ILO is a specialized agency of the United Nations

273rd Session
Geneva, November 1998

Working Party on the Social Dimensions of the Liberalization of International Trade



Executive summary

Overview of global developments and Office activities
concerning codes of conduct, social labelling and other
private sector initiatives addressing labour issues

I. Introduction

  1. Institutional context
  2. Scope and methodology
  3. Sources of private initiatives

II. General context

  1. Private actors and partnerships
  2. Private/public partnerships
  3. International standardizing activities and private sector initiatives

III. Codes of conduct

  1. Development of codes with labour provisions
  2. General trends
  3. Operational codes
  4. Model codes
  5. Content
  6. Implementation of codes

IV. Social labelling

  1. Structure and operation
  2. Content
  3. Impact
  4. Final considerations

V. Investor initiatives

  1. Screening of investment funds
  2. Shareholder initiatives

VI. Activities of the Office

  1. Enterprise and Cooperative Development Department (ENTREPRISE)
  2. Bureau for Employers' Activities (ACT/EMP)
  3. Bureau for Workers' Activities (ACTRAV)
  4. Multinational Enterprises Branch (MULTI)
  5. Working Conditions and Environment Department (TRAVAIL)
  6. Sectoral Activities Department (SECTOR)
  7. International Institute for Labour Studies (INST)
  8. Office for Inter-Organization Relations (REL/INT)
  9. Office of the Legal Adviser (JUR)

VII. Private initiatives and the ILO: Issues for discussion

  1. Expectations concerning the ILO and its possible role
  2. Possible implications for the achievement of the ILO's objectives
  3. Possible positions and actions and follow-up on the discussion

Appendix: Tables

I. Introduction

1. Institutional context

1.  At the 270th (November 1997) Session of the Governing Body, the Working Party on the Social Dimensions of the Liberalization of International Trade requested the Office to prepare a broad overview of global developments and Office activities concerning codes of conduct, social labelling programmes, and other private sector initiatives relevant to labour issues. The Working Party agreed on the desirability of including an analysis of the origin, range, and types of practices, the characteristics differentiating the various initiatives, and the merits and demerits of differing approaches. At both the 270th and 271st Sessions, a range of positive and negative views on the effects and uses of codes of conduct and social labelling approaches was expressed. It was, however, agreed that further discussion and understanding of the phenomena would help define the possible level of ILO involvement.

2.  The initiatives at issue are distinguished by their non-governmental, or private sector, character. With the increasing globalization of private sector economic activity, growing networks of enterprises operate across national boundaries in an astonishing array of contractual, equity and joint venture arrangements. The result, in which economic activities are beyond the reach of any one national system, has been described as a "mismatch of regulatory scope and actual economic structure".(1)  This limits further the action, both of sovereign States, and also of private actors with crucial interests, including workers, local employers' organizations, and others. At the same time, enterprises confronting the pressure of globalized and/or regionalized competition have undergone dramatic restructuring which, besides its positive returns, has resulted in downsizing, layoffs and other negative social returns. Many commentators believe the changes in enterprise environment which fuel social initiatives are here to stay.

3.  In today's globalized economic environment, the emergence of a social role for enterprise corresponds to a notable shift in power from the provident State to privatized and multinational enterprise, concomitant with a waning influence of governments and local workers' organizations in the mobile marketplace. Under such circumstances, the growth of pluralistic societies and independent media have contributed to increasing public demands for accountability for the social and environmental impact of enterprise operations. In general, these demands arise through the efforts of non-governmental organizations (NGOs), and the continual evolution of strategic alliances of local-level workers' organizations and international trade secretariats (ITSs). Enterprises are facing growing pressure to be accountable for non-financial benchmarks in what has been termed the "triple bottom line", a reference to economic, social and environmental performance.(2)  Enterprise is responding, at the level of the multinational and individual enterprise, as well as in enterprise associations of businesses and industries and, to some extent, employers' organizations, particularly at the national level. Partnerships between enterprise, workers, and/or NGOs, sometimes facilitated by governments and intergovernmental organizations (IGOs), are negotiating selected labour practices to improve, and devising methods for external monitoring of enterprise progress. The results are particularly critical for developing country enterprises, especially those of small and medium scale, which must find cost-effective ways to make the technical and managerial changes needed to obtain access to foreign markets or large international buyers who impose social requirements.(3) 

4.  As the number of voluntary initiatives multiplies, so do calls for ILO action, as well as calls to refrain from action, in this area. On a regular and steadily increasing basis, the Office receives requests by enterprises and their partners contemplating private initiatives to address labour practices. Some seek information, others request review and comment on proposed codes of conduct, and still others wish to have advice on social inspection methods, for internal application or for use as professional auditors. In a hearing at the European Parliament on 2 September 1998, an ITS representative, in the name of workers, called for the ILO to establish procedures for the accreditation of professional auditors to certify enterprise performance, keep a register of accredited auditors, and develop training programmes, on a self-financed basis from outside funds. An industry representative at the hearing stated that enterprises themselves should be responsible for the implementation of such initiatives as voluntary codes of conduct, but acknowledged that ILO training programmes and independent auditors could enhance transparency and accountability. The International Organization of Employers (IOE) itself has begun to consider whether, and how, it should assist enterprises on these matters.(4) 

5.  The Director-General noted in his Report to the 1994 Session of the International Labour Conference that these phenomena confront the ILO with the need to adjust its approach to its constituents;(5)  Traditionally, the approach has relied heavily on commitments and actions taken by member States. Nonetheless, although the ILO Constitution, Conventions and Recommendations directly apply to member States only, the principles underlying international labour standards reflect social goals which are defined and achieved only with the active cooperation of the non-governmental social partners. That cooperation is made all the more necessary by the increasingly supra-national framework of economic activity. Some 20 years after its adoption and despite long-standing efforts within the UN and elsewhere, the ILO Tripartite Declaration concerning Multinational Enterprises and Social Policy stands as the only international consensus among governments and interested non-governmental parties on social policies and measures to be taken, on a voluntary basis, by enterprise actors in the private sphere. The Tripartite Declaration is intended to guide the voluntary commitments of multinational enterprises and home and host country workers' and employers' organizations, as well as governments.(6) 

2. Scope and methodology

6.  The first part of this paper highlights trends and issues emerging in private sector initiatives involving voluntary methods of addressing labour practices in enterprise operations.(7)  The term "private sector initiatives" refers to actions which, while not enforced by law, may seek to enhance or supplement behaviour required by law. The term "labour practices" encompasses, in principle, all conditions of labour and rights at work within the scope of the ILO mandate. As a practical matter, however, this overview focuses on practices relevant to fundamental labour principles and rights,(8)  wages, working hours, occupational health and safety, employment security, and job training. The initiatives reviewed in this paper include codes of conduct (section III), social labelling programmes (section IV), and investor initiatives (section V). Definitions appear at the start of discussion of each initiative. Following a review of outside initiatives, section VI of the paper reviews Office activities and publications relevant to enterprise social responsibility and, in particular, to the private sector initiatives discussed. Section VII contains final questions for consideration. Any specific reference to an enterprise or programme in this paper does not constitute an endorsement by the Organization of any product, service, enterprise, or programme.

7.  To prepare this overview, the Office conducted a review of primary literature produced by enterprises and other actors involved in private sector initiatives, and of available secondary sources. The secondary sources included studies, published surveys, and commentaries on private sector initiatives, and were derived from a wide variety of sources, including employers' and workers' organizations, enterprises, governments, intergovernmental organizations (IGOs), non-governmental organizations (NGOs), and professional academic and research institutions. In addition, interviews were conducted with a broad range of officials of relevant programmes and services in the Office and with representative workers' and employers' organizations. Compilation and analysis of data and literature were facilitated by coordination with ILO field offices and Office research initiatives in progress, and in particular the research programme on multinational enterprise (MNE) codes.(9)  Contacts were also made regarding research initiatives on related matters within the International Chamber of Commerce, the Organization for Economic Cooperation and Development (OECD), the European Commission, the European Parliament, various UN agencies including UNCTAD and UNEP, national government agencies, and various private research institutes.

8.  Attention was paid during research to the need for a balanced database of information in terms of representativity of activities and interests by region, size and sector of enterprise, and the varying perspectives of governments, workers and employers. Nonetheless, the database of available information, particularly from Internet sources, reflected a predominance of sources based in North America and Europe, although some significant data were available from Asia, Latin America and Africa. The inevitable gaps in information, which could be corrected through field and/or survey research, are noted where pertinent as fertile areas for further research and, as much as possible, were taken into account when describing current trends and drawing up final questions for consideration.

9.  Preparation of this paper has demonstrated the "work in progress" status which characterizes the field of voluntary social initiatives of enterprise. Major disparities persist in definitions and terminology, standardization of data sets, units and criteria for analysis, and methodologies for research and evaluation. Significant work therefore remains to be done in developing and applying standardized criteria and methods for research and analysis in this field. Further conceptual work could be aided by case study research, and collection and analysis of more reliable and comparable statistical and other data.

3. Sources of private initiatives

10.  Private sector initiatives generally spring from the concept of enterprise social responsibility, which has emerged in various forms in recent decades. In 1965, the Delhi Declaration, developed at an international seminar in India on "Social Responsibility of Business", articulated a prototype model of what has evolved as "stakeholder theory". In the United States, theorists emphasized the need for enterprise to accept the community as a stakeholder in governance while, in Japan, enterprises developed a sense of linkage to national values, community life, and the environment.

11.  Desire to add value to the enterprise is a key factor in the development of voluntary initiatives. The need to preserve or legitimize a reputable public image, including the reputation of brand names, has fostered such initiatives among sectors active in outsourcing consumer goods internationally and those with labour-intensive operations, at all stages from production to retail. Producers of capital goods and intermediary products, who do not independently respond to consumer demands, may be pressed by intermediary buyers to satisfy the demands of consumers. In addition, evidence that good corporate behaviour may enhance rather than diminish financial performance, has tended to reinforce moral incentives for enterprise to act as "good citizens".

12.  Voluntary initiatives may serve enterprise interests by displaying efforts to improve workplace conditions, thus helping to defend against potential, or actual, consumer boycotts as well as formal accusations of unacceptable or illegal business practices. Such initiatives may also obviate the need for government regulation by demonstrating that industry best practice satisfies the public interest. Indeed, in some cases, governments may encourage enterprise initiatives as a substitute for social regulation of trade or business. Developing countries may view such initiatives as a way to leverage limited public resources. For exporting countries, the initiatives may be welcomed as improving labour practices and a nation's public image for trade and other purposes, while debate continues over whether other considerations, such as protectionist instincts, may contribute to some importing government efforts to encourage voluntary initiatives. There are also questions relating to the impact of such initiatives on international labour standards, which are addressed subsequently in this paper.

II. General context

13.  This section presents a general context for understanding private sector initiatives, including the identity, interaction, and strategies of key actors and partnerships involved. Discussion of specific tools used by these actors, including codes of conduct, social labelling, and shareholder actions, is found infra sections III-V.

1. Private actors and partnerships

14.  A diverse range of actors participates in private sector initiatives across global, regional, national and local levels. Traditional actors include the enterprise and its closest "stakeholders":(10)  its employees (and their representative workers' and employers' organizations), investors, consumers, and partners in contract or equity relationships. Activity on the part of enterprise associations, NGOs and other private groups has recently increased, along with new coalitions between these more recent stakeholders and the traditional actors. Some cases feature government and IGO partners as well. Partners in these new coalitions may share very similar interests or may constitute ad-hoc platforms of common interest only for the sake of the goal at hand.

15.  Initiatives led by enterprises or enterprise associations appear to be the most rapidly proliferating; these vary widely in membership, activities, and priorities. Efforts are being developed by individual enterprises, business associations, sectorally identified industry associations, and employers' organizations across diverse economies. Large, well-established individual enterprises initiate social programmes on their own; medium and small enterprises tend to form sectoral, national, cross-sector and/or cross-national coalitions. New associations, focused solely on social agendas, are emerging while some established enterprise associations provide their members with guidance on social issues as well as their traditional financially oriented services.

16.  Workers' organizations are integrating enterprise social initiatives more often into their methods of representation in an economic environment subject to increasing globalization and regionalization. Their aim is to enhance or supplement, rather than replace, results of consultations and collective bargaining processes. Workers' organizations are applying such strategic approaches within specific enterprises as well as in sectoral, cross-sectoral, and multi-sectoral enterprise groupings. Worker campaigns and media initiatives have taken up private sector initiatives as objectives to address specific concerns. Local and international unions have incorporated the tools of enterprise social initiatives into their policy-making, strategy-building, worker education, and outreach programmes. Participation of international trade secretariats (ITSs) has increased through an evolving type of international "solidarity negotiation" with multinational enterprises (MNEs) and across commodity chains and sectors, with varying degrees of involvement by local unions in negotiation and implementation (see infra section III). Workers' organizations increasingly seek out coalitions with investors, NGOs, and governments, reflecting the growing sense that trade unions need consumer, investor and other pressure to gain leverage in negotiations with MNEs and also SMEs operating in worldwide economic operations.

17.  Non-governmental organizations and coalitions cover the range of collaborative and combative methods to encourage labour-related private initiatives. The collaborative methods include awards programmes or public recognition for best practices, "dialogue" with specific enterprises, and development of codes, labels, and other initiatives discussed infra. One of the earliest successful NGO coalitions, in the 1980s, brought Del Monte operations in the Philippines to negotiate monitoring and reporting arrangements with the Swiss-based Migros cooperative food chain and its constituency, consumer groups and other NGOs in Switzerland. Some NGOs provide direct services to those affected by enterprise labour practices, particularly in developing countries.

18.  NGOs may also use methods invoking external pressure, such as campaigns to expose poor labour practices, including boycotts or threats of boycotts. A well-known and long-standing debate persists on the efficacy of such methods. Some argue that boycotts hurt the workers they are intended to help, while others believe that, at times, negative publicity operates to complement the collaborative efforts of others in developing codes, labels, and other initiatives (see infra sections III and V). NGOs also take legal action against particular enterprises, a tactic which appears to draw attention to labour practices but may not result in desired improvements or compensation for alleged harms.(11)  In 1998, the Clean Clothes Campaign (CCC), operating as an expanding, European-based network of NGO coalitions seeking improved labour practices in apparel and sportswear supply chains, orchestrated a series of well-publicized cases against apparel and sportswear MNEs before the self-appointed "People's Tribunal"; adverse "judgements" were entered against all named enterprises after a review of lengthy factual "briefs" and testimonial "evidence" heard by private individuals acting as "judges".

19.  Professional consultants, auditors, and educational enterprises appear to be seizing the opportunity to play a role in the trend toward socially responsible business. Very few have any experience or background in labour standards or labour relations. These enterprises, which sell their services, include associations focusing specifically on socially responsible business, traditional accounting firms expanding into social accounting services, specialized social accounting organizations, inspection and monitoring firms embracing social as well as technical standards, and universities. The various services they provide cover: specialized information resources, including surveys on enterprise practices and guides on best practices; training and education for employees in socially responsible practices and internal monitoring and reporting of labour practices; external services in social accounting and inspection; and accreditation of aspiring social accountants and inspectors. Such services are provided on self-defined bases, given the lack of standardization of criteria and methodology for accounting, inspection, and accreditation. Among educational efforts, one university offers a Master of Science course in "Responsibility and Business Practice", others have special centres for social accounting research, and many provide a steady stream of opportunities to attend conferences and summer programmes on theory and practice of social responsibility in enterprise.

2. Private/public partnerships

20.  The factors and incentives leading to the emergence of a social role for enterprise, described supra section I.3, is reflected in the interplay between private sector initiatives and the role of the public sector in setting social policy and addressing social needs. The new public/private partnerships are distinct from public sector initiatives which employ purely public tools. The latter are beyond the scope of this overview.(12) 

21.  Voluntary enterprise partnerships with governments have collaborated for some decades, particularly in developed countries, in such joint efforts as technical codes of practice addressing safety and health issues. Enterprise may sometimes make use of governmentally provided information resources, including guides in code development, and governmentally sponsored conferences and Best Practices award or ranking programmes. In some cases, enterprises, workers, and/or NGOs operate in coalition in their own territory with the help of government financing. Some cooperative arrangements in exporting countries are also facilitated by the governments of importing countries. In certain developing and transition economies, enterprise and government collaborate in inspection and other ways that leverage limited public resources.

22.  Voluntary enterprise initiatives with IGOs appear to be increasing in number and scope, building on established tools as well as new collaborative methods. Such initiatives may involve operational programmes, or agreements to follow guidelines on enterprise social policies or technical standards affecting conditions of work, health and safety. Two sets of guidelines on enterprise social policies exist at the intergovernmental level: the ILO's Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy ("Tripartite Declaration"), and the OECD Guidelines on Multinational Enterprises. The Tripartite Declaration, adopted in 1977, applies globally to governments, MNEs, employers and workers. Lists of ILO Conventions and Recommendations particularly relevant to enterprise operations are annexed to the Tripartite Declaration. The implementation of the Tripartite Declaration is reviewed through periodic surveys and a procedure permitting requests for interpretation in specific cases. The system as a whole offers an international framework for the development of voluntary labour-related commitments in codes of conduct and other social initiatives. In contrast, the OECD Guidelines on Multinational Enterprises apply only to enterprises operating within OECD countries. The OECD Guidelines are interpreted through a clarification procedure, under which the OECD seeks to interpret the Guidelines consistently with provisions and interpretations given in the Tripartite Declaration. The outcome of several pending proposals for other social guidelines, including the UNCTAD Draft Code of Conduct on Transnational Corporations and proposed labour clauses in the draft OECD Multilateral Agreement on Investment, is unclear, as is that of the WTO Working Group on Trade and Investment.

