INTRODUCTION
Corporate codes of conduct are gaining increasing prominence in the debates over fair labour standards, particularly in relation to employment by multinational enterprises (MNEs).
There two kinds of code of conduct, and two kinds of labour standards. Actual labour standards - what people are paid, how jobs are designed, the levels of safety and structures of supervision and so on - are created every day in every workplace in the world through the interaction between workers, employers' representatives and the labour process itself. Whether or not the firm has adopted a formal code of conduct, its representatives display a de facto 'code' in their dealings with workers. So in this first sense, all firms have a code of conduct and all workplaces generate labour standards.
In the second sense, codes of conduct are formal policies which purport to shape corporate conduct in certain ways. Labour standards, in their second meaning, are the fundamental conditions which should apply across workplaces - for example the labour standards of the International Labour Organization (ILO). This paper is concerned with the ways in which formal codes of conduct impact on workplace behaviour, and their value in enabling workers to secure conditions of work no less favourable than those provided for in internationally sanction labour standards.
The action which shape labour standards within a firm are influenced, to a greater or lesser extent, by the external environment, which includes the systems of national and international regulation. Where legally binding rules (or some other strongly normative processes) do not exist in the external environment of the firm, and where management is strongly placed to dictate actual standards within the firm, the mechanism of the corporate code of conduct becomes a significant, perhaps the only, instrument for attaining fair labour standards. It will be shown that the phenomenon of 'globalization', discussed in Section 2 below and by Robert Kyloh elsewhere in this volume, has challenged national and international regulatory schemes and led to just the combination of internal and external factors which accords a greater significance to the code of conduct.
Codes of conduct are thus paradoxical and controversial. They are a creature of the firm, yet are used to temper the power of the firm in relation to its dealings with its employees. It will be shown that such self-limitation on employer power generally occurs when the interest calculation of the firm shows an advantage in doing so. Codes of conduct are controversial because they may be used to alter the balance of power within a firm towards the employer (for example by replacing collective bargaining with the operation of the code), and because there is likely to be contestation between the firm and interest groups as to the way in which the employer's power should be expressed in the code. This element of contestation can be heightened where the 'firm' is a multinational enterprise spanning very different cultural, social and political systems. Alongside the complex structures of international business, equally complex networks of interest groups are developing at international and regional levels. Even taking a simple example (say, consumer pressure on a retailer in Denmark to show that goods imported from Zambia were produced under humane conditions) we can see the depth and complexity of the issues raised by MNE codes of conduct: how wide does the net of responsibility for labour standards extend, should national sovereignty be subverted by international standards, what rights do workers have to self-determination over such regulatory instruments, what is the validity of international labour standards, to what extent should a firm be driven by economic imperatives only, what institutional framework should exist to manage such issues, and so on.
This article seeks to place codes of conduct within the current political and economic developments referred to as globalization. The history of codes is discussed in Section 1, and the ways in which current institutional structures and rules are inadequate to meet the challenges of industrial development are then considered in Section 2. The constituent elements of codes of conduct, and what is required to make them work are discussed in Section 3. Two cases studies then follow, which attempt to show how individual codes of conduct arose from their particular historical circumstances, and the ramifications of these processes for the general debate.
In summary, the paper concludes that a number of elements should be present before a code of conduct can operate equitably and effectively. These are that:
- the content of the code, and the processes by which it is determined and implemented, involves and empowers the workers covered by it;
- the code reflects the local needs of workers and as an absolute minimum guarantees the core standards of the International Labour Organization (ILO);
- the company which adopts the code is genuinely committed to its implementation, and provides the resources, training, monitoring and reporting mechanisms to make it work;
- the company's behaviour towards its employees (and those whose employment is directly dependent upon the company through subcontracting and other arrangements) is transparent, and adherence to the code is subject to independent verification by qualified assessors.
By extension, codes of conduct are unlikely to perform a useful function, and indeed may be harmful and counterproductive, when
- the code unilaterally imposes outcomes on the workforce, and stifles the processes of worker participation and negotiation at the workplace;
- the code is vaguely worded and/or ignores international standards;
- the code is just a piece of paper issued by head office, with no real impact on actual company operations, policies or people;
- the company's labour practices are closed to external audit.
