global market. ‘Advancement of labour rights’ is translated into ‘compliance with labour code’ as business strategy. Second, the introduction of an ethical code for worker protection leads paradoxically to the firm using the Code to encourage workers to cooperate with the management to avoid the loss of contracts and hence future employment opportunities. This is achieved by narrating a little management tale that links workers’ employment with the firm’s profits and opportunities. This tale not only encourages the workers to enter into tactical alliance with the management in the face of social auditing; but also suppresses (without resolving) the increasing contradiction between the social and the economic in the workplace. Third, the incorporation of the labour code as part of its business strategy has lead the firm to build elaborate managerial and audit/documentation systems to defend the subcontractor against charges of infringing the Code. This institutionalization of ‘soft’ procedures and systems produces the paradoxical result that more effort goes into paperwork than into actual advancement of labour rights protection. Based on the advice by auditors, management showcases its files, registers and records as proof of compliance. The second main conclusion concerns the commodification of ethical codes in workplace practices. Our case study suggests that there is a tendency towards the managerialization of the social agenda so that codes of conduct and CSR become marketing devices and commodities that mask the contradiction between global capitalism and social inequality in the workplace in developing countries. This managerialization/economization of codes of conduct is what leads to the above-mentioned focus on paperwork, records and documents. This raises the question of whether such initiatives have created a ‘market for ethics’, a ‘market in ethics’ or an ‘ethics for the market’. The first market would involve the recognition by TNCs, service-oriented NGOs, monitors, and southern subcontractors that ethics can become the basis for accumulating reputational capital. Thus they seek to turn politically-sensitive global labour questions into profitable managerial-ethical knowledge and auditing practices such as constructing codes, monitoring performance, writing reports, setting up task forces, filling in forms and keeping files/records/documents. This in turn creates a ‘market in ethics’, that is, opportunities for business (e.g. auditors, lawyers, ethical officers) and for service-oriented NGOs (to demonstrate their effectiveness and raise funds).


N.-L. SUM and P. NGAI

Together, the primacy of managerialized ethics and the routinized practices of the emerging code-of-conduct industry are giving rise to a neo-liberal globalization with an audit (and not a human) face. These managerialization and auditization7 practices regarding codes may foster reputational/material benefits for the TNCs, service-oriented NGOs, monitors, and subcontractors. But this occurs at the expense of diverting attention from the workers’ welfare and undermines the original motivation behind labour codes, namely, to develop and institutionalize an ‘ethics for the market’. This would advance the rights of workers in developing countries on an equal footing with those of transnational capital. On the basis of the present study, this aim of civic activism in developed countries is still far from being realized in developing countries. Acknowledgements We are grateful to the two anonymous reviewers for their insightful and helpful comments and to Bob Jessop, Jeff Henderson, Julie Froud and Alvin So for making this article possible. The usual disclaimer applies. Notes
1 The UN proposed an international code in 1980 but it never reached consensus and finally failed after a decade of unsuccessful negotiations. The ILO has had a Declaration of Principle on Multinational Enterprise and Social Policy since 1977 but it failed to endorse an agreed international code of conduct for all industries and companies. Women Working Worldwide (2000) Company Codes of Conduct, available at http://www.poptel. There is mounting criticism of Kofi Annan’s Global Compact on the grounds that the United Nation has provided moral legitimacy for TNCs. It has thereby helped to ‘bluewash’ the TNCs, which means the use of the reputation of the United Nations to present a more humanitarian image for the profit-oriented TNCs. The reason why China is chosen is because China is increasingly being seen as the ‘factory of the world’. With China joining the WTO since December 2001, it has become a major player in the international garment industry. In March 2002, the Financial Times predicted that after the 2004 phase-out of apparel export quotas currently in place under the Multi-Fibre Agreement, global apparel chains will stretch to China so that it will become the producer of half the world’s garments by 2005 (Brandsma 2002). The Code of the European Corporation required all its ‘contractors, their sub-contractors, principal suppliers and licensees to observe the standards when producing or distributing products or components of products for XXX’. Moreover, the Corporation demanded from all its suppliers that they comply with applicable labour laws in the countries where they operate. The production manager often referred to the Corporation Code as ‘paradoxical’ and ‘hypocritical’. The idea of ‘auditization’ stems mainly from Michael Power (1997).

2 3



6 7

Ascoly, N. and Zeldenrust, I. (2003) Challenges in China: Experiences from Two CCC Pilot Projects on Monitoring and Verification of Code Compliance (Amsterdam, SOMO).