23.  Private enterprise also engages in voluntary initiatives with IGOs on matters involving technical standards relevant to working conditions.(13)  Within the WTO, non-governmental as well as governmental standardizing bodies may accept the commitments relating to international trade found in the Code of Good Practice annexed to the Agreement on Technical Barriers to Trade (TBT).(14)  Other IGOs have sponsored a range of technical standards, some of which are implemented through voluntary industry practice and others through the monitoring of member governments.(15)  Of particular note is a framework for assessing the social performance of industries piloted by UNIDO with the textile, clothing, footwear and leather industry (TFC) in several developing countries.(16) 

3. International standardizing activities and private sector initiatives

24.  Some voluntary commitments of enterprise are developed through the activities of international standardizing bodies, particularly the International Organization for Standardization (ISO), a private organization composed of national standards institutes or organizations from some 120 countries.(17)  ISO has developed more than 11,000 international uniform standards to facilitate international exchange of goods and services, primarily in technical fields. In recent years, ISO has developed two series of generic standards applicable to a wide range of industries and services: in 1989, a quality assurance systems management (ISO 9000) series, and, in 1996, an environmental systems management (ISO 14000) series.(18)  Enterprises on a worldwide scale have sought third-party certification of conformity with the two systems management standards, but some enterprises in developing countries have reported limitations in access to recognized certification services. ISO itself does not verify that its standards are being implemented properly by users; rather, ISO national members, or independent ("third party") certification bodies registered with the national standards organizations, inspect and issue certificates of "conformity assessment" with particular standards. Despite the fact that ISO does not certify or police its standards, ISO does influence conformity through guidance documents and training activities. Its published guidelines address procedures for selection and accreditation of certifying bodies, establishment of criteria for certification, and acceptable methods for enterprises to use in publicizing conformity with ISO standards.

25.  Although no ISO standards directly address the social or labour field, the ISO model is relevant, in certain respects, to the development of codes and social labels addressing labour practices which are the subject of this paper. Under ISO's methods of operation, actors other than governments participate in developing standards, conducting or financing assessment of compliance, and providing technical assistance, particularly to standardizing bodies in developing countries. Recent work to develop international standards on eco-labelling, primarily in response to the calls for expansion of environmental labelling schemes adopted at the 1992 UN Conference on Environment and Development (UNCED), is noted in section IV infra.(19)  In 1997, ISO decided not to launch the development of a management system standard for occupational health and safety,(20)  an area in which the ILO itself is currently active (see infra section VI). However, without ISO involvement, some non-governmental entities have sought to adapt ISO-style operations to social standards for enterprise. A US-based research institute, the Council for Economic Priorities (CEP), has launched a generic management system standard for enterprise, with a third-party certification process, called "Social Accountability 8000" (SA8000). The standard is based in part on ILO Conventions and Recommendations and inspired by the ISO model.(21)  In addition, a Swiss NGO, Bread for All, has suggested an ISO 21000 series for social management, a proposal which has not been endorsed by ISO.(22) 

III. Codes of conduct

26.  For the purposes of this paper, the term "code of conduct" refers to a written policy, or statement of principles, intended to serve as the basis for a commitment to particular enterprise conduct.(23)  By their very nature, voluntary codes contain commitments often made in response to market incentives with no legal or regulatory compulsion. However, as public statements, codes usually are considered to have legal implications under laws generally regulating enterprise representations, advertising and, in cases of joint enterprise action, anti-competition.

27.  The International Organization of Employers (IOE) estimates that 80 per cent of codes of conduct fall into the category of general business ethics with no implementation methods. This paper focuses only on codes of conduct that address labour practices, including forced or child labour, employment discrimination, freedom of association and collective bargaining, level of wages, occupational health and safety, and other workplace-related issues. This section reviews key trends in the development, content and implementation of those codes. Many of the codes reviewed in this section were adopted by an enterprise with the intention of being applied internationally; indeed, some of the codes are used as "sourcing guidelines" to specify requirements for the workplace conduct of business partners with which the enterprise is engaged. In such cases, the code of, for example, an international buyer or retailer is applied to workers in supply (or "value") chains who are not necessarily employed directly by the enterprise adopting the code.

1. Development of codes with labour provisions

28.  Codes of conduct that address labour practices have become a key element in the debate over improving the conditions of workers worldwide. Modern codes of conduct originated in the early twentieth century, with model codes on advertising and marketing practices developed by the International Chamber of Commerce in the late 1930s. During the 1960s and 1970s, labour provisions in enterprise codes of conduct were primarily directed at internal management practices,(24)  and the first major third-party "subscription" codes for enterprises developed in the United States.(25) 

29.  The late 1980s and 1990s have seen a rapid proliferation in use of codes of conduct with several new trends. First, the application of traditional, and typically MNE, codes to headquarters only or, at most, to wholly owned foreign subsidiaries, has expanded. As public pressure for enterprise accountability increases, enterprises feel more vulnerable to the labour practices of their business partners in the commodity or service chain. The "self-applied" model has been replaced by the development or application of codes in a growing variety of business partnerships involving more than a single enterprise. Codes now apply to joint ventures, licensees, and other contractual relationships resulting from the chain of transactions necessary for the products and services purchased through foreign direct investment. Concomitantly, an increasing number of actors are participating in code initiatives with enterprise, including in particular workers' organizations, industry associations, NGOs and other private groups.(26) 

30.  Another striking feature is the prominent role that such codes now play in internal enterprise decision-making. As noted by one commentator, "previously the codes were domiciled in the corporate affairs and shareholder relation departments, and were not distributed widely to suppliers. Within the past year, however, the issues have now become compelling enough for United States companies that they assign higher-level managers with direct functional responsibility for sourcing and supplier relationship ... Discussions over implementable codes of conduct are now being held outside and within corporate boardrooms ...".(27)  In part, media interest and the resulting reactions of shareholders, consumers and other stakeholders have prompted this enterprise response.

2. General trends

31.  There is no precise information on the extent to which enterprises around the world have adopted codes of conduct with labour-related provisions in their operations -- domestic or foreign.(28)  Available information focuses on the significant number of large enterprises and various types of enterprise associations that have done so across the commodity or service chain of international operations. Information on domestic operations mostly appears to be limited to enterprises in developed economies. Most surveys of codes available to date are specific to particular enterprises in developed economies, to only a few labour issues, or both.(29)  Case-study approaches by sector or issue exist, with notable existing contributions and further work still to come from the ILO.(30)

32.  Codes of conduct with labour-related provisions are known to exist in nearly all 22 ILO sectors of activity. Sectoral participation in codes, however, varies widely. Sectors that deal directly in consumer products appear more conducive to their development, including textiles, clothing, leather and footwear (TCF), commerce (retailers and home manufacturers of consumer products), food and beverage, and the chemical and toy industries. The chemical and forestry industries have proportionately high representation, focusing largely on HSE operations. Emerging sectors include hotel, financial services, telecommunications, high technology, agriculture, and oil and gas industries.

33.  Two general types of codes exist. Operational codes apply directly as commitments to specific conduct within enterprises or their partners, or through subscription systems sponsored by third parties (typically enterprise associations or coalitions of enterprise, trade union, and/or NGOs) to which enterprises commit themselves, either on their own behalf and/or for their business partners. "Subscription codes" may involve monitoring and/or reporting systems to be performed by subscribers or by the external code sponsors themselves. Model codes are issued by enterprise associations, trade unions, NGOs, and/or governments for others to use as a basis in developing their own codes. Although model codes do not operate directly within enterprise or through enterprise subscriptions, model codes may be reflected, in whole or in part, in codes adopted by enterprise.

34.  Most of the codes which have propelled the topic into the spotlight are operational codes in large retailing and manufacturing operations engaged in international trade. As well as large retailers and manufacturers, a number of industry associations of export suppliers or import retailers have issued codes of conduct. These codes are frequently for use by SMEs whose resources may be insufficient to develop their own codes of conduct. A small but increasing number of known codes are the result of collaboration of enterprise with workers' organizations, NGOs, governments, or a combination of these partners.

35.  In increasingly globalized relationships, the structure and operation of commodity chains play a defining role in the development and implementation of codes.(31)  Some codes develop by virtue of common ownership ("equity") between enterprises in the chain (e.g., tea plantations and packaging). Other codes are applied as conditions to contractual arrangements, which may represent long-term relationships, or fragmented, highly opportunistic outsourcing arrangements. The closer and more long-lasting the relationship between retailer and supplier, the easier the application of the code appears to be. The longer the chain of production and more complex the levels of contractors, subcontractors and buying agents, the more challenging the situation is. Manufacturers with wholly owned facilities abroad or ongoing relationships with contractors have built-in arrangements for code implementation, while retailers removed from suppliers must use bargaining power to require compliance with code standards. Retailers who directly contract out manufacture of private-label merchandise can directly influence their contractors. Workers' organizations have developed strategies for negotiations on codes that apply across commodity chains, based on solidarity campaigns among trade unions in production, packaging, distribution and/or retail.(32)  The chain-wide model depends for effectiveness in part on the structure of the commodity chain -- the more cohesive, the more likely an initiative can be adopted and implemented. A significant problem, for which insufficient evidence to date has been collected, concerns the costs associated with the monitoring of codes, particularly for producers subject to varying requirements of different buyers and retailers. Although, in many instances, the added costs of code systems are reflected in a retail mark-up, nonetheless, if the cost is significant, the point at which it is met in the first instance could distort the competitive position of enterprises.

3. Operational codes

36.  Operational codes are developed by enterprises, enterprise associations, workers' organizations, NGOs or, in some cases, a combination of one or more of those entities. Available information suggests that the world's largest MNEs, and in particular US-based MNEs in the TCF and related commerce sectors (e.g., manufacturers, retailers including department stores, mass merchandisers, specialty stores and mail order clothing companies), have led the trend toward usage of codes as a means of responsible sourcing. Japanese and Korean MNEs of comparable size, for which little information is available relating to single-enterprise conduct, appear to operate with creeds or philosophies of a very general nature directing, for example, enterprise respect for employees.

37.  Enterprise-association codes may be developed by business associations, industry groups or employers' organizations. In contrast to single enterprise codes, these codes reflect a negotiated consensus among association members. Available information suggests that, with a few exceptions, business associations, while becoming more aware and active on issues of social responsibility, in general do not adopt codes as tools to address labour concerns. This trend may be the result of lack of consensus among widely differing business sectors represented in the membership of such associations. In contrast, industry associations, both sectoral and multi-sectoral, have developed a notable number of initiatives relating to codes but few contain references to international labour standards. At the global level, associations in the toy, tea, sporting goods and chemical industries, among others, have developed codes with labour provisions. Examples include Athletic Footwear Association, Tea Sourcing Partnership, Responsible Care (chemical industry), International Council of Toy Industries and World Federation of the Sporting Goods Industry. Some of these have been criticized for inadequate provisions for implementation, and insufficient incentives to encourage retailers to invest in, and work with, suppliers. European regional initiatives in industry associations appear distinctively characterized by collaboration with workers' organizations. This feature may be attributed in part to the effect of pre-existing regulations which nurtured a culture of consultations among labour and management, thus assisting a predisposition to extend discussions and ultimately arrive at code agreements to new issues.(33)  Industry associations at the national level, particularly those comprised of SME members, appear to adopt codes in part to facilitate a common public image, avoid unnecessary costs, and prevent adverse competitive effects. This trend can be noted at both ends of the commodity chains and in developing as well as developed economies.

38.  Employers' organizations appear to be most active on codes at national and local levels in Latin America, where some have sponsored code-like structures in which individual companies may voluntarily participate. A Brazilian employers' organization helped to sponsor the Abrinq Foundation, which has had considerable effect in using a certification system in Brazil's fight against child labour (see section IV infra). Similarly, employers' organizations in Colombia and Guatemala have instituted policies on child labour, each with a follow-up system for signing up company members or recognizing individual adoption of codes thereafter.

39.  In a trend that appears to be increasing, workers' organizations are joining with enterprises and enterprise associations, to negotiate and implement joint codes of conduct with labour provisions. In general, codes that have developed with worker participation incorporate international labour standards in a more consistent pattern than do other types of codes reviewed. Most such codes also appear to utilize some type of external monitoring and certification system and many use the subscription method. As discussed above, worker/enterprise codes appear to have advanced the farthest, in terms of implementation, within the European Union. A notable example of collaboration between a workers' organization and an MNE, the Danone group, is discussed infra (footnote 71). Non-European agreements between workers' organizations and enterprise associations present widely disparate results, including codes focused on a single issue (usually child labour) and codes containing a full set of fundamental labour standards.

40.  Available information suggests that joint codes have been negotiated by workers' organizations in internationalized consultations between international trade secretariats (presumably in consultation with local trade unions) and MNE headquarters management. Evidence on what effect, if any, these "code agreements" have on local collective bargaining agreements is as yet unavailable. In any event, it would appear that coordination between ITSs and local workers could serve to provide greater participation by those directly employed in the situations to be covered by a code of conduct than is the case in codes developed solely by enterprise and/or NGOs without worker involvement.(34)  The international code agreements are presumably subject to general contract law obligations, but it is not certain that, in contrast to collective bargaining agreements, those obligations would be made effective by national legal systems in each country of the MNE's operations. One innovative approach has been to commit the enterprise, within the "code agreement" to incorporate that agreement into its existing local level collective bargaining agreements. This type of approach has developed in Europe in the context of the EURATEX/ETUF-TCL agreement and in part reflects the particular national law and practice concerned relating to collective bargaining relationships. The strategy of transposition to the national level of collective bargaining agreements brings with it, not only legal benefits, but also existing implementation systems and relationships, and a method for achieving uniformity of standards among enterprise operations in various countries and circumstances.

41.  With increasing frequency, hybrid code systems involving enterprise, workers' organizations, and NGOs are developing with focuses on labour practices. These hybrid codes, initiated by NGOs in many cases, appear to be growing, while codes proposed exclusively by NGOs appear to be decreasing. Most hybrid codes operate through the subscription method, sometimes prompted to do so through NGO dialogue or campaigns. Experience to date suggests that the participation of NGOs in codes of conduct involving labour practices has shifted the traditional balance between worker and employer at the negotiating table. Negotiations tend to be more piecemeal, with public demands contributing to determining which sectors, which issues, and which safeguards will be the principal subjects. The new demands of civic groups are often negotiated in processes that are marked by few rules of procedure, little transparency and are devoid of legitimacy. In recent times, workers' organizations have increasingly been accepted by enterprise management and NGOs as important contributors to the initiatives. In several high profile efforts, workers' organizations are noticeably outnumbered in comparison with their enterprise and NGO counterparts(35)  but recent trends show otherwise. For example, the recently born UK-based Ethical Trading Initiative has equal representation among enterprise, workers' organizations and NGOs. Within the enterprise sector, however, SME representation in the formulation and oversight of hybrid codes appears quite limited.(36) 

42.  The content and implementation of hybrid codes largely depend upon the identity of the various code partners and the leadership they provide. The partners in the hybrid system significantly affect its start-up influence and range: where large enterprises dominate, the potential for broad influence exists but may be diluted in part by the divergent interests among the members.(37)  Several efforts with more balanced roles among the partners report difficulty in achieving consensus on difficult issues. The Ethical Trading Initiative in the United Kingdom has met with problems over such issues as freedom of association and child labour but recently adopted a Base Code for its member companies to adopt or incorporate into their own codes, referencing ILO standards on those and other issues. The United States Apparel Industry Partnership encountered public tensions over the living wage issue and is now facing obstacles on agreeing to a system for independent, external monitoring and verification.

43.  Sensitive to public and media attention, hybrid code partners tend to concentrate on labour practices in particular sectors and countries. Areas of particular interest include TCF (Clean Clothes Campaign, Apparel Industry Code), forestry (Forest Stewardship Council, discussed further in section IV infra), and food and agriculture (Fair Trade Organizations, Draft Banana Charter). Several hybrid codes apply to labour practices of MNEs in specific countries (MacBride Principles for Ireland; Sullivan Principles for South Africa; Maquiladora Principles for Mexico). A few recently developed systems are multi-sectoral in nature. The SA8000 system aims at a generic certification of any enterprise, discussed supra section II.3. The Ethical Trading Initiative focuses on implementation of code requirements.

44.  Governments and IGOs tend to play supportive roles in selected hybrid code systems, serving as catalysts, facilitators, endorsers, and even financial supporters. The Ethical Trading Initiative receives its core funding from the United Kingdom Government, which also sends a representative to observe the Board meetings. The Apparel Industry Partnership was developed with the encouragement of the United States Government. The code prepared by the Federation of International Football Associations (FIFA), with ITS assistance, led to an agreement to run cooperative projects to eliminate child labour from particular industries in South Asian countries, through the collaboration of ILO/IPEC and UNICEF with industry groups, workers, governments and NGOs (see section VI infra (TRAVAIL)). In Brazil, UNICEF has cooperated to develop the Abrinq Foundation with its child labour social labelling programme (see section IV infra), and a state government in Brazil works with enterprise and UNICEF to collaborate in the provision of schooling and the elimination of child labour from rural activities.

4. Model codes

45.  Model codes, which continue to be issued by business and industry associations, trade unions, NGOs and hybrid coalitions, serve as a conceptual basis for the development of codes within enterprises or simply as a general influence on enterprise. Behind the emergence of some of these codes is a strategy to promote use of consistent language in codes of conduct and benchmarks for evaluation and comparison of unilaterally adopted codes, but available evidence is inconclusive as to their various effects. Model codes issued by business associations tend to be ethically oriented, some with general references to labour practices.(38)  Model codes developed by workers' organizations appear to aim both at guiding trade unions in the negotiation of codes in particular sectors,(39)  and at serving as general guidance for all industries.(40)  With a few known exceptions, NGO model codes tend to take on a "benchmark" quality, providing a set of principles, minimum standards, or guidelines to be followed in the process of adoption, and the actual content of codes of conduct by enterprise.(41)  In the public sector, the United States Government has proposed its own Model Business Principles(42)  for United States companies doing business abroad.(43)

5. Content

46.  This section addresses general trends in the labour-related content of code provisions, with special attention paid to the choice of labour practices in codes, sources of reference for defining the level of labour practices to achieve, and actual content of the practices aspired to in various codes. As demonstrated below, a comparison of codes reflects discrepancies in the various labour issues selected for attention, the sources of reference used to define proper practices, and in the actual level of practices targeted through codes of conduct. The issues covered in this section comprise fundamental labour rights (freedom of association/collective bargaining, freedom of discrimination, forced labour and child labour), wage levels and occupational health and safety.

47.  A preliminary analysis of approximately 215 available codes(44)  demonstrated that significant hurdles in evaluating the labour content of codes of conduct exist at the present stage of research, making it extremely difficult to give reliable indications of trends. Survey methods of data collection have generally failed to obtain high rates of responses from companies, leading to uncertainty as to the degree to which codes addressing labour practices are used by enterprise. The limited information available suggests that, of companies responding to surveys, a majority that have international sourcing guidelines or codes of conduct also address labour practices.(45)  Where such codes are found, the reasons behind the selection of issues, and reference sources used to determine the level of practices to achieve, must be uncovered from the language of the code and the enterprise's internal operations and external environment.