Corporate codes of conduct are thus political instruments created through the interaction of usually competing forces in the workplace, community and regulatory networks. Such codes are by definition not able to take the place of universal systems of regulation: even an ideal code of conduct is limited in application to employees of a particular company, or at best to those employed by its suppliers or subcontractors. Given the very real limits on an instrument which is essentially self-policing, and the contemporary trends in industrial development what is required is an enforceable universal safety net of minimum labour standards to protect all workers.
1. HISTORY OF CODES OF CONDUCT
Codes of conduct which shape employer behaviour towards employees are not a recent phenomenon - employers have been guided by moral, religious and political 'codes' since the start of industrial production. For example, some of those who first campaigned for fair labour standards, the "utopian visionaries and sentimental dreamers", also prided themselves on the conditions enjoyed by their workers. The employer as paternalist is still a matter for popular mythologising today. Rather, taking a long view, it is probably better to regard the late twentieth century as uniquely a time when in some Western countries it is not clear what external belief systems inform people's ethical conceptions of employment relations.
The modern history of codes of conduct is usually said to start in the mid-twentieth century. Three phases of development can be identified: the period around and immediately after the Second World War, the 1970s, and the current explosion of interest in codes in the period from the late 1980s to the present.
The first codes were those created by the International Chamber of Commerce (ICC): it created a Code of Standards of Advertising Practice in 1937, followed by codes on marketing and direct marketing (finally consolidated in the 1974 International Code of Marketing Practice). The end of World War II brought a new set of international institutional arrangements, and an upsurge in the activities of MNEs, especially American MNEs in Europe. In the face of pressure to control this influx, and probably to cement its own role on the international stage, the ICC developed an International Code of Fair Treatment for Foreign Investment, which was largely directed at influencing the ways in which governments dealt with MNEs.
The next wave of development occurred in the late 1960s and 1970s. Kees van der Pijl pinpoints in the 1970s a "moment of regulation". In this period, there was a conjunction of opinion between developing nations and organised labour that the power of MNEs should be restrained by binding rules. The ITT scandal in Chile was a defining incident. The first call for international regulation of MNEs was made by the International Confederation of Free Trade Unions (ICFTU) in the late 1960s.
This mood for change was symbolised by the 1974 UN resolution advocating a New International Economic Order, and the Report of the Group of Eminent Persons convened by the UN's Economic and Social Council to report on the regulation of MNEs. The Eminent Persons' Report is a bold expression of faith in meaningful international action to regulate the power of MNEs, including the "concertation" of national labour laws to ensure global protection of workers. The Report gave rise to negotiations over the UN Draft Code on MNEs. In 1976 the OECD adopted its Declaration on International Investment and Multinational Enterprises, and in 1977 the ILO Tripartite Declaration on MNEs was agreed. At the regional level the Andean Investment Code was created, and in the European Community various proposals for worker participation in corporate governance and information and consultation first emerged. At the level of international industries, codes were also established. For example, the 1975 International Council of Infant Formula Industries (followed by the WHO/UNICEF code of 1981 on marketing of breastmilk substitutes, and the 1984 Nestle/NGO agreement about the implementation of the WHO code). The International Federation of Pharmaceutical Manufacturers Association produced a code on marketing, and the ICC first code on bribery was established in this period.
At the national level at this time, developing countries were attempting to assert control of MNEs and nationalisation of industries occurred in some countries. Sikkink reports that 22 developing countries passed legislation controlling MNEs in the period 1967 - 1980. Developed countries also established regimes: for example, the Canadian Government's Some Guiding Principles of Good Corporate Behaviour for Subsidiaries in Canada of Foreign Companies (1967), the Code of Behaviour for Japanese Investors Overseas (1973), and the US Foreign Corrupt Practices Act (1977). The Canadian Guidelines were later adopted by the US and Canadian Chambers of Commerce as the Precepts for Successful Business Operations Procedures in Canada and the US. In 1975 USA-BIAC issued its Review of Standards and Guidelines for International Business Conduct. Like the Business Roundtable and the US Chamber of Commerce, it concluded that a general cross-industry code would not be effective, rather individual companies should "set for themselves standards of responsible conduct", reflecting the strong preference of business for non-centralised, voluntary regulation which persists today.