Barrientos, S. (2002) Mapping codes through the value chain: from researcher to detective, in R. Jenkins, R. Pearson and J. Seyfang (eds) Corporate Responsibility and Labour Rights: Codes of Conduct in the Global Economy, pp. 61–78 (London, Earthscan). Bossone, B. (1999) The role of trust in financial sector development, World Bank working paper no. 2200, 1 October (New York, World Bank). Brandsma, E. (2002) Made in China, Ethical Corporation, 9, pp. 34–37. Culter, C., Haufler, V. and Porters, T. (eds) (1998) Private Authority and International Affairs (Albany, State University of New York Press). Dicken, P. (2003) Global Shift: Reshaping the Global Economic Map in the 21st Century, 4th edition (London, Sage). Dwyer, T. (1996) These sneakers really stink, Daily News (New York), 18 July. Fombrun, C. (1996) Reputation: Realizing Value from the Corporate Image (Boston, MA, Harvard Business School Press). Frenkel, S. J. (2001) Globalization, athletic footwear commodity chains and employment relations in China, Organization Studies, 22(4), pp. 531–562. Gereffi, G. (1999) International trade and industrial upgrading in the apparel commodity chain, Journal of International Economics, 48, pp. 37–70. Gereffi, G. and Korseniewicz, M. (eds) (1994) Commodity Chain and Global Capitalism (Westport, CT, Greenwood Press). Gereffi, G., Gracia-Johnson, R. and Sassser, E. (2001) The NGO-industrial complex, available at: (accessed on 4 December 2003). Global Unions (2003) Institutional investors driving worker exploitation, paper presented at the World Economic Forum, Davos, available at: 2003/NewsCo. . ./InstitutionalInvestors.htm (accessed 23 July 2003). Haufler, V. (2001) A Public Role for the Private Sector: Industry Self-regulation in the Global Economy (Washington, DC, Carnegie Endowment for International Peace). ICHRP (International Council on Human Rights Policy) (2002) Beyond Voluntarism: Human Rights and the Developing International Legal Obligations of Companies (Versoix, Switzerland, ICHRP). Jackson, K. (2004) Building Reputational Capital (New York, Oxford University Press). Jenkins, R. (2001) Corporate codes of conduct: self-regulation in a global economy, Technology, Business and Society Programme Paper No. 2, April (New York, United Nations Research Institute on Social Development). Jenkins, R., Pearson, R., Seyfang, G. and Howitt, R. (2002) Corporate Responsibility and Labour Rights: Codes of Conduct in the Global Economy (London, Earthscan). Klein, N. (2000) No Logo (London, Flamingo). King, D. and MacKinnon, A. (2001) Who cares? Community perceptions in the marketing of corporate citizenship, Journal of Corporate Citizenship, 1/2, pp. 37–53. Lipschutz, R. D. (2002) Doing well by doing good? Transnational regulatory campaigns, social activism and impacts on state sovereignty, in J. Montgomery and N. Glazer (eds) Challenges to Sovereignty: How Governments Respond, pp. 291–320 (New Brunswick, NJ, Transaction). Lipschutz, R. D. (2003) Regulation for the rest of us? Global social activism, corporate citizenship, and the disappearance of the political, available at: (accessed 4 March 2004). Lum, T. (2003) Workplace codes of conduct in China and related labour conditions, Report for Congress, 23 April, available at: (accessed 4 February 2004). O’Rouke, D. (2003) Monitoring the monitors: a critique of PriceWaterhouseCoopers’ labor monitoring, available at: (accessed 23 July 2003). Power, M. (1997) Audit Society: Rituals of Verification (Oxford, Oxford University Press).


N.-L. SUM and P. NGAI

Sajhau, J.-P. (1997) Business ethics in the textile, clothing and footwear (TCF) industries: codes of conduct, ILO Bulletin, no. II-9, June. Scharage, E. J. (2004) Promoting international worker rights through private voluntary initiatives: public relations or public policy?, report to the US Department of State on behalf of Iowa Center for Human Rights, available at: (accessed 4 February 2004). SOMO (Centre for Research on Multinational Corporations) (2001) The changing terrain in the codes of conduct debate, available at: (accessed 2 February 2004). Taplin, I. M. and Winterton, J. (eds) (1997) Rethinking Global Production (Aldershot, Ashgate). Waddington, J. (ed.) (1999) Globalization and Patterns of Labour Resistance (London and New York, Mansell). Wick, I. (2003) Workers’ Tool or PR Ploy?, 3rd edition (Bonn, Friedrich Ebert Stiftung). Wilkinson, R. (2002) Peripheralizing labour: the ILO, WTO and the completion of the Bretton Woods project, in J. Harrod and R. O’Brien (eds) Globalizing Unions? Theory and Strategies of Organized Labour in the Global Political Economy, pp. 204–220 (London, Routledge). Utting, P. (2001) Regulating business via multistakeholder initiatives: a preliminary assessment, available at: (accessed 4 August, 2003). Utting, P. (2003) NGO involvement in promoting corporate responsibility: the role of ‘multistakeholder initiatives’, available at: Peter.doc (accessed 22 August 2003).

Sign up to vote on this title
UsefulNot useful