48.  Choice of issues. The labour issues included in codes of conduct often, but not necessarily, appeared to reflect the nature of publicized labour problems in the various industry and service sectors. As discussed below, codes reviewed in the TCF sector, for example, tended to concentrate on child and forced labour. While three-fourths of codes reviewed mentioned health and safety, special attention to that issue was apparent in the efforts of enterprises in the chemical, transport, TCF, mining, commerce, postal and toy manufacturing sectors. Approximately two-thirds of the codes reviewed, representative of most sectors of code activity, addressed the issue of employment discrimination. In contrast, relatively few codes addressed freedom of association and collective bargaining. A review of references to international labour standards yields similar conclusions.(46) 

49.  At present there is insufficient data to determine the actual impact of codes on labour practices. Nonetheless, the selective approach by which most codes address labour problems appears to promote uneven implementation of fundamental labour rights, both within a single enterprise and across the world's labour force. Given the interdependence of fundamental labour standards, a piecemeal treatment could make it difficult to sustain progress in one selected area while leaving others behind. For example, without freedom of association and collective bargaining to promote a balance in employer-employee relations, it is difficult to achieve the effective abolition of child labour. Likewise, respect for hours of work and social benefits contribute to health and safety systems in the workplace.

50.  Sources of reference for defining the level of labour practices to achieve. A wide diversity, and at times divergence, exists among the methods used to determine the level of labour practices to achieve. Among the codes reviewed, code drafters tended either to create their own definitions of labour practice targets ("self-definition") and/or refer to one or more of the following sources: national law, international labour standards and industry practice. Self-definition appeared to be the leading method of establishing labour practice goals, particularly among enterprise-drafted codes. As exemplified below, the definitions varied widely in similarity or divergence from international labour standards on the given issues. Self-defined standards appeared most frequently, as discussed below, to set goals implicating level of wages, health and safety, and certain fundamental labour rights. Code provisions which only used portions of ILO instruments in many cases changed the meaning or intended protection of the instrument and qualified as self-definitions.(47) 

51.  National law was second in frequency of use as a reference source for code "standards". General provisions guaranteeing respect for national law appeared frequently in the commerce sector, and industry associations regularly appeared to refer to national law. A few codes mentioned national law as a minimum standard only. Of all issues addressed, the area where proportionately most frequent reference was made to national law was level of wages -- an area lacking an international standard. In contrast, references to industry standards, by matching or exceeding them, appeared to play a lesser role than national law or international standards in code provisions. The nature of code initiatives as tools for enterprise leadership may help explain this phenomenon.

52.  No more than one-third of the codes reviewed bore any reference to international labour standards, either generally or specifically mentioning the principles of an ILO Convention or Recommendation.(48)  References to ILO standards occurred proportionately more often in joint (enterprise/worker) and hybrid codes than in codes developed by industry associations, employers' organizations or enterprises alone. References to multiple fundamental international labour standards were most frequent among worker/enterprise codes and model codes, and appeared rarely in codes generated by enterprise alone.(49)  Many of the codes of enterprises considered to be pioneers in social responsibility had no references to international labour standards or ILO instruments.(50)  The Tripartite Declaration on Multinational Enterprises and Social Policy was referred to in only one code, a hybrid code in the postal and telecommunications sector. The 1998 ILO Declaration on Fundamental Principles and Rights at Work has already been incorporated as a reference point in one code, the Ethical Trading Initiative's Basic Code.

53.  Freedom from discrimination. Among the labour practices in codes of conduct reviewed, the issue of freedom from discrimination in employment appeared in approximately two-thirds of all codes, representing the second most frequently raised labour issue across all sectors of codes activity (after health and safety infra). A majority of these references were self-definitions. Very few codes actually referred to the ILO's Discrimination (Employment and Occupation) Convention, 1958 (No. 111), while six codes mentioned a particular national law on discrimination.(51)  Many of the codes treated freedom from discrimination in general allusions to respect and dignity of workers,(52)  or direct references to eliminating discrimination.(53)  Others enumerated grounds of discrimination,(54)  but few reiterated all the grounds of discrimination found in Convention No. 111.(55)  Some of the codes which mentioned freedom from discrimination went further to promote equal opportunity for advancement.(56)  Mention of equal pay for work of equal value appeared in only a handful of codes, all of which directly referred to Convention No. 100 concerning Equal Remuneration for Men and Women Workers for Work of Equal Value.(57) 

54.  Child labour. Slightly less than half the codes (about 45 per cent) highlighted the issue of child labour. Particularly active sectors which concentrated on child labour included food and drink, TCF, and toy and commerce, although several child labour prohibitions existed in codes in the forestry, basic metal production and construction sectors. The majority of these references were formulated as self-definitions. Two characteristics marked self-defined prohibitions: a lack of definition,(58)  or a self-determined age, usually 14 or 15 years.(59)  A proportionately higher number of the code references on child labour (nearly one-third) referred to national laws than in treatment of other labour issues. Some enterprises imposed their own standard or national law, whichever was higher.(60)  Others deferred to national law when a conflict arose between their self-defined age or an age set by international standard.(61)  Still others set their own age standard if no national law addressed the question of child labour. Less than 10 per cent referred to Convention No. 138 in defining a minimum age standard; these were roughly split between enterprise-generated codes, enterprise/worker codes and hybrid codes.

55.  Forced labour. Roughly one-quarter of the codes prohibited forced labour, with the overwhelming focus of those references in the TCF, commerce, and toy sectors. Four out of five of these codes incorporated a self-definition, primarily by prohibiting "forced labour" without defining it further.(62)  Most were limited to prohibiting forced labour in goods; services were added exceptionally.(63)  Many added prison labour, either as a qualifying or additional component,(64)  and some focused on corporal or mental abuse as well.(65)  None appeared to refer to national laws on forced labour. Most of those which referred to ILO standards cited either Convention No. 29 or Convention No. 105; one code alluded generally to the rules and guidelines of the ILO on the subject.(66) 

56.  Freedom of association and collective bargaining. Approximately 15 per cent of the codes contained references to freedom of association and/or collective bargaining; the paucity of treatment of this issue is remarkable, given the degree of unanimity on its essential nature between employers and workers at the ILO level. The references occurred most frequently in the TCF and commerce sectors, and to a lesser extent in the food and drink, forestry, utilities, chemicals, basic metal production, agriculture, mining and hotel sectors. A little less than half the code references, primarily those in single-enterprise codes, created self-defined targets. The self-definitions, which varied widely, included statements affirming freedom of association and collective bargaining,(67)  others alluded generally to respect between labour and management,(68)  still others supported the elimination of trade union activities,(69)  and some proposed a combination of approaches.(70)  About half of those which mentioned the issue referred to relevant ILO Conventions, primarily the Freedom of Association and Protection of the Right to Organise Convention, 1948 (No. 87) and the Right to Organise and Collective Bargaining Convention, 1949 (No. 98) and, in a few cases, the Workers' Representatives Convention, 1971 (No. 135). With the exception of the enterprise noted supra (KappAhl), these references appeared only in codes developed with the participation of workers organizations and/or NGOs.(71)  Less than half of the code provisions referred to national law, sometimes using it to qualify, or limit, respect for freedom of association and collective bargaining.(72)  One hybrid code, SA8000, opted for an internationalized approach except in situations where national law restricted freedom of association and collective bargaining.(73) 

57.  Wage levels. Somewhat less than half (40 per cent) of the codes formulated commitments relating to wage levels. The references mentioned national laws, industry standards, or chose to self-define an appropriate formula. Among all three types, the TCF and commerce sectors led in frequency of reference to wages, although other sectors demonstrated notable activity, including chemical, financial services, and food and drink sectors. References to national law occurred at a slightly higher rate than self-defined references. Those which referred to national law in many cases also referred to matching or, less often, to exceeding an industry standard.(74)  Self-defined standards for wage levels often invoked the principle of "fairness" in setting a general standard.(75)  A theme which emerged from some references required a wage sufficient to meet "basic needs" of life;(76)  some qualified the formula with local practices and conditions while others maintained an absolute "living wage".(77)  Some codes opted to default from a self-definition to national law, or vice versa. Although no explicit reference to an international labour standard on wage levels appeared, references to wages addressing basic needs would be consistent with the Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy, which advised multinational enterprises operating in developing countries to provide wages, benefits and conditions of work "at least adequate to satisfy basic needs of workers and their families".(78)  Some code references sought to ensure full and immediate payment of wages which, under ILO jurisprudence, could be relevant to preventing conditions contributing to forced labour.(79) 

58.  Health and safety. Roughly three-quarters of all codes contained provisions on occupational safety and health, the labour issue to which reference was most frequently made. Among the most active sectors were the chemical industry and the TCF, transport, mining, commerce, postal and toy manufacturing sectors. Nearly 70 per cent of these contained some sort of self-defining standard, alone or in coordination with national law. Self-definitions included general descriptions of the goal,(80)  specific performance requirements for achieving the goal,(81)  and/or management systems approaches.(82)  Less than one-third of the references mentioned national law; some sought to maintain both national law and self-defined standards;(83)  others looked to self-defined standards where no adequate national health and safety law existed.(84)  Very few references to the relevant international labour standards appeared in codes of conduct; these were all found in hybrid codes, such as SA8000.(85) 

59.  Overall considerations. The above review of the selection of labour issues and sources of reference suggests that labour practices are largely covered by ad hoc negotiations between the various parties interested in drawing up codes. The content of hybrid codes, and differences in emphasis among sectors, reflect the unequal access to information among interested parties, and the compromises arrived at in negotiating such codes. In one example, the Clean Clothes Campaign found itself committed to the narrower Fair Trade Charter for Garments, agreed to among social partners in the Netherlands, when it attempted to promote the broader, more recent Code of Labour Practices for the Apparel Industry. Ultimately, the lack of a firm basis can lead to the selection of only certain labour standards, or even only portions of a particular international labour standard, and thus jeopardize the traditional collective bargaining processes. They could also make it difficult, if not impossible, for smaller producers and other enterprise partners to satisfy the demands of code proponents, often key business partners. The cost in attempting to comply, and the damage to credibility if the attempt is unsuccessful could be detrimental to fair competition and productivity. Consumers and investors, faced with divergent claims and no way to assess or compare the commitments of enterprise, may in turn withdraw support, affecting economies of scale and enterprise development.

6. Implementation of codes

60.  Not infrequently, codes launched with much publicity in an import country are unknown, unavailable or untranslated at producing facilities; even where available, workers may have no way of reading the code or reporting non-compliance without disciplinary treatment or dismissal.(86)  A United States study found that few workers were aware of existence of codes and that formal training about codes was not common among suppliers overseas, but was more likely to occur where a supplier produced largely for the company with the code.(87)  With such discoveries and the advent of environmental accounting, among other developments, the area of implementation of codes of conduct has become the subject of intense experimentation in the 1990s.(88)  Approaches as divergent as the type of codes, identity of actors, and nature of industries mark the terminology and procedures currently in use.(89)  Experience suggests that the current lack of standardized principles and procedures hinders good quality in implementation of codes, and prevents the reporting of comparable data to help measure progress in the enterprise, and within or across industry sectors. The lack of standardization also contributes to suspicions of third parties about internal monitoring processes and exploitation of labour conditions due to cultural relativity in MNE practices.

61.  Attempts to set principles and procedures for the implementation of labour-related codes have multiplied as rapidly as adoption of codes themselves. Principles for code implementation are found within codes, in appendices and guidance documents published for use with specific codes(90)  and in general principles proposed by coalitions which have adopted codes or offered for code users in general.(91)  One attempt specifically aimed at the perceived needs of enterprises in developing countries has recently been piloted.(92)  Common to most of these efforts are emphases on the need for participation of all interested parties, comparability of reporting data, periodic evaluation and revision of methods, transparency -- including disclosure of reports, and the need for external auditing to verify and validate processes used.

62.  Guidance for procedures to implement codes appears to be set generally by reference to unspecified "best practices" in the industry, or by definitions created or negotiated by the code sponsors. Exceptionally, direct references to the ILO appear which, when used, tend to hold considerable weight in the implementation of a code. The most recent code of the Clean Clothes Campaign, for example, refers to international labour standards and ILO jurisprudence in interpretation of code provisions or disputes as to whether a practice actually violates the code. Other provisions refer generally to ILO expertise in labour standards and labour inspection methods and, in the case of SA8000, recommend code users to specific ILO field offices for information and assistance. References to national law in implementation of codes are rare.(93) 

63.  In practice, implementation of codes involves, in the first instance, internal management systems. Although mention of compliance and enforcement procedures is common, "[v]ery few ... codes emphasize or discuss in detail internal oversight and personal integrity."(94)  Evidence suggests that companies with well-functioning quality assurance management systems (including compliance with the ISO 9000 series) also tend to have effective systems of implementation for labour-related codes provisions. Other standard systems of relevance may include financial accounting, environmental accounting, and HSE management;(95)  ILO-recommended occupational health and safety management systems;(96)  and environmental management standards of the ISO 14000 series. These various management systems share basic steps in common, including setting a clear and detailed enterprise policy, allocating managerial resources for effective dissemination of the policy (including translation), developing and implementing tools for monitoring, reporting, and taking corrective action, including training programmes. Reports from practice suggest that code implementation systems often lack sufficient human resources, adequate participation by workers, and transparency in application. An emerging trend among MNEs seeks to allocate responsibility within enterprise for social management through the establishment of a focal point as both a public signal and management mechanism.

64.  Internal systems for social accounting range from self-declaration of code compliance by managers, suppliers, or affiliates, to active monitoring, evaluating and reporting processes. Self-certification can be done by written acknowledgements of contractors, suppliers or buying agents, certification on shipping documents, verification that certain actions have been taken, requirements to maintain documentary proof of compliance (with a right reserved to carry out on-site inspections), or simply a contractual provision. In contractual relationships, some enterprises engage buyers, or other brokers in a commodity chain, to certify compliance of suppliers; this mechanism places the enterprise with the code at a greater distance from the actual supplier and evidence suggests it is not as effective. Active monitoring may be conducted by enterprise personnel or consultants, or externally by third parties or professional inspectors. Enterprises generally favour internal processes because of concern about leaks of confidential information. However, according to one study, the more vertically integrated the enterprise is (i.e. owning or controlling all steps of the production process), the more likely monitoring will be facilitated by an in-country or regional presence facilitating frequent inspections.(97)  Less monitoring occurs with less integrated relationships in the chain or in the company's operations. Public reporting of results may serve to allay suspicions about internal monitoring to some degree.

65.  External monitoring or inspection may refer to various types of third-party assessment. Some such monitoring is conducted with information and methods controlled by the enterprise under study; other external monitoring operating outside such constraints is considered independent of enterprise control. The assessment of independence depends on the circumstances as well as the actors involved. One approach involves the use of industry association and even employers' organizations as third-party monitors. Another approach, that of retaining a professional inspection or auditing firm, is becoming more common as such firms adopt social portfolios to enhance other services offered. The firm retained commonly provides the company with other inspection or auditing services already. Some evidence indicates that traditional financial accounting firms may be less independent due to inexperience in the detection of workplace violations and pre-existing contractual relationships with enterprise management. In another approach to external monitoring, MNEs and others are entering into agreements with NGO-led coalitions to monitor and produce reports; the financing of such initiatives is unclear, and trade unions are noticeably absent from many of these arrangements. Another development, too recent to judge, has trade unions and enterprises setting up joint structures for monitoring and even certification.(98)  In a further approach to external monitoring, industry or employers' organizations sponsor an accreditation structure to be used by member enterprises at lower cost and potentially greater visibility due to the common membership front. Third-party models with NGOs and workers' organizations pursue accreditation systems with certification fees; in this structure, an enterprise faces a less costly system but potentially has less input into the process of certification than in any of the other models.

66.  The two processes -- internal and external -- can be considered mutually complementary. Internal monitoring appears to be a necessary aspect of rooting a code in enterprise culture; external monitoring and inspection provides a useful litmus test of that culture. For both, a key question involves standardization of criteria and procedures to be used to assess the situation in the workplace under the code standards. A variety of audit checklists, vendor questionnaires and worksheets have been published by code sponsors to suggest quantitative and qualitative means of measuring a diverse range of issues -- from adequate ventilation and sexual harassment to determination of a worker's age, or freedom to assemble and associate with other workers.(99)  In addition, a lack of sufficient safeguards governing procedures for collection of information marks a number of initiatives in both internal and external monitoring; this insufficiency merely aggravates what are in many cases the already questionable indicators of measurement.

67.  The content of labour provisions in codes naturally affects the type of criteria, and even procedures, used for internal monitoring or external inspection. Recognizing this, some independent inspection companies have approached the ILO, requesting its assistance in using international labour standards to establish basic criteria for inspection, and seeking its views on the codes of conduct regarding inspection. More such requests are probable in the short term, as experimentation in social accounting and inspection pushes forward to address "the fragmentation of standards and metrics in this field [in order] to obtain company capacity to manage and to report to stakeholders".(100)  A number of observers predict the trend will become established practice in the future.

IV. Social labelling

68.  The term "social labelling" has come to connote a means of communicating information through a physical label(101)  about the social conditions surrounding the production of a product or rendering of a service.(102)  Social labels may be affixed to products or their packaging or displayed on shelving or shop windows at the retail site. Some labels are assigned to enterprises, usually producers or manufacturers. They are aimed at consumers and/or potential business partners. Social labelling programmes are considered to be voluntary responses to market incentives (including the demands of business partners) rather than to public law or regulation.

69.  This section reviews recent forms of social labelling which address specific labour practices, including their structure and operation, potential impact, and relationship to issues within the scope of the ILO mandate.(103)  General trends are described below. (104)  A company's or programme's inclusion in this section does not constitute endorsement by the ILO of any product, service, company or programme.

1. Structure and operation

70. Civil society processes, including campaigns and public demands, create the perceived market for social labels and represent a key influence in the development of social labelling programmes. Labels which operate independently of any one enterprise (independent labels) have been developed and administered by NGOs, workers' organizations (union labels), industry and trade unions or other enterprise associations, or hybrid partnerships of one or more of those sectors. Single enterprises engaged in production, export or retail sales,(105)  or enterprise/government partnerships, have also developed independent social labels. Social labelling programmes run by NGOs or hybrid partnerships tend to be dominated by organizations in developed countries, particularly in the initial phases; those led by enterprise associations or public/private partnerships have included coalitions in developing countries, such as Abrinq in Brazil and Kaleen in India.