There was also increased activity at the level of individual businesses. A sample of these was reviewed by Kline in 1985 who found that most corporate codes of conduct were directed at internal company behaviour and were therefore "particularly inappropriate to the diverse cross-cultural demands of the international market place". Those codes which were externally directed were "aimed primarily at social critics, the government and the media", and lacked "operations specificity or credibility". "Framework" codes directed to specific issues were instituted during this period, including in 1977 the Sullivan Principles, directed at the behaviour of US companies operating in South Africa. In 1984 the MacBride Principles were created with the aim of influencing the behaviour of US firms in Northern Ireland.
This period of intense code activity petered out during the 1980s. Both Sikkink writing in 1986 and Kline writing in 1985 describe a "stagnation" of code efforts and of research into them. This decline coincided with the shift towards deregulatory market-driven policies in both developed and developing countries, lifting pressure on business to account for its actions and abide by externally approved norms.
There has been a resurgence in the pressure for and use of codes of conduct in the late 1980s and 1990s. A number of factors seem to have brought this about. First, political pressure from organised lobby groups, political parties and individuals has created a new imperative for accountability. This has been seen not just in relation to labour standards but also in relation to environmental issues.
The phenomena of socially-responsible investment and shareholder action on social issues have become more widespread. In May 1997 shareholders of Shell were lobbied to agree to a motion which would require the company to submit its environmental and human rights practices in Nigeria to external audit. The Times commented:
Shell is not the evil corporate monster that the protesters would have us believe, but it must rank as a public relations disaster area...The fact that 11% of Shell's shareholders were persuaded to vote against the company's board yesterday was a huge blow to the company and carries a strong message to industry generally. The demand for external auditing will grow, as a younger generation fired up over environmental issues and convinced that big business is exploitative overseas, becomes more vociferous.
Growing economic integration within regions has highlighted the international disparity in wages and labour costs, and the availability of efficient electronic telecommunications has narrowed the distance between the so-called First and Third Worlds; these factors have underpinned the growing internationalisation and education of community and other interest groups, including trade unions, consumer groups, churches and other social justice campaigners. Economic uncertainty and high unemployment in developed countries have led to calls for protection at a time when barriers to trade are coming down. All these developments have focused public attention on the sites of global production, with the resulting scandals and high profile campaigns. Symptomatic of this new mood, the Clinton Administration created its Model Business Principles of May 1995, which call for "fair employment practices, including avoidance of child and forced labour and avoidance of discrimination based on race, gender, national origin or religious beliefs; and respect for the right of association and the right to organise and bargain effectively".
Another likely cause of this code activity is the concerted push to achieve a global system of regulation of labour through the World Trade Organization. This seems to have led to heightened awareness of labour standards, and the need for firms operating transnationally to account for their labour practices if for no other reason than to assist in the campaign of international business to avoid this regulation. At its Singapore meeting, the Trade Ministers of WTO member states renewed their support for "the observance of internationally recognised core labour standards".
Finally, there has been a growing recognition that labour standards are a matter of financial interest to companies. It has been shown that labour standards are an input to economic growth at both the macro and micro levels. And there is emerging evidence that firms which adhere to labour standards are more efficient and better supported by investors. A recent survey of 100 German firms found that those which adopted policies of enhancing worker skills, security, flexibility and initiative gave a greater return to shareholders than their competitors in the same industry:
This has already begun to happen in the United States. The successful California state pension fund, Calpers, which manages more than $(US)100 billion, takes "workplace practices" into consideration and prefers to invest in companies which demonstrate a strong employee focus.
A number of high profile firms have adopted codes in relation to their own employees and their key production suppliers. But other methods are also gaining prominence: industry codes covering multi-employer groups, retailers seeking to ensure that the goods they sell are made without exploiting labour, framework codes such the model code of the International Confederation of Free Trade Unions (ICFTU), collectively bargained agreements between firms and trade unions in relation to supply networks, bi-partite contracts establishing company-wide standards, composite codes which combine model codes of conduct with implementation strategies, licensing schemes for producers, public education campaigns about fair labour standards,and alternative trade networks.
2. GLOBALIZATION AND THE CHALLENGE TO REGULATION
Questions about the internal and external regulation of corporations have arisen sharply in the context of global economic integration and the very significant role now played by MNEs. Whereas in the 1970s, concern over MNEs focused on potential interference with national policies (and we will see that this remains a central concern of the MNE international "regulatory" system), today there is a more fundamental worry: that economic activity is being conducted on a truly transnational plane which by-passes the 'geographical logic' of the nation state altogether.
Some of the important trends in this big picture:
The "gigantism" of MNEs. Some 37,000 MNEs control about one-third of the world's private sector productive assets, and have world-wide sales worth $4.8 trillion.
There is a marked asymmetry of foreign direct investment (FDI) flows: 97% originates in developed countries which are also the recipients of 75% of it. China is now the largest host nation in the world, accounting for nearly three-quarters of the FDI flows to less developed countries. Between 1986 and 1991, the share of the least developed countries in FDI flows reduced from 2.9% of the world total to 0.6%.Flows to sub-Saharan Africa were on a par with those to New Zealand.
Systems of global production have replaced trade as the major means of international economic interaction. This involves the "intra-firm international division of production", and with it an international secondary labour market which does not share in the diffusion of technical skills and wealth generated within the global networks.. This new international division of labour thus leads to widening gaps in income and opportunity. For example, a common pattern for Japanese MNEs in East Asia is to conduct sophisticated research and development functions in Japan, medium-end production in Malaysia, and low-end production in the Philippines. There is also an increase in 'arms-length' production through subcontracting, licensing and franchising, adding to the complexity of organizational authority and responsibility.
"Exponential growth of cross-border alliances or collaborative agreements" has created international networks between the MNEs, blurring even further the geographical foundation for economic activity. Bellack refers to the "substitution of external markets by internal markets of operation (such as transfer pricing, exchange of technology, take-over, merger, contractual agreements) creating a sphere of action which is almost inaccessible to governments."
In the critical area of technological developments, MNEs are now "the main actor in determining national level technological development." For example, the Finnish MNE Nokia has a research and development budget two-thirds the size of the research funding of the entire Finnish university sector.
The highly competitive international environment, together with the diffusion of production over different sites, makes MNEs 'footloose': ready, willing and able to relocate to utilise the comparative advantage of different sites, including low labour costs.
Programmes of privatisation of state functions have allowed MNEs to penetrate into areas of economic activity usually regarded as close to the national interest and identity of individual countries. For example, it was recently reported in the UK that an American MNE was bidding for the right to operate part of this country's social security programme.
MNEs can thus shape economic and social outcomes to an extent previously unknown. They have enormous power because of their mobility: they can demand, and get, huge amounts of government support to attract, or secure, their presence. For example, "...when Ford recently threatened to relocate the production of Jaguar cars from the UK to the US last year, it was granted £80 million in government subsidies... to entice them to stay." MNEs can bid down employment conditions between potential locations, or simply move to areas with cheaper labour costs. Bellack discusses the "blackmailing" of workers by Hoover when it announced that one of its three national plants was to close, without specifying which one. And Pollert reports that during a dispute at the Volkswagen-Skoda factory in the Czech Republic a one-hour stoppage occurred:
Moderate though the one hour strike was, management aired the usual threats of mobile global capital, arguing that Mexican wage costs were the same as Czech and production could move there... skilled Czech workers still get only one-tenth of the average German VW wage.
Many case studies of transnational involvement in manufacturing show that such threats are often followed by action to shift at least some elements of production to lower-cost sites. For example, Kuruvilla shows how Mattel's toy production moved from Hong Kong to Malaysia when labour costs rose in Hong Kong. When costs rose in Malaysia, 30% of production was contracted to India.
This 'jurisdiction shopping' encourages states to create an attractive regulatory framework for MNEs. The creation of Export Production Zones (EPZs) or Special Economic Zones (SEZ) in which workers' rights are waived to attract foreign companies is one of the starkest examples of this regulatory competition. MNEs wield this power with only limited accountability, as will be seen below. There is a "yawning democratic deficit" in what some are calling the global "single economy and single market".