71.  Complex market relationships exist among social labels, codes of conduct, company names, licensing agreements, and other private voluntary initiatives. In some cases, small and medium enterprises, or enterprises in industries with little brand loyalty, have shared the costs and higher visibility of a common independent social label, usually administered by an enterprise association or hybrid partnership on the basis of a commonly accepted code of conduct. In such situations, the label may be affixed to the product as, for example, with RUGMARK, or may consist of a trade name used by certified companies, as for example, Responsible Care (see supra section III). On the other hand, an enterprise which has a good reputation established on the basis of a well-known code of conduct may find that its company name acquires, over time, the status of a brand name label (e.g. Gap).(106)  In some cases, retailers or private label manufacturers license the use of their logos or trade names to contractors which meet preset standards, often found in the code of conduct developed by the retailer or manufacturer. In such cases, the logo or trade name operates as a means of reporting code compliance to buyers and vendors along the chain of production as well as to the general public and consumers. Recent licensing examples include procurement policies of US-based Duke University and Notre Dame University,(107)  FIFA soccer balls used in the World Cup and athletic equipment procured for the Sydney Olympics. Some programmes which start as multi-stakeholder subscription codes (see supra section III) may eventually adopt a certification label as, for example, the Clean Clothes Campaign.

72.  Social labels influence labour practices in a selective manner. Rooted in concerns of consumers, media and civil society campaigns, many social labelling programmes target consumers in developed countries and producers in developing countries. The programmes appear primarily in export markets involving retail trade,(108)  and with market "niche" products, affluent consumers and eye-catching circumstances. Products vary in adaptability to social labelling depending on the price sensitivity of the sector, the role of brand recognition, and concern of consumers respecting social issues in the supply chain involved. Some labels apply only to highly specific sectors such as hand-knotted rugs, soccer balls or cut flowers; more general labels cover various products in the clothing industry or selected agricultural products. Labels may be more likely to develop with products bought and consumed publicly or associated with social identity (such as clothing, footwear, food and luxury goods) or discrete production processes (such as tea).(109) 

73.  The costs of running a social labelling programme have not yet been proven to operate on a self-sustaining basis. Costs normally include start-up and administration expenses, such as physical production of the label, establishment of the verification structure, and consumer outreach and promotional publicity. Once in operation, a small percentage of profits may be channelled back into local improvements in the workplace, the region of production or, in some child labour oriented programmes, educational and rehabilitation programmes for former child workers. In many cases, programmes are subsidized by licensing fees imposed on users of the label (producers or distributors). In addition, programme costs may be borne by fees assessed proportionately based on such factors as value of labelled products exported or earnings gained from their exportation. Importers may also pay levies on labelled products; in such cases, the cost may be passed on to the consumer or, particularly with programmes run by individual enterprises such as Reebok, absorbed in some other internal way. Some programmes, such as the fair trade labelling programmes, are specifically designed for consumers to pay a premium which operates as a mark-up that ultimately shifts to the consumer some of the extra costs of labelling. Grants and in-kind contributions by way of monitoring or other resources often help to finance new programmes. These contributions come from governments of both exporting and importing countries, and from NGOs, IGOs, and in a possible new trend, workers and employers themselves. For more information, see "Sources of financing" columns in the Appendix.

2. Content

74.  As with codes of conduct, social labels may be single issue (child labour, freedom of association) or multiple issue, containing a set of labour standards or a set of standards including labour(110)  as well as other issues like fair trade or forest conservation. The issue most frequently targeted among programmes reviewed was child labour (see Appendix), and second most frequent was level of wages. Some of the issues most frequently covered in the general review of codes of conduct discussed in section III, supra, such as health and safety and freedom from discrimination, appeared less often in social labelling programmes reviewed (see Appendix). Like codes, roughly one-third of the labelling programmes included some reference to international labour standards.

75.  Some programmes originally aimed at non-labour issues have added labour issues to their agendas as workers' organizations become more involved in the programmes. This is the case with fair trade labelling programmes, which have begun to collaborate, to a limited extent, with workers' organizations as they expand, on the production side, from smallholders to plantations and, on the distribution side, from alternative trade shops to mainstream retailers.(111)  Similarly, the Forest Stewardship Council (FSC), which began as an effort to address technical aspects of forest management, has added principles on freedom of association and occupational health and safety; at the national level, its affiliates, with trade union participation, adopt more specific labour criteria to fit local issues. In contrast, the labour issues addressed by the Flower Label Programme, developed by the German BGI (Flower Wholesalers and Importers) to ensure environmentally and socially correct production of imported flowers, were influenced at the start by trade union involvement. The participation of workers' organizations in the implementation of labelling programmes may include an opportunity to provide information to monitors (Abrinq), comment on certifiers' reports (FSC), or representation as observers (FLO tea registry).

3. Impact

76.  Research and understanding of the effects of social labelling is visibly impeded by the lack of standardized principles and criteria from which to report data and compare programmes. General studies on social labelling are limited in number, scope and methodology, due in part to the limited years of experience with social labels.(112)  In contrast, research on the effects of ecolabelling, in operation for more than 20 years, spans private and public organizations.(113)  Despite the head start, environmental labelling research has been criticized as "anecdotal".(114)  Without assessing their exact relevance to social labelling, selected experiences of ecolabelling which may shed light on future analysis of social labelling programmes are included in this section.

77.  Experience to date suggests that social labelling programmes bring direct and indirect effects, both helpful and adverse.(115)  On the helpful side, labelling programmes may improve working conditions and raise funds for educational and rehabilitation programmes for former child workers. They may also build consensus among industry groups, NGOs, international organizations and governments about labour practices to be addressed. Responses may take the form of more conscientious compliance or enforcement of labour laws or enterprise adoption of codes of conduct to forestall use of the label. Social labelling programmes may also bring adverse effects, including financial difficulty among participating enterprises and consequent loss of jobs. Higher prices of labelled products may result in lower penetration of the market. Some child labour labelling programmes have been criticized for driving child workers into less formal sectors where the elimination of exploitation is more difficult. Legal risks also exist, including issues of inconsistency with national anti-competition or truth in advertising laws and international trade law.

78.  Debate on how to determine the effects of social labelling has just begun. Some elements for analysis include market share of the label, consumer recognition, percentage take-up of participating companies, number of beneficiaries involved and magnitude of change involved in beneficiaries' income and consumer spending.(116)  Others attempt to capture patterns relevant to actual shifts in the mainstream, non-participating market or in public policy in government(s) involved. The latter approach is reflected in the proposition that, once labelled products reach 10 per cent or more of the market share, the reflected attitude of consumer groups becomes important to major producers and retailers, which then react with alternative initiatives such as codes of conduct. Economic theory, unsupported by empirical data, has been used to propose, in the ecolabelling context, that negative effects could arise when successful labelling schemes drive down the price of unlabelled goods resulting in increased demand for those goods.(117)  Opponents argue that the increased sale of labelled goods would result in increased economies of scale in production and consequently lower prices, which could be expected to result in turn in diminished sales, reduced economies of scale and higher prices for unlabelled products,(118)  and that awareness raising may result itself in increased demand for labelled goods.

4. Final considerations

79.  What appears evident is that the market alone, without a coherent international framework, has been ineffective to date in developing uniform and generally accepted standards that could help promote benefits and prevent the risks of labelling efforts. In the context of ecolabelling, the International Organization for Standardization (ISO), discussed supra section II.2, is developing standards for ecolabelling criteria, symbols, and verification processes.(119)  This effort, promoted by UNCED's Agenda 21, seeks to arrive at guidelines for standardizing labour criteria, label development and verification systems. Proponents of such standards argue that they could help prevent confusion among consumers arising from the conflicting diversity of criteria used, lack of clarity of meaning among various labels, risk of lost credibility with unverifiable claims, and chance of illegality under national or international regulation.

80.  Social labelling programmes share some of the potential benefits and drawbacks applicable to codes of conduct generally. On the positive side, labelling programmes appear to stimulate social concern among enterprise and consumers and provide a market-based financial (rather than regulatory) incentive to improve labour conditions. However, this benefit applies selectively, both to the labour issues involved in the criteria and to specific sectors of enterprise concerned, as discussed supra. On the negative side, labelling programmes tend to lack transparency and methods for independent verification of the claims behind the labels,(120)  promote "outside" intervention in national standard setting, and discriminate against producers in developing countries who face undue costs or other constraints in the process of certification of conformity assessment.

81.  As with other private initiatives, social labelling programmes may prove more effective and more equitable when combined with other programmes to form a comprehensive, transparent and reliable social policy. In addition, international trade rules, and in particular the Code of Good Practice of the Technical Barriers to Trade (TBT) Agreement of the World Trade Organization,(121)  offer guidelines on how voluntary standards can be developed and implemented in ways that prevent unjustifiable non-tariff barriers. That Code, discussed supra section II.2, encourages transparency and cooperation in the development of standards at local, central and regional levels and, in the long run, equivalence, mutual recognition and harmonization of public or private standards on "as wide a basis as possible" by reference to international standards. ILO standards, such as those on working conditions and environment, occupational safety and health, equality of treatment between men and women, non-discrimination, rights of tribal and indigenous peoples and employment appear relevant to the interpretation of the TBT Agreement and its annexes.(122) 

V. Investor initiatives

82.  Investor initiatives concerning labour practices of enterprise are part of the "socially responsible investment (SRI)" movement that has grown recently in importance in certain developed countries. Although there is no single accepted definition of SRI, the term generally indicates investment-related decisions that seek social change while maintaining economic returns. The desired targets of social change, however, vary considerably and appear to be based on highly subjective judgements.

83.  This section reviews two SRI approaches: the screening of investment funds and shareholder initiatives. Investment fund screening is the inclusion (investment) or exclusion (divestment) of publicly traded corporate securities from investment portfolios based on the social performance of the company in areas including labour relations.(123)  Shareholder initiatives involve the exercise of rights based on an established share ownership as a means of influencing company behaviour. Formal shareholder action includes submitting and voting shareholder resolutions(124)  and asking questions at companies' annual meetings; informal action involves attempts to reach agreements with management through dialogue and negotiation. A growing "spirit of compromise" has been observed recently, together with a desire to avoid embarrassing publicity and the high costs of management time in dealing with the resolutions.(125)  The two approaches differ in timing and orientation. Fund screening occurs at the time of purchase or sale of company stocks and may involve withdrawal of money from a company deemed socially irresponsible. In contrast, shareholder initiatives occur after purchase and seek to influence corporate policy decisions by maintaining a relationship rather than pursuing a "boycott" approach.

1. Screening of investment funds

84.  Available evidence suggests that the phenomenon of SRI is growing but its impact on labour practices remains inconclusive. After a lead in the United States,(126)  SRI funds were set up in the United Kingdom and France in the early 1980s(127)  and spread in the late 1980s to Australia, Austria, Germany, Japan and Switzerland and other developed country markets.(128)  SRI is still growing in the United States, led by screened mutual funds,(129)  and much of the available research information derives from that country's experiences. Investors that use social screens tend to be public retirement and pension funds, colleges and universities, religious groups, social investment funds including mutual funds. In some countries, legal rules regarding the fiduciary duties of institutional investors, particularly pension funds, have restricted their consideration of any factors other than financial performance when considering investments.(130)  Although controversy persists over the financial gains of SRI portfolios, a growing number of authors tend to review them positively.(131) 

85.  Labour criteria used in screening appear to be unspecific and no standardization is apparent among screens. About one quarter of the American socially screened mutual funds use labour-related criteria such as no sweatshop labour, advancement of women and people of colour in the workplace and issues of concern to unions.(132)  The criteria may serve as bases of exclusion or inclusion. Investors often base their decisions on information gained from research institutions(133)  and that provided by the company itself, usually as a reply to "screening questionnaires". In deciding how stringently to apply the criteria, fund managers usually compromise between a strict application of the criteria that would exclude any enterprise with problems and traditional investment criteria.

86.  Insufficient evidence, coupled with the lack of standardization of criteria, make it impossible to determine a measurable impact of fund screening on labour practices. Nevertheless, it appears clear that the criteria used by various investment funds tend to differ, impeding the ability to send a clear message to enterprises; indeed, in some cases, the act of divestment could amount to nothing more than a silent withdrawal of money.(134)  In addition, as some advocates of these initiatives acknowledge, a combination of instruments may be necessary for screened funds to have an effect. In most cases, screened funds do not own enough shares in single companies to hit them hard by damaging their share prices but they may make a difference combined with campaigns and other social strategies. For example, the impact of divestment from apartheid South Africa in the 1980s was enhanced by the fact that the initiative used widely recognized labour principles (Sullivan Principles) and occurred in the context of a coherent, highly publicized campaign.

2. Shareholder initiatives

87.  Like fund screening, shareholder initiatives focusing on labour practices appear to be growing in number. Similar limitations in available data and lack of standardization or specification of criteria make it difficult to assess their exact impact beyond anecdotal evidence. Since the 1970s, shareholders, mainly in American companies, have regularly acted on labour-related concerns, through formal actions such as filing shareholder resolutions and submitting questions at annual shareholder meetings, and through informal communications with or to management, other shareholders and/or the press and the public, including "shadow annual reports" and disruptive action. Formal shareholder activism reached a record level in 1996 in the United States, but also appears in Japan -- with an overwhelming proportion of environmental and only few labour-related proposals -- and Germany, distantly followed by other European countries and Canada.(135) 

88.  The geographic trends reflect various cultural and economic factors, including civil society involvement in share ownership, level of development of publicly held corporations, and legal restrictions on grounds for shareholder activism. United States publicly held corporations are owned by a great diversity of shareholders, including church investors, while, in contrast, controlling shareholders in Canada(136)  and Europe, such as governments and banks, may have competing interests with their roles as lenders and pension funds.(137)  In Japan and Germany, the system of cross-holdings, in which big companies own shares in each other to protect managers from outside challenges, affects shareholder interests.(138)  As with investment screening, laws disallowing shareholder action on grounds considered not strictly financial present impediments to shareholder activism in Canada.

89.  Enterprises involved with labour-related shareholder proposals correspond in many respects with those active in private initiatives generally (see supra sections II and III). They are likely to be involved in outsourcing production internationally. In the United States, the two most active industry sectors appear to be oil and gas production and textiles, clothing, leather and footwear, including toy manufacturers;(139)  in the service category, commerce sector activity appears dominant. In Germany, activity in the chemical industry, particularly enterprise engaged in foreign direct investment, appears dominant.(140) 

90.  Coalitions among different types of institutional investors have emerged as the predominant partnership employed by many successful shareholder actions. The most important sponsors of labour-related resolutions are institutional investors, including insurance companies, pension funds, church funds, union funds, local authority funds and also fund managers who manage investments for other people. Public employees' pension funds and church investors are particularly active generally although evidence is inconclusive on specifically labour-related resolutions. Notably, labour shareholders as a group, such as labour unions and union pension funds, only occasionally sponsor labour-related shareholder proposals, although they do support such resolutions submitted by other sponsors.(141)  Shareholders appear influenced to take action by media coverage, boycotts against the shareholders' company, support by public officials for private initiatives, and reports of research organizations, universities, and NGOs. ILO findings of "violation[s] of internationally accepted labor standards" have been noted in proposed resolutions.(142) 

91.  Although shareholder resolutions address a wide range of issues, the number that directly address international labour issues is relatively small.(143)  The impact of those resolutions is most visible when the effort results in an agreed plan of action with management leading to the withdrawal of the resolution.(144)  Even a modest vote in favour of a resolution, however, can aid proponents in their discussions with management since it demonstrates that a significant number of shareholders support the spirit of the resolution. Measurable progress may take several years since proponents often put a resolution to vote several times if they are able to obtain a certain share of the vote legally required.

92.  In proposed resolutions, concerned shareholders usually request management to take some voluntary initiative involving labour practices within the enterprise. In some instances, shareholders have asked the enterprise to develop a code of conduct, adopt a third party code such as the Sullivan Principles, and even amend the articles of incorporation to include international labour standards as binding provisions. Shareholders have also asked for the enterprise to increase activity under a code of conduct, or to monitor and report on contractors' compliance with the enterprise's code of conduct and have even accepted an active role in the monitoring process. In other efforts, shareholders have requested enterprises to restate the formal company policy in response to various labour concerns. Companies that have a code of conduct have simply referred to the code and, in some cases, described their monitoring practice, when requested. Others have interpreted company policies to meet the concerns expressed. Frequently, shareholders ask enterprise to prepare a report on the enterprise's practice regarding its own labour practices abroad or its supplier standards. Recent resolutions requesting reports tend to give so many details and directives to management that it appears the end goal is not only to be informed on enterprise practice but also to influence the enterprise to adopt solutions to already identified labour problems. Notably, as a means of circumventing a legal rule excluding shareholder proposals concerning the "ordinary business" of the company, American shareholders have developed proposals to link executive compensation to social performance of the company.

93.  It is difficult to measure the impact of labour-related objectives in shareholder proposals due to the variability of external and internal environments of enterprises concerned, and the lack of standardization of goals and verification methods to measure progress toward those goals. Under such circumstances, enterprises face unpredictable parameters in designing and implementing private initiatives to address labour practices, and in reporting to, and negotiating with, concerned shareholders. In a similar fashion, shareholders have no accurate ways to assess enterprise improvements on issues of concern. Furthermore, most such initiatives do not involve the participation of the workers who could be affected by the initiative, except in unusual situations in which workers' organizations are acting as concerned shareholders for the very enterprise in which they are employed.

VI. Activities of the Office

94.  This section recapitulates, without elaboration, the activities of the Office which appear to be relevant to the private initiatives under review. The recapitulation is offered as an up-to-date view of what is being done in the Office, with no intention of making an assessment of those activities. Each department has reviewed the elements pertaining to its activities for accuracy and comprehensiveness, an exercise in which the Office has demonstrated a great deal of cooperation. This compilation may prove extremely helpful in that Office activities relating to codes of conduct, social labelling programmes, and other private initiatives are not consolidated in any one place or in any Office publication, whether in the programme and budget or on the Internet website.

95.  Only Office activities most directly relevant to the focus of this paper are included in the departmental lists below. The lists encompass activities directly focusing on codes of conduct and social labelling initiatives which address labour practices, or involving assistance to enterprises (in such forms as guidelines or training programmes) in taking voluntary initiatives relating to labour practices, monitoring and third-party (non-governmental) inspection systems. Notable activities beyond the immediate subject of this paper include programmes focused on such matters as social initiatives of enterprise outside the workplace,(145)  tripartite participation and social dialogue in the context of privatized enterprise-based training (EMPFORM), investment and improved employment relations in EPZs (RELPROF), workers' representation and information in MNEs (RELPROF), and the interaction of regional integration and labour relations (RELPROF). In addition, the lists below relate generally to private sector initiatives, and do not encompass Office activities relating to enterprise initiatives in which the governmental sector plays a leading role together with non-governmental partners. Such activities comprise, for example, Office coordination of the UN/private sector Joint Action Programme,(146)  and industry/government partnerships on codes of practice or training of government labour inspectors (SECTOR)(147)  and TRAVAIL (child labour).(148) 

1. Enterprise and Cooperative Development Department (ENTREPRISE)

Relevant activities

96.  The programmes of the Enterprise and Cooperative Development Department are guided by, among other matters, the need to promote good enterprise-level human resource management and industrial relations, based on tripartite consensus, and management approaches leading to corporate social and human standards which reflect ILO values. The department's activities in this regard include launching the ILO Enterprise Forum for business leaders, entrepreneurs, academics, employer and worker representatives, senior government officials, and others. The first Forum in November 1996, with 600 participants from 98 countries, considered "promoting social progress and enterprise competitiveness in a global economy". The second Forum, in November 1999, will focus on "enterprise competitiveness, corporate citizenship and employment challenges in the twenty-first century". Preparation for the second Enterprise Forum includes plans for exchanging views with enterprise actors on corporate citizenship and social initiatives, involving discussions with representatives of employers and workers. Workshops are also planned with business leaders and academics in cooperation with the International Institute for Labour Studies (see infra INSTITUTE). Other activities illustrating the department's approach include the International Small Enterprise Programme (ISEP), which seeks to generate high-quality jobs in small enterprises by focusing on, for example, voluntary initiatives on the part of SMEs, as referenced in the 1998 Recommendation concerning general conditions to stimulate job creation in small and medium-sized enterprises.(149) 

Publications and past activities

2. Bureau for Employers' Activities (ACT/EMP)

Relevant activities

97.  The Employer Meeting on Corporate Citizenship and Social Initiatives (New York, 1-2 October 1998), is debating the issue of corporate citizenship and social initiatives, in particular codes of conduct and social labels, to assess the implications for employers' organizations. The Employers' Activities Bureau is one of the organizers of the meeting jointly with the United States Council for International Business and the International Organization of Employers. ACT/EMP's technical and advisory programmes for employers and their organizations focus on, among other matters, integrating action on child labour into their organizations' activities through the Integrated Programme for Building Partnerships and Capacity Against Child Labour. ACT/EMP's work with employers' organizations include programmes relating to improving labour-management relations, the management of human resources, and occupational safety and health policies and practices.


3. Bureau for Workers' Activities (ACTRAV)

Relevant activities

98.  A general programme of the Workers' Activities Bureau seeks to promote more effective responses by workers' organizations at local, national and international levels to, among other matters, the implementation of codes of conduct for major enterprises and their suppliers. The programme includes advisory services, research, and training activities. One project within the programme, entitled Developing national and international trade union strategies to combat child labour provides support to international trade secretariats in seeking to establish codes of conduct with major companies through collaboration among unions along the supply or "value" chain.


4. Multinational Enterprises Branch (MULTI)

99.  Relevant work of the Multinational Enterprises Branch focuses on, among other matters, the Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy (Tripartite Declaration). Full-scale quadrennial surveys are conducted to monitor the extent of application of the Tripartite Declaration by governments, employers, workers and MNEs. A summary of the replies received, with an analysis of them, are submitted to the Governing Body for discussion. The Branch also oversees requests for interpretation of the Tripartite Declaration, which may be submitted in the event of disagreement over the application of the Tripartite Declaration, and are governed by the Procedure for the examination of disputes concerning the application of the Tripartite Declaration.(150)  In addition, the Branch is conducting a research programme on codes of conduct among multinational enterprises which aims to compare the contents of such codes with the provisions of the fundamental Conventions referred to in the Tripartite Declaration (see Publications in preparation below). The Branch also conducts promotional and technical advisory activities, including the facilitation of documents of intention among tripartite partners, which take the form of memoranda of social understanding (for an example of such an understanding, see GB.271/MNE/1).


In preparation

5. Working Conditions and Environment Department (TRAVAIL)

100.  Activities of the Working Conditions and Environment Department include policy-oriented research, data collection and dissemination of information on enterprise initiatives relating to conditions of work, child labour, and occupational safety and health.

Relevant activities -- Conditions of work

101.  An Action Programme on Social Initiatives by Enterprises, implemented in cooperation with ENTREPRISE, aims to increase awareness among the social partners and interested parties concerning human resources and enterprise practices and strategies which improve conditions of work, competitiveness and productivity, and employment security. The outputs envisaged include a comparative analysis, a multimedia package and an Internet database. In addition, a training programme on management practices concerning conditions of work, entitled Work Improvements in Small Enterprises (WISE), is conducted on a tripartite basis, with the cooperation of ILO field offices.

Relevant activities -- Child labour

102.  Work on child labour covers policy-oriented research as well as operational programmes. In mid-1997 TRAVAIL published a preliminary study entitled Labelling Child Labour Products. The study contained a general discussion of social labelling, a more detailed discussion of the issues involved in social labelling as a way of combating child labour, and an emphasis on inherent problems and the need for more in-depth examination of the issue. TRAVAIL research undertaken during the current biennium will follow up on this study by exploring whether labelling has had an impact on child labour and how these schemes would have to work in order to be effective.

103.  Through the International Programme on the Elimination of Child Labour (IPEC), the Office designs and implements field programmes aimed at the progressive elimination of child labour by strengthening national capacities to address child labour problems, and by creating a worldwide movement to combat it. IPEC works together with governments, employers' and workers' organizations, NGOs and other relevant parties to develop and implement measures which focus on preventing child labour, withdrawing children from hazardous work and providing alternatives and rehabilitation for them. Particularly relevant IPEC projects include the following:

104.  In other initiatives, the department, together with the Bureau of Labour Statistics, has been developing statistical instruments to document, inspect and monitor child labour at the national, sectoral and industry level, within the framework of the Statistical Information and Monitoring Programme (SIMPOC). IPEC also oversees the Cooperation Agreement between the International Olympic Committee (IOC) and the International Labour Organization, of January 1998, in which the ILO and the IOC agreed to cooperate in "promoting social justice and human dignity", with particular attention to the elimination of poverty and child labour. Proposals for specific application of the Agreement are under discussion in a joint working group.

Relevant activities -- Safety and health

105.  In its safety and health programmes, the department conducts an Action Programme on Safety Culture which aims to review current practices and elaborate and test a technical standard on occupational safety and health management systems (OSH Guidelines) and a practical guide on safety culture. The management standards are to be based on the relevant international labour standards adopted by the ILO such as the Occupational Safety and Health Convention, 1981 (No. 155) and the Occupational Health Services Convention, 1985 (No. 161). The department is also launching the Proposed Global Programme on Occupational Safety, Health and the Environment (OSHE) to facilitate the improvement of OSHE delivery systems, management tools and programmes, monitoring and information services for prevention of occupational accidents and diseases, and protection of workers and their families. National-level action is aimed at employers' and workers' organizations, SMEs and government entities.


In preparation

6. Sectoral Activities Department (SECTOR)

Relevant activities

106.  Several key tripartite meetings of the Sectoral Activities Department have dealt with voluntary initiatives, particularly in the context of industry/government partnerships. See supra note 3. The tripartite meeting for the Chemical Industries, to be held in February 1999, will review a discussion paper in preparation by the Office on voluntary initiatives in that sector, especially training and education in health, safety and environment. The work is done pursuant to resolution No. 69 concerning future work of the ILO in the chemical industries, Chemical Industries Committee, 18 May 1995.


7. International Institute for Labour Studies (INST)

Relevant activities

107.  Along with the Business and Society programme (see footnote 145 supra), the International Institute for Labour Studies is conducting a Labour and Society programme focusing on "Organized Labour in the 21st Century". The programme follows a double approach: consultations and exchange of views among union practitioners and academics; and case studies to investigate the challenge facing trade unions, assess their varying responses, and identify the policies and activities which have proved particularly successful in different regions of the world, including voluntary initiatives in the private sector. Issues to be covered include union membership among traditional and new constituents; regional and global action by unions; collective action and social alliances; and a new agenda for unions involving action for human rights and partnership in development programmes. Case studies are being organized in Brazil, Chile, United States, Canada, Germany, Sweden, Spain, Lithuania, Japan, Republic of Korea, India, Israel, Ghana, Zimbabwe, Niger and Tunisia.

Conferences, workshops and public lectures


In preparation

8. Office for Inter-Organization Relations (REL/INT)

108.  The Deputy Director-General for Inter-Organization Relations attended and made contributions concerning codes of conduct and social labelling at various meetings sponsored by private associations, including a 1998 annual meeting of the American Society of International Law and a university-sponsored conference on social labelling. She also participated, with the Assistant Director-General responsible for promotion and coordination of enterprise activities, in the 1998 Workshop on Corporate Social Responsibility organized by the World Business Council for Sustainable Development.

9. Office of the Legal Adviser (JUR)

109.  The Office of the Legal Adviser has been charged with the responsibility for preparing this paper for the Working Party on the Social Dimensions of the Liberalization of International Trade, as well as assisting in the development of a standard response to enterprises and others requesting information or assistance pertaining to private voluntary initiatives. In addition, the Legal Office represented the Organization in testifying at a hearing at the European Parliament on codes of conduct, held upon private initiative, on 2 September 1998. In that testimony, the Legal Office emphasized that the issue of codes of conduct was the current subject of discussion within the ILO and, although specific responses would necessarily await the Organization's own policy consensus, the general experiences of the ILO and its unique tripartite perspective were relevant to any consideration of the topic. The testimony of workers and industry representatives is discussed supra section I.

VII. Private initiatives and the ILO :
Issues for discussion

110.  As demonstrated by the previous sections, private initiatives aimed at promoting certain principles and social objectives have become a universal reality with a variety of forms, and constitute an important element in the international debate on the social dimensions of economic development. It was hence particularly appropriate that, following the discussion at the Conference in 1997 on the Report of the Director-General, the Working Party decided to conduct a more detailed examination of the subject.

111.  Even if, for the reasons set out in the previous sections, one may question the viability or sustainability of some of these initiatives, it is reasonable to think that this phenomenon will continue to develop whatever the ILO's position on the matter. Under pressure from the public and from consumers, taken up and maintained by an increasing number of NGOs and by the development of a world network facilitating distribution of information, enterprises will undoubtedly react with increasing numbers and more rapidly by making commitments of various kinds and at various levels to offset risks or benefit from advantages that a good social or environmental image can have in commercial terms. Whether the responses bear any sustainable impact in the workplace, and in what way, are separate and important questions.

112.  The above analysis shows that such initiatives have a number of limits that are inherent to their origins and their very nature. These include:

113.  To return to the question raised by the Chairperson of the Working Party in March 1998, but without at this stage claiming to give any definitive answer, it would seem useful in this final section to examine how this diverse phenomenon, with its positive aspects and current difficulties, may be of interest to the ILO's activities and the achievement of its objectives. This "interest" can perhaps be defined on the basis of the following three questions:

1. Expectations concerning the ILO and its possible role

114.  Judging from the references made to the ILO and its standards in various codes and labels, the interest and expectations given to the Organization in this field are somewhat marginal. However, this should not result in hasty conclusions: on the one hand, many of the direct initiatives seem to confirm a genuine interest in various forms of assistance or engagement; on the other hand, more generally, the proliferation of such initiatives seems to give rise to a growing need for an external source of reference and legitimacy to which the ILO may have the means to respond.

(a) Specific requests addressed to the Office

115.  It is difficult to account for all the requests that have been received and handled by the various Office units. However, a large number of them have been the subject of consultations between several of the units concerned. In the first section we saw that their origins and purposes can vary significantly. Often they come from individual enterprises wishing to obtain information on the ILO's standards in fields of interest to them. Often these intend to consult the ILO on a draft code being prepared in the enterprise in question or to obtain the ILO's opinion concerning systems of accreditation and verification. In other cases, the requests come from associations of enterprises or employers' organizations seeking technical assistance in the design of programmes for the inspection of worksites and for the training of future inspectors. Finally, there are those from trade unions seeking ILO assistance to strengthen their communication and negotiation capacity at the global level. In this connection reference can also be made to the suggestions at the recent hearing in the European Parliament (mentioned in section I.1) by an employers' representative, speaking on behalf of the UNICE, and by a workers' representative.

116. So far the Office has in its replies limited itself largely to providing information on its standards and on the Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy, and has drawn attention to the fact that the phenomenon is being studied in the Organization and may in time be the subject of more detailed replies. However, it is difficult to continue to give such replies, particularly because they only generate further requests. Consideration must hence be given to whether the Office should provide substantive replies and in what form. The answer to this question naturally depends on the following considerations.

(b) The general need for an external frame of reference

117.  In section III it was noted that the proliferation of different initiatives in terms of the issues they address and their methods of implementation may pose problems of compatibility between such initiatives (particularly in the case of suppliers or subcontractors who must simultaneously meet differing requirements of different codes), and may cast doubts on the credibility in the mind of the public and of consumers regarding the real scope of the commitment made in relation to its purpose (for example, the elimination of child labour) and its implementation.

118.  In these circumstances, what developments are possible? The first is that the confusion and scepticism caused by the widely varying standards and systems of supervision lead to the weakening of the phenomenon. This is probably not the most plausible development given the diversity of interests involved in and served by the phenomenon. The second might be that the need for standardization and harmonization will be addressed by the market itself, but this seems more or less plausible according to whether one examines, respectively, the verification systems or the content of the codes. As regards verification systems there are a large number of auditing firms and councils which are keen to offer their services in this expanding market, and the prospect of the market's taking up this need seems initially quite plausible. As we have seen, this possibility gives rise to a number of questions concerning, on the one hand, the lack of experience and specific competence of such services in social matters and, on the other hand, their cost, which may restrict them to enterprises in developed countries and will disadvantage local enterprises in developing countries, as can be seen by the partiality of certification systems in effect for ISO 9000 and ISO 14000 (see section II.2). It is understandable in such circumstances that people should naturally turn to the ILO to obtain such services, thinking that, in addition to its almost axiomatic competence in this field (particularly in view of the role it has played with governments in developing labour administration), the ILO offers the guarantee of an objective and non-profit-making examination.

119.  As for the content of principles, the prospect of their privatized standardization seems even more haphazard. In the present context standardization means defining the exact meaning of the concept or objectives, such as the elimination of child labour, or freedom of association, or discrimination, instead of leaving everyone free to choose the most convenient definition. As noted in section II.3, a number of commentators have expressed the opinion that the experience of the ISO, particularly with the ISO 14000 series on environmental systems management, demonstrates the limitations of the voluntary standard-setting process in fields that are not strictly technical. The consensus method of work, and the predominant position in technical committees of MNEs and industrialized government member bodies, have been blamed for results limited to procedural rather than substantive requirements and reflecting the least common denominator rather than more stringent targets.

120.  It is not therefore surprising to find that, here too, the quest for a reference framework and for legitimacy with regard to the content of principles results quite naturally in some of the more ambitious initiatives turning to the ILO and its standards.(151)  However, there is a risk that the use of ILO standards is done in a unilateral and selective manner and may indeed serve the appearance of legitimacy by giving the impression to the public and the consumer that the ILO is associated with the initiative.

121.  In any case, it seems that, whatever its wishes, the ILO will have difficulty behaving as if a phenomenon that invokes its name does not exist. Naturally, it is perfectly free not to react. However, it seems important that any position adopted should be the outcome of considered examination, and not a stop-gap measure. This supposes that the ILO must assess the possible impact of the phenomenon and its development -- with or without its involvement -- in terms of its own objectives.

2. Possible implications for the achievement of the ILO's objectives

122.  It should be clear that the demand for and possible expectations concerning various forms of ILO action, if they materialize, cannot in themselves justify the ILO's necessarily responding positively, even if for the ILO it represents a considerable source of possible income and influence. The role that the ILO could be called on to play should largely derive from how it can direct the impact of the phenomenon towards its own values and the achievement of its objectives. This involves considerations on two levels: the purpose of the initiatives, that is, their motivation; and their effects, that is, their concrete impact on the achievement of the ILO's own objectives.

(a) Motivation/purpose

123.  The ILO should welcome a phenomenon which, as noted, seems initially to derive from social concerns similar to its own and to reflect the gradual acquisition by the public and by enterprises of its values as expressed in its Constitution, even if this is not expressed specifically in the wording of its instruments.

124.  However, one question may be asked, given that the development of such initiatives is at least partly a market phenomenon, and that in the marketplace motivations may be varied and even contradictory. They may equally derive from a genuine "market ethic" (which should logically result in some degree of convergence with the objectives and activities of the ILO) or from an "ethics market", with demand coming, or supposedly coming, from the public and consumers, and supply coming from enterprises acting either spontaneously or in response to market demands as communicated to them by pressure from NGOs and other associations. It should be fully realized that the logic of this latter tendency means ultimately bypassing the ILO: it would suffice to inform consumers of the conditions in which the goods or services they purchase are produced for consumers to exercise their own judgement and choose from among preferences that it would in such circumstances be impossible to arrange in any hierarchy of values. There would hence be absolutely no need for any national or international prescription of measures; it would suffice to ensure informed choices by means of appropriate standardization, which by the same reasoning may be thought capable of being ensured by the mechanisms of private competition, even if this might give rise to the questions discussed above.

125.  It hence seems preferable, in assessing the value of such initiatives to the ILO, not to become too concerned with their vaunted or supposed motivation, but instead to concentrate more pragmatically on their concrete impact on the achievement of the ILO's objectives.

(b) Concrete impact

126.  The question of the impact of private initiatives on the implementation of the ILO's objectives, as developed and spelled out in its standards, may initially seem largely irrelevant. For enterprises, codes of conduct are an issue of good management, and not a question of standards. By contrast, the ILO's instruments are not addressed as such to enterprises. Some of their provisions may be relevant to enterprises, but most of them concern the adoption of certain objectives, programmes and general policies which are only meaningful to governments.