The trends towards 'arms-length' commercial relationships - such as sub-contracting, licensing and franchising - within and across international companies also challenges the current regulatory norms. Most national systems of labour law (and the international systems from which they grew) are based around the mutual obligations arising from the employer/employee relationship. But it is increasingly the case that MNEs have an important impact on the lives of many people employed by someone else in the backward- and forward-linked processes associated with the MNE operations. Thus national laws which govern the relations between a sub-contractor and his/her workers are unlikely to have jurisdiction over, say, the contractual negotiations between that sub-contractor and the MNE even though the terms of this contract may directly lead to the violation of workers' rights and entitlements. MNE employment must also be viewed in the context of the failure of many systems of national law to protect the interests of workers who are, for whatever reason, invisible to regulation. Homework's, those employed in the informal sector, and even so-called atypical workers within the formal sector are often disregarded by extant national legal processes.
But the problem is not just one of the inadequacy of national systems. International industrial processes call for an international framework of regulation which is beyond the scope of any one state to create. Take for example a company with a divisional structure in which production of a particular product or service takes place in several countries. For the purposes of forward planning, employment policy, line management, resource allocation and internal determination of working conditions the company may regard this divisional unit as one entity. For workers to meet management on a level playing field, they need access to all or some of the following: protection against discrimination for undertaking transnational trade union activities (for example, organising in more than one country), the right to strike over matters affecting the division as a whole, the right to strike in support of workers in other parts of the division or in other internationally organised industries (without such action being regarded as "secondary" industrial action), the right of entry to the other countries in which the division operates when on union business, the right to collectively bargain with divisional management or whoever has the power to determine matters within the division, and the right not to have strike action undermined by the transference of production to other parts of the business.
Clearly, the problem is not just that workers in the same industrial situation may be subject to different laws within the different national regimes: what is needed is a transnational regulatory schema which set down the positive rules for international industrial behaviour.
No such international regulatory schema exists. The Report of the Group of Eminent Persons convened by the UN in the 1970s to consider regulation of MNEs made recommendations on many of these matters: for example, it recommended that national governments "follow liberal rather than restrictive policies" in relation to international sympathy action, that MNEs be required by national laws to provide certain important information within each host country (including health and safety measures taken at home), the creation of consistent international accounting structures, that MNEs be prevented from entering countries which violated workers' rights unless they had obtained permission to apply international labour standards to their own workforce, that access to real decision-makers be guaranteed and so on.
Some writers conclude that national power is so compromised that a new set of international rules and institutions must be found. Fatouros asks "... whether it is not time to start thinking about the possibility of constructing a common legal framework for all world economic activities, one that would cover all the separate issue areas and cope with their interrelationships." In the case of international labour standards, this case has been articulated by the ICFTU and some governments. The link of both labour and environmental standards with trade has been debated within the World Trade Organization (WTO). The OECD is hosting negotiations over a Multilateral Agreement on Investment (a "GATT" for FDI), which, in the view of international trade unions, should also include labour standards and industrial relations matters.
This mismatch of regulatory scope and actual economic structures not only limits the effectiveness of state control: it also limits the capacity of other actors to deal with MNEs from a position of strength. Wedderburn's conclusion from the 1970s is relevant today: "Until some such recognition of the international realities (i.e. of the right to engage in "international" industrial action) is undertaken by each system of national law, it will remain the social function of the law to assist in the fragmentation of the international trade union movement, and by so doing to inhibit the development of industrial action across national frontiers..." The fact is that there are currently no supranational institutions which regulate labour practices of MNEs. This reflects the preference of MNEs and the countries which have supported their interests over the years. It is in the context of this vacuum of regulation that the code of conduct issue must be seen.
Of course, the big picture of globalization is an aggregation of many smaller pictures, some of which show important deviations from the conclusions reached above. (For example, the scope of MNE activity varies across industries; MNEs still tend to reflect the governance structures, culture and policies of their home nation; there is an increasing number of small and medium-sized MNEs; the "race to the bottom" has not cut a straight path through current industrial development; there is evidence of growing international networks of interest groups which are attempting to deal with MNEs at the international level; the international trade union movement is undertaking an increasingly important role in co-ordinating and advising transnational action.) In the smaller frame we can observe the actual presence of MNEs in particular locations, employing real people who live in real communities. These people, management and workers, create labour standards each day. Whether the head office has produced a document called "code of conduct" or not, workers are confronted with a set of behaviours in all their dealings with the MNE. The internal regulation of labour standards within a MNE is to be welcomed as a valuable mechanism to influence actual behaviour at the workplace and in respect to labour policy at the head office. However, it should not be mistaken for more than that. What the current international economic system calls for is a binding and consistent set of labour standards which are subject to appropriate monitoring and enforcement.