127.  In reality, however, things are more complex. First, there is no watertight separation between the two. In 1977 the Governing Body of the ILO adopted a text -- the Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy -- which was addressed not only to the ILO's traditional constituents, but also to multinational enterprises, drawing to their attention a number of instruments that it presented as being of global relevance (but without distinguishing specific provisions that might concern them more directly) so as guide them in their social practices. Moreover, as noted in section I, although it is governments that ratify Conventions, enterprises are increasingly seeking guidance on how to apply the standards in the workplace.

128.  It would also seem that, in certain circumstances, private initiatives and international labour standards can operate both in parallel and in competition with one another in relation to governments and enterprises. For example, it is perfectly conceivable that a local enterprise in a country that has ratified a Convention might offer superior conditions to those foreseen in a code of conduct or label by scrupulously respecting the provisions of national law incorporating the provisions of that Convention. However, that will not suffice for the local enterprise to obtain the advantages deriving from a code or label, that is to say, access on an equal footing to markets in developed countries, as it is not linked to the network of subcontractors and affiliates covered by those codes and labels. For this reason, according to UNIDO,(152)  it is more advantageous for export enterprises to follow ethical codes than international standards on social and environmental issues, including those of the ILO. At the same time, it might be thought that such a claim could risk discouraging governments themselves from ratifying or applying ILO Conventions.

129.  Nevertheless, one cannot dispute the fact that, in the case of a country that has not ratified the corresponding ILO Conventions, private initiatives automatically bring an undeniable advantage in terms of social progress. Where all workers are not offered the protection resulting from ratification of the relevant Conventions, such initiatives at least enable workers in the sectors concerned by the codes or labels to enjoy conditions closer to those provided by those standards, and perhaps in this way to bring about gradual improvements in working conditions in the country. Again, this advantage inevitably raises a number of questions:

130.  Is better complementarity possible? The issues described above do not mean that the initiatives cannot make a positive contribution to the attainment of the ILO's objectives. They simply suggest that such initiatives and the attainment of the ILO's objectives will not automatically converge. Perhaps some better form of complementarity could be envisaged between, on the one hand, priorities which are defined externally and on a selective basis and, on the other hand, the process of social progress which develops within each Member and is inclusive in approach. In this regard the application of a small number of simple principles that are already stated in other ILO texts would appear relevant. These include:

131.  The answer to this question depends on the assessment made by the Working Party and the position it wishes to adopt as a result.

3. What possible positions and follow-up action might be envisaged?

132.  In the light of the above considerations and the requests and suggestions made, various positions on this issue are possible. They could range widely, from a relatively minimalist position to a more active and even interventionist position, or perhaps merely one of providing support services. It seems useful for the discussion to illustrate these various positions, but it is stressed that the intention is not to propose decisions, but rather to permit as complete a discussion as possible.

(a) The minimalist position?

133.  This is the position that the Office has so far adopted in response to the requests received and approaches made to it. In responding to the requests, the Office has been guided by the well-established principles that guide all its activities, including:

134.  It would seem possible, without essentially departing from this minimalist position, to envisage a number of activities that are purely and simply part of the responsibilities of the Office regarding information and analysis. In particular, the Office could establish a resource centre on the Internet to track, report and analyse voluntary initiatives relevant to enterprise labour practices. It could also use its sectoral expertise to collect and analyse information concerning industry-specific developments and their implications for ILO activities. Sectoral meetings for that purpose, which are being proposed, would facilitate further research and offer an opportunity for sectoral constituents to engage in discussion of these initiatives, in which many find themselves unrepresented due to the lack of transparency in their development and implementation.(153) 

135.  A research agenda could include pilot studies with a diverse set of enterprises to analyse the impact, if any, of codes of conduct on labour practices and approaches used to administer voluntary initiatives. Particular focuses could include assessment methods and the ways in which enterprises, particularly in developing countries, respond to the challenges, both technical and managerial, inherent in responding to the demands to use codes of conduct made by their business partners.

(b) A role in providing support services?

136.  In so far as the assessment of private initiatives seems positive from the point of view of the achievement of the ILO's objectives, the Office could offer information and advice in two areas:

137.In this general perspective, the Office might consider one or more of the following activities or publications, some of which would, however, be based on the assumption that it had already conducted some of the applied research projects mentioned above:

(c) A proactive position of engagement?

138.  This position would mean drawing on the interests and aspirations expressed in such initiatives and more specifically directing them towards the ILO's aims and purposes, including, where appropriate, endeavouring to rectify some of the problems they pose. It should be stressed that this position may derive equally from either a positive or less encouraging assessment of the global impact of the phenomenon on the ILO's objectives. Setting aside for the moment the possibility that the ILO might encourage enterprises that desire to express their support for certain social principles or rights, and the possibility that they might do so by contributing to the ILO programmes that will have to be established to provide assistance to constituents to improve the application of fundamental principles and rights in the framework of follow-up on the implementation of the ILO Declaration of June 1998, this active position could result in two different but possibly complementary means.

  1. A text that would aim to recommend, to enterprises wishing to make certain commitments or initiatives, certain forms of good practice regarding the content and methods of giving effect to their commitments. However, there is still the question of how the ILO can do this, since its jurisdiction is normally exercised in relation to and through States. While it is not in legal terms inconceivable to take action on the phenomenon by means of classic standard-setting action,(154)  nevertheless such a solution rests on the supposition that there is a consensus which for the moment is not apparent. However, there is another possibility, the model for which is offered by the Tripartite Declaration concerning Multinational Enterprises and Social Policy.

    As is well-known, this text is addressed not only to the ILO's traditional constituents (governments and employers' and workers' organizations), but also, directly and specifically, to multinational and local enterprises so as to call on them to lend consideration to a number of principles on a voluntary basis. Naturally, this would not mean revising this text, but simply to draw on it and perhaps supplement it so as to call on enterprises wishing to draw up codes of conduct or labelling systems on social issues to draw on a number of uniform principles in the content of their codes (taking account, for example, of the principles set out above in section VII.2) and in the effective methods of implementation and assessments of performance.

  1. A second possibility within this proactive position might consist in supplementing recommendations concerning the content of the initiatives with recommendations on how to implement them transparently and credibly. In this regard it is legitimate to ask whether, account being taken of some of the requests, the ILO should propose, to enterprises wishing voluntarily to subscribe to commitments based on its recommendations, a reference system and procedures that would make it possible to guarantee genuine verification without thereby necessarily going so far as itself to supply the auditing and certification services, which if provided on a large scale might take it beyond the limits of its resources and even of its competence and status.

* * *

139.  It should again be stressed in conclusion that the distinction between the different possible positions that can be adopted is largely of didactic value. There is no clear-cut division between the different alternatives, nor is it a question of suggesting that the position to adopt can be determined once and for all. The choice can easily evolve. In view of the wide range of interests at stake and the fact that the chief comparative advantage of the ILO in this debate is that it can, through its various bodies, offer a forum in which most, if not all of those concerned can make their voices heard, it would seem quite natural for the discussion to take place in the ILO so as ultimately to identify the elements of a consensus on the types of action and the contribution that it could usefully undertake in relation to this phenomenon. For the moment, the issues and ideas discussed above are not intended as a basis for decisions, but only for an initial exchange of views. According to the guidance emerging from that exchange, the Director-General might be invited to present more specific proposals to the Governing Body at a forthcoming session.

Geneva, 19 October 1998.

Appendix: Tables


1. J. Murray: "Corporate codes of conduct and labour standards" in R. Kyloh (ed.), Mastering the challenge of globalization: Towards a trade union agenda, ACTRAV Working Paper (ILO, 1998), p. 60.

2. The "triple bottom line" concept has received criticism for its potential to compromise three disparate forms of value on one balance sheet. See, e.g., N. Mayhew: "Trouble with the triple bottom line", in Financial Times, 10 Aug. 1998.

3. For difficulties facing developing country enterprises in the context of various standards requirements, see Restrictive Trade Effects of Standards, Technical Regulation and Conformity Assessment Procedures, WTO document G/TBT/W/42, background paper prepared by the secretariat, 28 Apr. 1997. See also R. Kumar, N. Gessese and Y. Konishi: Responding to global standards: A framework for assessing social and environmental performance of industries: Case study of the textile industry in India, Indonesia, and Zimbabwe (UNIDO, 1998).

4. See, e.g., A. Wild: A review of corporate citizenship and social initiatives: "Social citizenship -- What's going on ... and why?", Preparatory reader for the Employer Meeting on Corporate Citizenship and Social Initiatives, 1-2 Oct. 1998, New York.

5. Defending values, promoting change: Social justice in a global economy: An ILO agenda, Report of the Director-General, International Labour Conference, 81st Session, 1994.

6. See further discussion of the Tripartite Declaration in section II.2 infra.

7. The role of enterprise in addressing issues of social or community development beyond specific issues encountered or reflected in the workplace is beyond the scope of this study.

8. As recognized in the ILO Declaration of Fundamental Principles and Rights at Work, adopted at the 86th Session of the International Labour Conference, principles concerning fundamental labour rights include freedom of association, the right to collective bargaining, elimination of all forms of forced or compulsory labour, the effective abolition of child labour and the elimination of discrimination in respect of employment and occupation. (Declaration, para. 2.)

9. The study in the Multinational Enterprises Bureau is described in GB.270/WP/SDL/1/3, para. 20. See also infra section VI.

10. For the purposes of this paper, the term "stakeholders" refers to "individuals and groups who may affect or be affected by the actions, decisions, policies, practices or goals of an enterprise". A.B. Caroll: Business and society: Ethics and stakeholder management (Cincinnati, Ohio, South Western College Publishing, 1996), p. 74. Some theorists have characterized the relationships of various stakeholders to enterprise as "primary" (workers, investors, customers and business partners/suppliers) and "secondary" (all others); others refer to categories such as "core" (essential for enterprise survival), "strategic" (vital to organization), and "environmental" (all others). idem., pp. 72-82. The activities of four categories of primary stakeholders are discussed in this paper: (1) employees (and their representative organizations); (2) owners or shareholders/investors (and their representative organizations); (3) customers or consumers; and (4) suppliers or any other business partners in contract or equity relationships with enterprise. Other stakeholders whose activities are relevant to private initiatives include non-governmental organizations, political and social groups, communities, governments, and intergovernmental organizations. For one description of "stakeholder theory", see L. Preston, "Redefining the corporation: Stakeholder theory in international perspective", Occasional Paper No. 78 (College Park, Maryland, University of Maryland, 1996).

11. UNOCAL, a US-based oil company, is facing allegations of forced labour in the construction of an oil pipeline in Myanmar in a lawsuit in the United States calling for civil damages. The sporting goods company, Adidas, has been sued in Hong Kong by Chinese dissidents who claim to have made Adidas soccer balls as prisoners in a Chinese labour camp, e.g., "Adidas said to use slave labor", in Washington Post, 19 Aug. 1998. The US-based multinational, NIKE Inc., is facing a lawsuit by private citizens in California alleging negligent misrepresentation, fraud and deceit, and unfair business practices, for adopting and not properly implementing a code of conduct.

12. Public initiatives excluded from the scope of this study are, among others, policy decisions of the OECD, G-7, G-8, G-15 and other multilateral groupings, developments cornering social charters in the context of regional economic groups (e.g. MERCOSUR and SADC), laws providing preferential treatment and terms of trade (e.g. NAFTA, European Union and United States labour conditions on granting of Generalized Systems of Preferences), import restrictions (e.g. United States Tariff Act of 1930, as amended by, inter alia, P.L. 105-61, 10 Oct. 1997 (child labour import ban)), selective public procurement and preference laws, and unilateral or multilateral sanctions imposed as a matter of public policy.

13. Within the ILO, in the past several decades, more than 20 codes of practice have addressed issues of occupational safety and health in offering guidance to employers' and workers' organizations as well as governments. ILO codes of practice are developed through a tripartite process of negotiation and approval of the Governing Body. Sectors of focus include mines, agriculture, forestry, construction and public works, shipbuilding and repairing, iron and steel, prevention of major industrial and maritime accidents, while specific risks addressed include ionizing radiation, noise and vibration, exposure to airborne harmful substances, safety in the use of chemicals at work.

14. Commitments include providing national treatment to products from any WTO member State, ensuring that no unnecessary obstacles to international trade are presented through voluntary standard-setting activities, and using relevant international standards where they exist. See Code of Good Practice for the Preparation, Adoption and Application of Standards, paras. D-F, Annex 3 to Agreement on Technical Barriers to Trade, item 7 to the Agreement establishing the World Trade Organization. The TBT Agreement itself is binding on all WTO member States as a part of the WTO framework agreement. See also WTO document G/TBT/CS/2/Rev.4 and Addendum, 11 Feb. 1998 (reporting 92 standardizing bodies from 69 Members accepting Code of Good Practice since 1 Jan. 1995).

15. For example, FAO Code of Conduct for Responsible Fisheries (1995), WHO International Code of Marketing of Breastmilk Substitutes, FAO International Code of Conduct on the Distribution and Use of Pesticides. However, according to a recent UNIDO report, "... for export-oriented firms, the dictates of large international buyers reflected in their own codes of ethics are more instrumental in bringing about improvements in social and environmental performance than guidelines recommended by international agencies such as the International Labour Organization, World Bank, World Health Organization, etc." Responding to global Standards, op. cit., p. 5.

16. In an attempt to pioneer a "practical and feasible" method for use in developing countries, UNIDO has applied principles from ILO instruments, together with additional criteria, to create a UNIDO-defined "IDEAL" set of "social performance indicators" used to assess and compare ("benchmark") national legislation and enterprise practices. See Responding to global standards, op. cit., pp. 45-46, and Annex 3, pp. 45-57.

17. Each member body represents the national body "most representative of standardization in its country". The member body is to serve as the liaison between interested parties in their country and the process of developing and adopting standardization in ISO's many technical committees, subcommittees, and working groups. See Internet Approximately 70 per cent of member bodies are government agencies or public organizations; some are private organizations, in many cases, subject to some public regulation. ISO, ISO memento (Geneva, International Organization for Standardization, 1991), p. 3. ISO processes have been analysed as dominated by the participation of MNEs and developed government member bodies. See, e.g., N. Roht-Arriaza: "Shifting the point of regulation: The International Organization for Standardization and global lawmaking on trade and the environment", in 22 Ecology Law Quarterly 479 (1995). An estimated 600 regional, national and local standardizing bodies operate in specific technical fields.

18. The ISO 9000 quality control standards contain guidelines for an enterprise to use both in its own operations and in specifying contract requirements for suppliers and subcontractors. The ISO 14000 series addresses environmental systems management but, in a result required by the consensus method in which ISO works, does not contain substantive targets, such as specific emissions standards. Some commentators have argued that ISO 14000 demonstrates a limitation of voluntary standard-setting processes: the consensus required results in acceptance of a "least common denominator" rather than more stringent targets set in mandatory regulation. See generally R. E. Cheit: Setting safety standards: Regulation in the public and private sectors (1990).

19. See Agenda 21, Ch. 4(B)(c), 14 June 1992, reprinted in United Nations, Report of the United Nations Conference on Environment and Development (1993). Agenda 21, a legally non-binding document, was signed by 156 countries at UNCED.

20. Considerations leading to the negative decision included fear of a "certification circus" if another generic management standard were added to ISO 9000 and 14000. ISO 9000 NEWS 5/1998, p. 11.

21. Unlike the ISO accreditation system which engages multiple national accreditation bodies, SA8000 certification is offered only through an accreditation agency set up by CEP and working with private inspection enterprises which it accredits under non-public criteria. See "SA8000: Management system standard for social accountability", ISO 9000 NEWS 5/1998, p. 13. The accredited agencies themselves are not necessarily certified as in compliance with SA8000. SA8000 is discussed further in section III infra.

22. M. Egger: "ISO 21000" -- private Normen für die soziale Verantwortung der Unternehmen? Ein Diskussionsbeitrag (Brot für alle Schweiz, Bern, 1996).

23. Employers' handbook on child labour; A guide for taking action, IOE, op cit., p. 49. Documents such as guides, handbooks, manuals or auditing checklists may exist to guide the implementation of the code in practice, but in many cases, there are no instructions as to how to implement or enforce code provisions which may be quite general in nature. Manuals or handbooks that exist without codes are beyond the scope of this paper.

24. See, e.g., Statement of the United States Council for International Business: Codes of conduct: Old solutions to old problems (1997).

25. Subscription codes, discussed infra, provide an external, or third-party, basis of commitment and/or monitoring for enterprises which choose to endorse them. The Sullivan Principles (1977) aimed to influence United States companies doing business in South Africa, and the MacBride Principles (1984) sought to influence United States firms operating in Northern Ireland.

26. See generally section II supra.

27. H. White: "Global outsourcing and corporate accountability", in Accountability Quarterly 4, Summer 1997.

28. Notably, UNCTAD sets at 44,000 the number of "parent enterprises" operating in the world today. World Investment Report, 1997, p. xv.

29. See, e.g., C. Ferguson: A review of UK company codes of conduct (Department for International Development, Aug. 1998) (survey of 18 UK company codes and three "sample" codes); Council for Economic Priorities: International sourcing report (survey of 360 companies for workers' rights in codes) (1998); C. Forcese: Commerce with conscience? No. 1 (1997) (survey of 98 Canadian enterprises); International Chamber of Commerce, Code survey (1998); United States Department of Labor: The apparel industry and codes of conduct: A solution to the international child labor problem? (1996); Investor Responsibility Research Center: The sweatshop quandary: Corporate responsibility on the global frontier (1998) (child and forced labour).

30. For ILO work, see section VI infra. See also World Investment Report: Transnational corporations, employment and the workplace (UNCTAD, 1994), and Open trading: Options for effective monitoring of corporate codes of conduct (1997). A project in progress at the OECD, commissioned by the Trade Committee, promises an inventory of codes used by enterprises based in OECD countries focused on issues including "fair employment", "fair business practices" and "respect for human rights". Similarly, the European Commission has commissioned a forthcoming study, and a private initiative by Mr. Howitt at the European Parliament focuses on implications from an overview of trends.

31. See, e.g., G. van Liemt: The social policy implications of codes of conduct, with particular reference to the relations between companies adopting such codes and their supplier and subcontractors, presented at Global Production and Local Jobs: New Perspectives on Enterprise Networks, Employment and Local Development Policy, Geneva, 9-10 Mar. 1998, and relevant discussion chaired by Mr. A. Abate at that conference.