2.1 The International Framework
I will now turn to the current international framework which regulates MNEs and their employment practices, the systems of the OECD and the ILO. It is argued that these are essentially voluntary structures which are inadequate to meet the current needs of workers and business. Five key conclusions are drawn from a comparison of the international codes governing MNE behaviour:
- the international systems are not fully consistent with each other;
- the rules governing MNE behaviour are not legally binding and their application cannot be enforced;
- the systems defer to national practice, and therefore fail to create a truly international framework;
- the ILO and OECD guidelines are based on an artificial notion of the 'symmetry' of obligations between employers and employees;
- a double standard is evident when the international rules governing MNE's labour relations are compared with those concerning financial transactions: labour rights are given less protection than the firm's entitlement to "fair competition".
2.2 The OECD Declaration and How It Works
The OECD's Declaration on International Investment and Multinational Enterprises (the Declaration) was created in 1976. It takes the form of a recommendation from OECD Governments to multinational enterprises to abide by a set of Guidelines for Multinational Enterprises (the Guidelines) which are annexed to the Declaration.
The Declaration requires Member countries, subject to some aspects of national interest and international law, to treat MNEs in ways "no less favourable than that accorded in like situations to domestic enterprises" (the principle of national treatment). The Guidelines are cast in terms of the benefits to economic interests of Member countries of MNEs, and the need for Member countries to co-operate to "resolve the difficulties to which their various operations may give rise." The Guidelines state that "observance of the guidelines is voluntary and not legally enforceable". The Guidelines are intended as "good practice for all", that is both MNE and domestic industries: "accordingly, multinationals and domestic enterprises are subject to the same expectations in respect of their conduct wherever the guidelines are relevant to both." In the most recent review of the Code, the Committee stated: "Their objective is to provide guidance to MNEs by setting standards addressed to these enterprises. In particular they should help ensure that their operations are in harmony with the national policies of the countries where they operate and to strengthen the basis of mutual confidence between enterprises and governments."
The General Policies section outlines a number of requirements which are vaguely expressed (for example, "... enterprises should take fully into account established general policy objectives of the Member countries in which they operate") and are insufficiently clear to establish actual standards of behaviour (for example "enterprises should abstain from any improper involvement in local political activities"). The pattern of weakly expressed requirements is broken in the case of bribery. Here definite behaviours are called for: "Enterprises should not render - and they should not be solicited or expected to render - any bribe or other improper benefit..."
Under the Disclosure of Information section, MNEs should provide "a sufficient body of factual information on the structure, activities and policies of the enterprise as a whole", and so provide at least annually not only financial statements but also "other pertinent information" including organizational information.
The Employment and Industrial Relations Guideline sets out a framework for MNE behaviour "within the law, regulations and prevailing labour relations and employment practices, in each of the countries in which they operate". The terms of the Guidelines are set out in Appendix One.
The final Guideline relates to Science and Technology and makes vague references about the need for MNEs to "endeavour to ensure" certain outcomes in relation to the transfer of technology.
2.2.1 Follow-Up Procedure
Regular review: There is a procedure for regular reviews of the Guidelines, and over the years a number of changes have been made to the text, which are discussed further below.
The complaints procedure: The Committee on International Investment and Multinational Enterprises is the body which processes matters arising from the Guidelines. Some commentators refer to the Committee as tripartite, but in fact the labour (TUAC) and business (BIAC) lobbies only have consultative status. The Committee can deal with matters raised by the BIAC or TUAC or the Member countries. Matters raised are dealt with as hypothetical cases and the decisions of the Committee take the form of "clarification" of the Guidelines. "The Guidelines themselves are expressly stated to be non-legal and voluntary. Clarifications are accordingly general and hypothetical explanatory statements, which do not accept or reject the truth of any given factual situation and do not pass judgement on anybody's conduct." This is a severe limitation on the practical enforcement powers of the Committee.
In practice, most of the Committee's work has been as a result of matters raised by TUAC and Member countries arising from the application of the Employment and Industrial Relations Guidelines (see Appendix One). Although the output from the Committee may be a clarification extrapolated from a factual situation, commentators note that most OECD members are aware of the facts of the case, and as with the ILO, the process of public airing of complaints sometimes works to secure improved behaviour through public humiliation. The Committee itself claims for the OECD schema the following: "The Guidelines have proved to be a realistic, balanced and flexible framework... while the work of the Committee on the Guidelines has helped to demystify the multinational phenomenon and, in doing so, alleviated or dissipated earlier fears or concerns."