32. For example, the International Union of Food, Agricultural, Hotel, Restaurant, Catering, Tobacco and Allied Workers Associations (IUF) brought tea producer and distributor affiliates together in 1997 to spur enterprise initiatives in the tea industry. In addition, multi-sector coalitions involving codes of conduct are emerging in efforts against child labour developed through a coalition of workers in the surgical instruments industry in Pakistan and public sector health care workers in developed countries.

33. See, e.g., EU Council Directive No. 94/45/EC, 22 Sep. 1994 (Official Journal of the EC, 30/09/94, No. L 254, pp. 64-72), amended by EU Council Directive No. 97/74 of 15 Dec. 1997 (Offficial Journal of the EC, 16/01/98, No. L 10, pp. 22-23), implementing 1989 European Community Charter of Fundamental Social Rights of Workers, which mandated multinationals operating in Europe to establish works councils for information and consultation with workers on selected issues in enterprise operations.

34. ITSs involved in developing codes report that they are devoting considerable financial and human resources to communication with workers at local levels. Such communication may demonstrate, in contrast to prevailing situations where codes are developed without trade union involvement, that the development of codes between ITSs and multinationals has tended to enhance local workers' awareness of the negotiations and content of codes that cover their industry or workplace.

35. See, e.g., United States Apparel Industry Partnership (which consists mainly of US-based MNEs in the TCF sector and US-based NGOs). See also discussion of SA8000, infra.

36. The Board which runs the SA8000 initiative of the Council on Economic Priorities (CEP) contains one smaller company, Eileen Fisher, out of 16 enterprises in total, and some of the initiatives have SME subscribers represented through industry associations (e.g. Clean Clothes Campaign, Draft Banana Charter).

37. The comparatively longer track record of Project Equality, an initiative begun in the United States for review and disclosure of equal employment opportunity and treatment, incorporates religious organizations and corporations as sponsors and reportedly exerts influence by publishing an annual Buyer's Guide accrediting over 1,500 employers of those inspected.

38. Examples include the ICC Business Charter for Sustainable Development, and the Caux Round Table Principles for Business, developed in 1994 by the US-based Minnesota Center for Corporate Responsibility.

39. See, e.g., International Code of Conduct for the Production of Cut Flowers adopted in Aug. 1998 by the IUF and several local trade unions and NGOs.

40. The ICFTU Basic Code seeks to help develop the codes of individual companies or their subcontractors, as well as industry associations and employers' organizations. In a similar initiative led by Hyundai and Daewoo trade unions in Korea with the Korean-based NGO, People's Solidarity for Participatory Democracy, the partners developed their own charter as a means to monitor the performance of Korean enterprises operating overseas in response to the 1996 Declaration on codes of conduct for those enterprises issued by the Korean Council of Economic Organizations.

41. See, e.g., Business for Social Responsibility Matrix of Company Codes of Conduct, Oxfam Clothes Code Campaign Minimum Standards and Principles, Interfaith Center on Corporate Responsibility 1995 Principles for Global Corporate Responsibility and Amnesty International Human Rights Principles for Companies, which refers widely to international labour standards and the ILO Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy.

42. The United Kingdom Government, in contrast, plays an important role in the development of the Ethical Trading Initiative, including financing and observation on the Board. Other governments have opted for the development of guides or guidelines relating to codes of conduct. The Canadian Government's Voluntary Codes Guide, along with various other information resources, includes an "Eight-step model for developing codes" reportedly prepared "by a multi-stakeholder working group". The Australian Government has published a Guide to fair trading codes of conduct (1995), and New Zealand has Guidelines on developing a code of practice (1993).

43. The principles, which are quite broad and do not refer to ILO standards, are supported by the Government's Best Global Business Practices Program, which provides information resources, conferences and awards relating to socially responsible labour practices in enterprise.

44. The codes were collected by Office sources and through secondary literature (see Scope and methodology, section I supra). More than 80 per cent of the codes belonged to MNEs, primarily based in developed and newly industrialized countries; some were developed by industry associations and employers' organizations. Some of the codes were developed by enterprises or enterprise associations together with workers' organizations and/or NGOs. A systematic collection and review of guidance documents or manuals used with codes is still needed.

45. For example, Council on Economic Priorities, SCREEN, op. cit., reported that, of 38 per cent responding, 99 per cent had sourcing guidelines addressing basic workers' rights; United States Department of Labor survey, op. cit., reported that of 42 responding, 36 had child labour policies with half or less with other labour rights standards; C. Forcese, op. cit., survey of Canadian codes reported that, of 43 responding, 49 per cent had international codes of conduct with labour provisions in 46 per cent of those.

46. Dominant references appeared to reflect concern as to particular problems in the following industries: basic metal production (Convention Nos. 100, 111), chemicals (Conventions Nos. 111, 138), construction (Convention No. 138), forestry (Convention No. 111 with notable worker/enterprise codes including Conventions Nos. 87 and 98 as well as Conventions Nos. 29 and 105), mechanical and electrical engineering (Convention No. 111), oil and gas production (Convention No. 111 and hybrid codes indicating Convention No. 169), textile, clothing, leather and footwear (Convention Nos. 111, 138 with worker/enterprise and hybrid codes indicating all core labour standards), and transport equipment (Convention No. 111). Service sectors reflected the following concerns: commerce (Convention No. 111 primarily), financial services (Convention No. 111), postal and other communication services (Convention No. 111 primarily), utilities (a worker/enterprise code had all core labour Conventions), and hotels, tourism and catering (only a worker/enterprise code, with references to Conventions Nos. 87 and 98).

47. In one example using existing standards to create a self-defined standard, a UNIDO study developed a UNIDO IDEAL standard by incorporating both ILO norms and SA8000 standards, but found that the ILO and SA8000 systems fell somewhat short of the UNIDO-developed IDEAL social performance indicators for occupational safety/health conditions, non-discrimination practices (which included ILO Conventions Nos. 87 and 98), communication/work environment, working hours, wages and training/education/awareness. Responding to global standards, op. cit., p. 37 (table 2).

48. In a particularly marked reference, an appendix to the Ethical Trading Initiative's Base Code states that the "tripartite structure of the ILO ... together with the technical expertise of this Organization in all matters relating to the world of work, make the ILO the authoritative and legitimate source of international labour standards".

49. An enterprise in the TCF sector presented a rare exception: "Any workers producing products manufactured for and sold by KappAhl must be provided with fair wages and decent working conditions and the international labour standards according to ILO Conventions Nos. 29, 87, 98, 100, 105, 111 and 138 must be observed." Code of conduct: Labour relations of KappAhl (in pertinent part).

50. Among those were Liz Claiborne and Levi-Strauss. Reebok's code had only a general reference to international human rights standards.

51. The four with ILO Convention references were KappAhl and three hybrid codes (Clean Clothes Campaign, Fair Trade Charter for Garments and SA8000). Five of the six with national law references were corporations with United States nationality; the sixth was a model code. The primary United States law referenced, the Fair Labor Standards Act, has been determined by United States courts to have no extraterritorial effect.

52. "Everyone is entitled to be treated with respect as a person, regardless of role or individual differences." WMC Mining Company -- Code of conduct.

53. The Sullivan Principles, of course, went farther than the workplace in supporting "the ending of all apartheid laws in South Africa".

54. "Employees must not be discriminated against because of personal characteristics or beliefs." Jones Apparel Group -- Business partner standards; "[p]erceptions of equality naturally vary between different cultures and it is not the business of a company to smooth out the differences. However, within SKF, in all the locations where we are active, we strive to maintain equality -- between the sexes, between different generations, different nationalities, races and creeds. We view equality not only as an ethical principle but also as an efficient working rule. It promotes a good team spirit" SKF -- Our views on ethics and morals.

55. Convention No. 111 provides that, "for the purpose of this Convention the term 'discrimination' includes ... any distinction, exclusion or preference made on the basis of race, colour, sex, religion, political opinion, national extraction or social origin, which has the effect of nullifying or impairing equality of opportunity or treatment in employment or occupation ...", Convention No. 111, Art. 1(1)(a) and chapeau.

56. "We are fully committed to a policy of non-discrimination in employment and to the cause of equal employment and advancement opportunity for all ..." Reynolds Metals Company -- Business conduct guide.

57. The codes included those of the enterprise KappAhl as well as those drafted by Clean Clothes Campaign, SA8000 and the International Code of Conduct for the Production of Cut-Flowers proposed by IUF, several local trade unions and a few NGOs.

58. "We believe that children should not be unlawfully employed as labourers ... We believe that if children work, it should not interfere with mandated education." Starbucks -- Statement of beliefs (Framework for a code of conduct).

59. One enterprise specified the age of 14, except for a lower age for light work if permitted by law, in one of its codes but, in another, permitted children over 12 "to be employed part time if still in education". Compare Body Shop Declaration of Compliance with Code of Practice with Body Shop -- Trading Charter and Statement of Human Rights Principles. The ages of 14 and 15 roughly mirror the provisions of Convention No. 138, though not necessarily with attention paid to the conditions and safeguards contained therein.

60. "Sara Lee will not knowingly use suppliers who employ workers in violation of the local mandatory school age, or under the legal employment age in each country. In no case will Sara Lee procure goods or services from firms employing workers under age 15." Sara Lee Corporation -- Supplier selection guidelines.

61. "Child labour must not occur. Only workers aged 15 and over, or over the age of compulsory education if higher, may be employed (ILO Convention No. 138). Exceptions to this rule may only be made if national legislation provides otherwise." Agreement between IKEA and the IFBWW.

62. Member companies, and their contractors, are expected to agree that "no forced or prison labour is employed, that workers are free to leave once their shift ends ..." International Council of Toy Industries Code of Business Practices; vendors of K-Mart "must ensure that [no] merchandise is made in whole or in part using any ... forced or prison labour." K-Mart -- Corporation Vendor Agreement.

63. "Sara Lee will not knowingly use suppliers of either raw materials or finished products that have been produced by prison labor or forced labor, or services provided by such labor." Sara Lee Corporation -- Supplier selection guidelines.

64. "Columbia Sportswear will not enter into a partnership with any suppliers or contractors who use any form of forced labour -- prison or otherwise." Columbia Sports Wear -- Company standards and business practice guidelines.

65. "We will not do business with firms that use forced prison labor or with companies whose contractors use such labor. We will also not purchase raw materials from companies that use forced prison labor. Furthermore, we will not transact business with organizations that use corporal punishment or any form of coercion." L.L. Bean Inc. -- Code of conduct.

66. "We will avoid sourcing from manufacturers where labour (e.g. child labour) is deemed to be exploited contrary to the rules and guidelines of the [ILO]." Empire Stores Group-Redoute Group -- Aims and ethics. In one code, a verbatim incorporation of excerpts of the definition of forced labour under Convention No. 29 failed to include the essential exceptions of Article 2 of that Convention. See SA8000.

67. An industry association code read: "Members shall only do business with partners whose workers are in all cases ... allowed the right of free association and not exploited in any way." Athletic Footwear Association -- Statement of guidelines on practices of business partners.

68. The company aims to "... foster a corporate culture that enhances individual creativity and teamwork value, while honouring mutual trust and respect between labour and management" Toyota Motor Corporation Guiding Principles (1997 revision).

69. One company code, reported to be on every employee's desk, stated that enterprise policy sought "[t]o operate the business in such a way that employees don't feel a need for representation by unions or other third parties. Where employees have elected in favour of -- or are required by law to have some form of -- union representation, Caterpillar will endeavor to build a sound company-union relationship" Caterpillar -- Code of worldwide business conduct and operating principles. Another, less union-friendly enterprise code read: "The company believes in a union-free environment except where laws and cultures require [SKP] to do otherwise ... believes that employees themselves are best able to voice their concerns directly to management." Sara Lee Knit Products -- International operating principles.

70. Another code read: "Employees shall be encouraged by lawful expression of management opinion to continue an existing no-union status, but where employees have chosen to be represented by a union, management shall deal with the union in good faith" (followed by explanation of good faith dealings) DuPont -- Labour relations policies and principles.

71. These included hybrid codes between enterprise/workers' organization(s), a few model codes and one enterprise-generated code. In what may be the earliest enterprise/worker code, the BSN group, now Danone, in the food and drink industry, formulated a Common Viewpoint with the IUF in 1988, followed by a series of implementation agreements. Among other matters, the 1988 document declared the agreement of the two parties to promote the "implementation of trade union rights as defined in ILO Conventions Nos. 87, 98 and 135", Common Viewpoint IUF/BSN, 23 Aug. 1988. In 1994, the parties further agreed to specific means of promotion of trade union rights, including "efforts to provide economic and social education and information to the entire workforce as well as their representatives", and defined "workers' representatives" by quoting Article 3 of ILO Convention No. 135. IUF/BSN Joint Declaration on Trade Union Rights, 25 May 1994.

72. For example, the company "respects its employees rights to legally organize and bargain collectively" ALCAN Aluminum Ltd. -- Code of conduct (emphasis supplied); and "All workers should be free to join associations of their own choosing, and they should have the right to bargain collectively. We don't accept any disciplinary actions from the factory against workers who choose peacefully and lawfully to organize or join an association." Hennes & Mauritz -- Code of conduct (emphasis supplied).

73. The code, discussed supra at footnote 21, was created by a non-profit institute, the Council for Economic Priorities, through a working group predominantly comprised of enterprise yet including NGOs and workers' representatives. It states a dual obligation of the company to "comply with national and other applicable law" and to "respect the principles of the following instruments: ILO Conventions [29, 105, 87, 98, 100, 111, 135, 138, 155, 159, 177] ... Universal Declaration of Human Rights [, and] ... Convention on the Rights of the Child". In its section on freedom of association and collective bargaining, the "company shall respect the right of all personnel to form and join trade unions of their choice and to bargain collectively; ... [and] in those situations in which the right to freedom of association and collective bargaining are restricted under law, facilitate parallel means of independent and free association and bargaining for all such personnel ...", SA8000, ss. II and IV, 4 (in pertinent part).

74. Some sought to meet both national law and match the industry standard. Others aimed to pay national law levels or prevailing local wages, whichever was higher. Still others resorted to industry standards if national law on wages did not exist.

75. "Compensation must be fair and adequate ..." Johnson and Johnson Our Credo. "Fair pay for all, at rates which are reasonable within the local economy" Tesco -- Towards a better world. "KEOs should make efforts to ... improve their workers' wages ..." Korean Employers' Federation -- Declaration of principles concerning human resource management for Korean enterprises operating overseas (KEOs).

76. "We believe that wage and benefit levels should address the basic needs of workers." Starbucks -- Statement of beliefs (Framework for a code of conduct).

77. Wages should "address basic needs of workers and their families so far as possible and appropriate in light of national practices and conditions" Dayton Hudson Corporation -- Standards of vendor engagement. However, the "water companies acknowledge the right of their workers to receive a just and fair wage which will enable workers to provide a decent home and living standard for themselves and their families and which will enable them to contribute to the economic well-being of their communities" Public Service International -- PSI Water Code.

78. Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy, para. 34.

79. Such references would also be consistent with Convention No. 95 (Protection of Wages Convention). Similarly, code references directing respect for national laws on minimum wage could be consonant with Convention No. 131 (Minimum Wage Fixing) and related instruments in ratifying countries to the extent their procedural safeguards for the fixing of minimum wages were respected.

80. For example, "good and safe conditions of work" Shell Oil -- Statement of general business principles, or "maintain a safe and healthy workplace" UNOCAL -- Statement of principles: Code of conduct for doing business internationally.

81. "In one code, contractors are required to provide adequate and well identified emergency exits and training for employees for emergency evacuation, healthful and safe facilities and conditions, well-ventilated and well-lit facilities, adequate medical assistance in case of emergencies, education and training for designated employees in first aid and health." See International Council of Toy Industries -- Code of business practices.

82. Chemical company members of Responsible Care, for example, tend to define operational aspects of health and safety policy in codes or in separate health and safety standards accompanying general company codes. For example, "[s]afety ... requires an ongoing commitment to continuous improvement of risk-minimizing measures" Degussa -- Guiding principles for safety and environmental responsibility.

83. "Each business shall support this policy by maintaining compliance with applicable governmental laws and regulations, as well as the company policies and requirements which are set out in the environmental, health and safety policy, codes of practice and health and safety guidance." Bristol-Myers Squibb Company -- Standards of business conduct.

84. "The company's policy is to comply with all applicable health, safety and environmental laws and regulations of the countries in which we operate. Where there are none or those that do exist are inadequate, the company will abide by its own standards." Courtauld's -- United States Code of conduct.

85. The codes primarily referred to Convention No. 155, although one IUF-NGO sponsored code mentioned Convention No. 170 and Convention No. 110, International code of conduct for the production of cut flowers, supra.

86. See, e.g., G. van Liemt, op. cit.

87. United States Department of Labor, Codes of conduct, op. cit. Knowledge among host governments of codes applying to suppliers operating in their countries was found to be mixed.

88. For the historical development of practices in auditing the social impact of enterprise operations, see S. Zadek, P. Pruzan and R. Evans: Building corporate accountability: Emerging practices in social and ethical accounting, auditing and reporting (Earthscan, 1997).

89. For the purposes of this paper, the term "social accounting" refers to internal monitoring, evaluating and reporting of enterprise performance against specific benchmarks; the term "social auditing" refers to external verification of the accuracy or completeness of the internal process; and the term "social inspection" refers to external independent monitoring, evaluating and reporting enterprise performance. "Accreditation" refers to the process of determining the qualification of inspectors or monitors, and "certification" refers to determining compliance of an enterprise with specific standards.

90. For example, International Council of Toy Industries, which added three appendices to its 1996 Code of business practices when it revised its system in 1997, including methodology for evaluating compliance, a detailed audit checklist, and a corrective action plan; Council for Economic Priorities, Guidance document for SA8000.

91. For example, Canadian Government-issued Twelve Principles for the effective implementation of codes discussed in C. Forcese, No. I, op. cit., note 35: Business for social responsibility, eight principles of quality of the Institute for Social and Ethical Accountability.

92. See the Framework for assessing social performance in industry, in Responding to global standards, op. cit., table 2.

93. In the implementation of its "Vendor and subcontractor code", the British Toy and Hobby Association has taken a position against carrying out on-site inspections in other countries which it views as essentially the task of government.

94. M. Lefebvre and J. Singh: "The content and focus of Canadian corporate codes of ethics", in Journal of Business Ethics (1992), Vol. 11, p. 807 (almost 70 per cent of Canadian codes in study exhibited this tendency).