Despite this claim, the Committee's "clarifications" show an extremely cautious, even opaque, attitude to interpreting the Declaration, an unswerving adherence to the principle of national treatment, and the primacy given to national systems of industrial relations and law. Thus, despite finding that anti-union activism occurred in some MNEs, it does no more than "reaffirm the view already expressed in both the 1979 and 1984 Review Reports that the thrust of the Guidelines in this area is towards management adopting a positive approach towards the activities of trade unions and an open attitude towards organizational activities of workers in the framework of national rules and practices." Its response on the issue of non-trade union representation of workers is difficult to discern: "enterprises should take all practical steps towards addressing the objectives underlying paragraph 6 (of the Declaration) within the framework of national laws, regulations and prevailing labour relations practices." There has been some disappointment expressed by trade unionists at the Committee's handling of complaints, but its supporters claim that the lack of change in the Guidelines has the full support of business: "it is generally considered that the Guidelines' relative stability has been a crucial factor in their continued acceptance by enterprises."
National Contact Points: In addition, Member countries are required to establish National Contact Points to "promote the guidelines, deal with enquiries and help to solve problems which might arise between business and labour."
2.2.2 The Impact of the OECD System
Given the general nature of its principles, and the fact that the follow-up and review procedures have not significantly strengthened the guidelines through interpretation, it is not surprising that there are few concrete claims for the OECD system influencing actual behaviour. The OECD itself states that the Guidelines have become a "respected point of reference for a great majority of MNEs", and the Encyclopaedia of Labour Law reports "they are certainly accepted and followed by a great majority of enterprises" but little evidence is given to support these claims. There appear to be few references in the 1990s codes to the OECD Guidelines. The reluctance of MNEs to report on compliance with the code in their annual reports undermines the suggestion that the OECD schema has become assimilated into mainstream corporate practice. Rather, it is likely that the general support from business for the Guidelines reflects the fact that it does little to constrain them in any practical way, as argued by Grosse.
2.3 The ILO Declaration and How It Works
In 1977 the International Labour Organization (ILO) Governing Body issued a Tripartite Declaration of Principles Concerning Multinational Enterprises and Social Policy (the ILO Declaration). The Declaration relates to what it calls the "social aspects of the activities of MNEs, including employment creation in the developing countries"... It is addressed to tripartite parties of the ILO structures (i.e. the governments of Member states, worker and employer organizations) and to "the multinational enterprises operating in their territories". The Declaration includes a number of principles which "are intended to guide the governments, the employers' and workers' organizations and the multinational enterprises", compliance is "on a voluntary basis" and its provisions are not intended to affect obligations arising from ratification of ILO standards. "The aim is to encourage the positive contribution which multinational enterprises can make to economic and social progress and to minimise and resolve the difficulties to which their various operations may give rise..."
The ILO Declaration places great stress on the primacy of national sovereignty: "All parties concerned by this Declaration should respect the sovereign rights of States, obey the national laws and regulations, give due consideration to local practices and respect relevant international standards." Also mentioned are the Universal Declaration on Human Rights (and the other Covenants) and the Constitution of the ILO "according to which freedom of expression and association are essential to sustained progress".
The Declaration calls on Member States to ratify ILO Conventions 87, 98, 111 and 122, and to apply their principles anyway, along with Recommendations 111, 119 and 122.
A summary of the key elements of the ILO Declaration is in Appendix Two.
2.3.1 Follow-Up Procedures
An ILO Committee on Multinational Enterprises oversees the Declaration's implementation, and as necessary the Governing Body interprets the terms of the Declaration. The Committee receives reports from Member countries on the application of the ILO Declaration. The reaction of some developed countries to this procedure has been critical, reflecting in Baade's terms "the visceral opposition of some MNE home countries to the establishment of what has been tastelessly labelled a 'wailing wall' for complaints against the conduct of such enterprises."