95. For example, chemical industry codes of practice and Responsible Care, Coatings Care and Responsible Distribution.

96. See, e.g., Successful health and safety management (ILO Health and Safety Series booklet, 1997).

97. United States Department of Labor: Codes of conduct, op. cit. The study found that, in the apparel industry, retailers may concentrate internal monitoring efforts on facilities that produce private label merchandise or brands sold exclusively at their stores where more public image was invested.

98. For example, PSI International Water Industry Council under the PSI Water Code; Australian Homeworkers Code Councils under the Homeworkers Code of Practice which operates in the domestic TCF sector in Australia.

99. See, e.g., Sullivan Principles: "Statement of principles questionnaire", and International Council of Toy Industries Appendices I and II to Code (Audit checklist and methodology for evaluating compliance). One study reported that, among companies reviewed, only 27 per cent had a standard list of questions for auditing purposes and only 24 per cent conducted training programmes to familiarize their workers with the auditing guidelines. CEP SCREEN, op. cit.

100. Elkington: Profits and principles (Shell Oil, 1998), head of Sustainability NGO joining with Shell in applied research project.

101. The labels show symbols such as logos, trade marks and, in some cases, text, which seek to differentiate the product or enterprise. If the label has no text, the meaning of the label usually must be acquired elsewhere.

102. Social conditions refer to the impact of the processes of production or service on people involved, including workers, local communities, suppliers or subcontractors.

103. Where relevant, this section refers to the experience of union labels (which have existed in various forms since early in this century), labels with claims relating to animal testing, and environmental labels, which have evolved in the last 20 years. For more information on environmental labelling, see A. Appleton: Environmental labelling programmes: International trade law implications (Kluwer, 1998) and E. Staffin, "Trade barrier or trade boon? A critical evaluation of environmental labelling and its role in the "greening" of world trade", in Columbia Journal of Environmental Law, Vol. 21, p. 205 (1996).

104. Data on many relevant details were unavailable for all programmes reviewed such as copyright protection of the label at national or international levels, and penalties for non-compliance.

105. See discussion of other individual company labels, in United States Department of Labor: By the sweat and toil of children, op. cit., pp. 109-114 (e.g., Dunkin Donuts, K-Mart, Spalding Sports Worldwide, American Challenge, American Soccer Company).

106. An enterprise with a "brand label" is not likely to see the utility of an independent social label shared by other enterprises. "Our best social label is our brand label," reported Alan Christie of Levi-Strauss to the EU-US Symposium on Labour Standards in Brussels in February 1998. Where brands are trusted, use of a label may undermine consumer confidence or require changes in established business practices. However, independent and brand labels may appear on the same product, as with the fair trade labelled products sold under the brand names Cafédirect and Max Havelaar.

107. In both cases, the codes of conduct which apply to licensees that manufacture products bearing University trademarks include measures for the rehabilitation of child labourers, independent monitoring of factories and the publication of results of factory monitoring, as well as the option to terminate contracts where such conditions are not met.

108. Notable exceptions include Abrinq Foundation programmes for domestic market sectors in Brazil, and the Australian Homeworkers Code of Practice in that country's mixed textile sector, and union labels in the United States.

109. S. Zadek et al., Social labels: Civil action through the market (New Economics Foundation for the European Commission, forthcoming, 1998).

110. Notably, STEP, a joint initiative in social labelling established by an industry group with five NGOs in Switzerland, covers nearly all fundamental labour issues as well as wage levels and occupational health and safety. See Appendix.

111. Trade unions associated with tea and banana plantations or in the distribution, packaging or retail sides see fair trade labelling as a way to strengthen worker position in the agricultural sectors.

112. Available studies include the ILO study reviewed by the WP/SDL at its last session, and several others based on case-study, survey, and/or field research. See J. Hilowitz: Labelling child labour products: A preliminary study (ILO, Geneva, 1997); D. Haas: Mit Sozialklauseln gegen Kinderarbeit? Das Beispiel der indischen Teppichproduktion (Berliner Studien zur Internationalen Politik; 4; 1998); S. Zadek et al.: Social labels: Civil action through the market (New Economics Foundation for the European Commission, forthcoming, 1998) (literature review, survey questionnaire, dialogue-based research); United States Department of Labor: By the sweat and toil of children, Vol. IV: Consumer labels and child labor (1997) (non-random survey research, site visits to eight countries, public hearings, and hundreds of contacts); Open trading (New Economics Foundation and Catholic Institute for International Relations (1997); M.A. Dickson: Socially responsible consumer behavior in the apparel marketing system: Preliminary findings from a survey (Ohio State University, Dept. of Consumer and Textile Sciences, 1996); Garment workers study (Marymount University, Center for Ethical Concerns, 1996).

113. In the IGO sphere alone, research and programmes on the effects of ecolabels on trade, and guidelines on ecolabelling systems, have been addressed by Codex Alimentarius Commission (labelling organically produced foods), International Trade Centre (ecolabelling implications for developing countries' production of textiles and clothing), the OECD, UNCTAD and UNEP (UNEP Expert Group Meetings on Ecolabelling and Environmental Standards with call for international disciplines to govern ecolabelling to ensure no unnecessary obstacles to international trade, UNIDO (development of technical cooperation programmes to provide assistance to developing countries in standardization and necessary adjustments to meet internationally agreed requirements; surveys of ecolabelling effect on market access of developing countries, and views on solutions possible through harmonization of standards). See World Trade Organization, Committee on Trade and Environment: Ecolabelling: Overview of current work in various international fora, Note by the secretariat, doc. WT/CTE/W/45 (1997).

114. Appleton: Environmental labelling programmes, supra footnote 103.

115. To date, empirical data are insufficient to determine the effects of labelling programmes. The effects discussed in this paragraph are drawn from largely anecdotal evidence discussed in the studies cited supra footnote 112. The International Labour Office is preparing an empirical study of the effects of child labour labelling programmes on child workers. See infra section VI.

116. Zadek et al., supra, figure 6, at 31.

117. A. Mattoo and H. Singh: "Ecolabelling: Policy considerations", in KYKLOS, Vol. 47 (1994).

118. A. Appleton, supra footnote 103.

119. ISO Technical Committee 207 on Environmental Management classifies three categories of ecolabels depending on programme participants and design. The Organization is developing five standards: ISO 14020-14024 on basic principles (ISO 14020); self-declaration claims (14021-14023 covering terms and definitions, labelling symbols, and testing and verification methodologies); and 14024 on principles and procedures for certain third-party labelling systems (Type 1). ISO 14040-14043 address principles and guidelines for assessing environmental impact throughout the whole of a product's life from raw materials acquisition to production, use and disposal.

120. As the executive of a leading enterprise enaged in social labelling remarked in a research interview for this paper, "Most of these [child labour-free] claims are disingenious ... I hope somebody, whether it's the ILO or somebody, holds people's feet to the fire ..."

121. Whether voluntary social labelling schemes constitute "standards" covered by the Code of Good Practice is beyond the scope of this paper. The Code of Good Practice has not been determined to apply to product labelling that references processes and production methods which do not have an impact on the final product ("unincorporated PPMs"). See generally the discussion of the Code of Good Practice supra section II.2.

122. See, for example, Annex I to the TBT Agreement, article 4 (international body or system is considered to be a "body whose membership is open to the relevant bodies of at least all members") and ISO/IEC, Directory of International Standardizing Bodies, Seventh Ed., 1995 (listing ILO and the specified standards).

ISO/IEC Directory of International Standardizing Bodies (listing the ILO and the specified standards). See also Restricted trade effects of standards, technical regulations and conformity assessment procedures, G/TBT/W/42, 28 April 1997, sec. iv.

123. "Community investing", in which investment programmes select community- based financial institutions to support community development initiatives, is beyond the scope of this paper. See Social Investment Forum: 1997 Report on Responsible Investing Trends in the United States, 1997, s. I. <>.

124. Shareholder resolutions are formal, mostly non-binding, requests or recommendations to management which, under the regulations governing shareholder rights, are placed on the proxy ballot and voted by all shareholders, along with other business matters submitted by management

125. V.A. Zondorak: "A new face in corporate environmental responsibility: The Valdez Principles", in Environmental Affairs, Vol. 18:423, pp. 457-500, footnote 109 at p. 477.

126. SRI has long historical roots in the United States, dating back to the late 1700s when religious investors decided not to invest in companies engaged in alcohol, gambling and tobacco. In the 1970s and 1980s, SRI re-emerged in the United States as a means to put economic pressure on the South African Government to end apartheid.

127. See N. Smith: Ethical investment: A four-stage model and some empirical research, A final year project for BUS3800 at Middlesex University, 1996, section "The history of ethical investment". See also ADME: "Investissement socialement responsable". <>

128. See J. Conrads: Geldanlage mit sozialer Verantwortung (Wiesbaden, Gabler Verlag, 1994) <>. See also the webpage "Global Social and Ethical Investing, Consuming and Business" on the Good Money, Inc. website. < dirfrgn_index.htm>

129. In a study just released by the Social Investment Forum, a non-profit membership organization, more than $1 trillion is estimated to be under SRI management of some kind, of which $529 billion are invested in socially screened portfolios, including mutual funds. This represents a growth of 227 per cent of the socially screened portfolios during the period from 1995 to 1997 that largely outpaced the broad market. $96 billion of it is in screened mutual funds whose number has nearly tripled from 55 in 1995 to 144 in 1997. (See Social Investment Forum: 1997 Report on responsible investing trends in the United States, 1997, s. I. < InvSRItrends.htm>

130. See, e.g., Cowan v. Scargill (1985), I Ch. 270 (Cowan) at p. 287 (United Kingdom); M. O'Brien Hylton: "'Socially responsible' investing: Doing good versus doing well in an inefficient market", p. 42, in American University Law Review (1992), pp. 2, 37-45; A. Leigh: "'Caveat investor': The ethical investment of superannuation in Australia", p. 25, in Australian Business Law Review (Oct. 1997), 341 (Australia).

131. See M. O'Brien Hylton, op. cit.. See also N. Dunnan: "Doing well by doing good: Investment funds satisfy goals and social concerns", in ABA Journal (Aug. 1996), p. 96; and J.K. Glassman: "Ethical stocks don't have to be downers", in The Washington Post (23 Apr. 1995), p. H01.

132. Other selection criteria range from tobacco, alcohol and gambling to environment and animal welfare. See Social Investment Forum, op. cit., s. II; and Co-op America: Socially responsible mutual fund screens. <>

133. These include in the United States: Kinder, Lydenberg, Domini (KLD), best known for its Domini 400 Social Index (DSI), a socially screened, capitalization-weighted index of 400 common stocks modelled after the S & P 500 stock index; and the Social Investment Forum; in the United Kingdom: the Ethical Investment Research and Information Service (EIRIS) and the United Kingdom Social Investment Forum.

134. Exceptionally, some large socially screened funds also engage in shareholder activism, opening dialogue with companies in which they invest on labour- related and other issues and sponsoring shareholder proposals on these issues while retaining the divestment option as a last resort, e.g. the Calvert Group, Ltd. and Franklin Research & Development Corporation. See for the former Calvert Group Ltd.: Understanding the shareholder resolution and proxy voting process. <>

135. Number of shareholder-sponsored resolutions in -- United States: more than 650; Japan: 54; Germany: 39; United Kingdom: 4 (13 in 1995); Sweden: 5; Canada: 3; Denmark: 2; France, Switzerland, Ireland, Norway: 1, respectively. (See IRRC: "Shareholder action advances worldwide: Investors placed a broader range of topics on more non-US ballots in the 1996 global proxy voting season", press release, 29 July 1996, < shareholder.html> and "IRRC finds shareholder activism at record levels around the world", IRRC 1996 press release. <>

136. C. Forcese: Putting conscience into commerce: Strategies for making human rights business as usual (Quebec, International Centre for Human Rights and Democratic Development, 1997), p. 65.

137. See J. Tagliabue: "Compliments of US investors; New activism shakes Europe's markets", in New York Times, 25 Apr. 1998; J.-A. Schneider: "Le droit de l'actionnaire de proposer une résolution au vote de l'Assemblée générale", to be published in Aktuelle juristische Praxis AJP/Pratique juridique actuelle PJA, s. A, para. 4 and s. D, para. 16.

138. See J.C. Wilcox: "Can 'relational investing' really work?", <> reprinted from The New York Law Journal, 1993, and John Tagliabue, loc. cit.

139. Those two sectors both respectively account for one-fifth of the targeted enterprises. The first includes companies such as Texaco, Unocal, Chevron, Mobil, Exxon and ARCO, the second NIKE, Disney, Philips-Van Heusen, Mattel, etc. (see ibid.).

140. For example BASF, Bayer, Hoechst, Merck and Schering (see Dachverband der Kritischen Aktionärinnen und Aktionäre: Kritische Aktionärinnen und Aktionäre bei ….<>

141. See M.A. O'Connor, loc. cit., p. 1347. Exceptionally, the Union of Needletrades, Industrial and Textile Employees (UNITE) has simultaneously waged publicity and shareholder campaigns, working with labour rights activists, involving companies such as Gap, Disney, NIKE, Wal-Mart, Philips-Van Heusen and Guess. (See P. Varley (ed.): The sweatshop quandary: Corporate responsibility on the global frontier (Washington, DC, IRRC, 1998), p. 18.)

142. Resolutions filed with Mobil (1998) and Chevron (1997 and 1998) requesting guidelines for country selection and reports on the company's business in Nigeria, in ICCR: The proxy resolution book (1997), p. 75 (1998), pp. 74, 75.

143. In the United States, among the more than 650 shareholder resolutions the Investor Responsibility Research Centre (IRRC) has recorded in 1996, only 50 were linked to international labour issues. (See P. Varley (ed.): The sweatshop quandary: Corporate responsibility on the global frontier (Washington, DC, IRRC, 1998), see list of resolutions at pp. 27-29.)

144. Agreements to withdraw proposed resolutions occur in about one-third of the resolutions in the United States. In other situations, management finds legal grounds for omitting the proposal from the proxy ballot, or the proposal is voted. Partly because of the nature of shareholder proxy voting, resolutions never come close to achieving a majority vote. Proposals dealing with social matters rarely receive more than 15 per cent of shares voted and frequently less than 10 per cent. (See P. Varley (ed.), op. cit., p. 18. Exceptionally, in 1996, a resolution filed with the retailer and private label manufacturer J.C. Penney requesting a report on labour standards for overseas suppliers obtained the support of the management, and then received 87 per cent of the votes (see ibid., table at p. 28).)

145. For example, programmes on business and society and Building competitiveness and communities, by Jane Nelson, Prince of Wales Business Leaders Forum (Geneva, 24 Sep. 1998) (INSTITUTE); and the series of workshops reflected in various publications such as Manual de Balance Social (ILO and CONFIEP, Peru, 1997).

146. The Joint Action Programme brings business representatives together with UN agencies and entities at the UN Staff College at the ILO Turin Centre and has led to practical collaborative action programmes among ENTREPRISE, ACT/EMP, and other parts of the UN system.

147. Included in the industry/government partnerships are various significant initiatives of SECTOR. These include a series of national codes of practice being developed and implemented on a tripartite basis to address health and safety in the forestry sector. Such projects address the forestry sector in Brazil, Chile, Fiji, Finland, Latvia, Uruguay, Zimbabwe and a SECTOR Working Paper on the subject, by P. Blombäck, Codes of forest practices: A guide to formulation, implementation and monitoring, is in preparation. In the oil and gas industry sector, a 1993 Tripartite Meeting pertaining to offshore petroleum installations adopted conclusions promoting a self-regulatory approach to safety management. See Final report, Tripartite Meeting on Safety and Related Issues Pertaining to Work on Offshore Petroleum Installations, Geneva, 20-28 April 1993, paras. 5, 6, 12, in TMOPI/1993/9 and GB.273/STM/4/1. ILO cooperation with oil industry employers' and workers' organizations to promote those conclusions has included participation in international conferences on health, safety, and environment in oil and gas exploration and production. In the textiles, clothing and footwear (TCF) sector, workshops to discuss national codes of conduct (Indonesia, Philippines, Sri Lanka and Thailand) have followed up on conclusions of a 1996 Tripartite Meeting. See Note on the proceedings, Tripartite Meeting on the Globalization of the Footwear, Textiles and Clothing Industries: Effects on Employment and Working Conditions, Geneva, 28 October-1 November 1996, para. 14, in TMFTC/1996/11.

148. TRAVAIL and its International Programme on the Elimination of Child Labour (IPEC) oversees the provision of specialized training for governments in developing child labour regulations and conducting child labour inspection (e.g., projects in Indonesia, Pakistan, Philippines, Thailand, Turkey). The department is also coordinating the Tripartite Meeting of Experts on Labour Inspection and Child Labour to be held in 1999, which will draw attention to the critical role of labour inspectors in combating child labour, identify best practices and approaches and facilitate the sharing of experience in this area. Also within TRAVAIL are the tripartite codes of practice in occupational health and safety, and guides and manuals on methods of labour inspection of health and safety standards. See supra section II.2.

149. Recommendation concerning general conditions to stimulate job creation in small and medium-sized enterprises, adopted by the International Labour Conference, 86th Session, Geneva, 17 June 1998.

150. Procedure for the examination of disputes concerning the application of the Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy by means of interpretation of its provisions, Official Bulletin (Geneva, ILO), 1986, Vol. LXIX, Series A, No. 3, pp. 196-197, replacing original version adopted in 1980 by the Governing Body.

151. The SA8000 standard, for example, mentioned in sections II.3 and III above, refers specifically to ILO standards, but not without altering their scope (for example, freedom of association: see above, section III) while seeming to refer to an ISO system, which is probably not accidental.

152. The UNIDO report mentioned above (section II.2) states: "For export-oriented firms, the dictates of large international buyers reflected in their own codes of ethics are more instrumental in bringing about improvements in social and environmental performance than guidelines recommended by international agencies such as the ILO, World Bank and WHO, etc. ... as they are better able to exploit trade opportunities inherent in the demand for sustainably produced goods." Responding to global standards, see footnote 16 above.

153. See Programme of sectoral meetings, 2000-01, GB.273/STM/1.

154. It will be recalled that this was the sense of the solution discussed in the Report of the Director-General to the Conference in 1997: this was intended to enable member States to establish and mutually recognize a labelling system based on a Convention so as to mobilize public support, in a coherent perspective of strict reciprocity, for the application of Conventions.

Updated by VC. Approved by RH. Last update: 26 January 2000.