In an important decision in 1988, the Committee on MNE clarified the "balance" which the Declaration seeks to strike between the interests of MNEs and member states. This decision was endorsed by the Governing Body in that year. The attempts by the Declaration to deal with both the positive and negative impacts of MNEs "are interdependent and neither can be pursued as a separate objective and disregard of the other... To take the view that the minimisation of negative social repercussions per se fulfils the overall purpose of the Declaration is therefore not correct...Such action must also be in harmony with the counterpart goals of the Declaration of contributing to economic and social progress, and be compatible with national economic and social policy imperatives." It appears from this that the ILO Declaration cannot be used to redress misuse of MNE power, even when such power is used in breach of the Declaration itself.
2.4 The ICC Bribery Code and How It Works
In 1996 the International Chamber of Commerce (ICC) re-released its Rules of Conduct on Extortion and Bribery in International Business Transactions. The foreword establishes the need for the revision, namely "scandals involving extortion and bribery (which) were a significant factor in toppling governments in many parts of the world", and the requirement that free trade be supported by "fair competition". Thus, "in addition to being a crime, offering or giving bribes may constitute acts of unfair competition, which could give rise to actions for damages." Part One contains recommendations to governments and international organizations: "Each government should take concrete and meaningful steps to enforce vigorously its legislation in this area... The World Trade Organization (WTO) should involve itself with these issues to support the OECD in the implementation of its Recommendations." There follow detailed recommendations on what form national legislation should take, including the requirement that proper mechanisms are in place for "surveillance and investigation" and to ensure violators are "subject to prosecution with appropriate penalties". Further, "Governments should periodically publish statistical or other information in respect of such prosecutions." "Economically significant enterprises" must be audited by "independent professional auditors." The Rules also call on regional administrations, including the EU, NAFTA and ASEAN to "satisfy themselves that appropriate legislation and administrative machinery" is in place to combat bribery.
The "Basic Rules" for enterprises are said to be "without direct legal effect". However, the exhortation to abide by the terms of the code is stricter than in the case of the labour codes discussed above: "All enterprises should conform to the relevant laws and regulations of the countries in which they are established and in which they operate, and should observe both the letter and the spirit of these Rules of Conduct." Some of the Rules are expressed as mandatory - for example: "All financial transactions must be properly and fairly recorded in appropriate books of account.." Boards of directors are charged with regularly reviewing compliance with the Rules, and should "establish procedures for obtaining appropriate reports for the purposes of such review... Companies should develop clear policies, guidelines, and training programmes for implementing and enforcing the provisions of their codes." 2.4.1 Follow-Up
An international structure within the ICC has been established to oversee the code's implementation via its national affiliates. The ICC calls for companies to endorse the Rules and "publicise them in their local environment", and to take an active role in lobbying governments to adopt and enforce appropriate legislation. The 1996 code calls for a study of best international practice within two years, and biennial reviews of the application of the Rules. The Rules are thus voluntary but the code requires concerted action to achieve its goals.
2.5 Conclusions on the International System of
Regulation of MNEs through Voluntary Codes
2.5.1 Inconsistency
The demands placed on MNEs by the OECD and the ILO are different. Which code should be regarded as authoritative is unclear, as no overarching interpretative or enforcement body exists. In general, the ILO code is more detailed and in some cases calls for a higher standard than the OECD. For example, it requires MNEs to adopt a higher standard in relation to wages than the OECD. The OECD instrument does not mention health and safety and only refers to training in passing, for example. Even where the same subject matter is dealt with, the ILO tends to expressly refer to more elements. Even small differences in wording between the codes would result in different obligations if the regimes were applied properly: for example, the ILO requires that MNEs notify "the appropriate government authorities" in case of major organizational change, while the OECD says that such notification should be made "where appropriate... to the relevant governmental authorities".
2.5.2 Not Legally Binding
The differences between the codes have little practical import, despite the overlapping membership of the ILO and OECD, because the codes are not strictly applied and are not legally binding. Although both the OECD and the ILO have the power to create legally binding regimes (albeit the ILO additionally requires ratification and implementation by member states), both chose to adopt non-binding regulation of MNEs. Further, the codes' non-binding nature has been continually stressed in the follow-up proceedings, particularly when parties have attempted to hold MNEs to the standards set out in the codes. According to Rubin and Hufbauer, trade unions originally called for the OECD code to be legally binding, after which the USA, Switzerland and Germany made a counter-claim for a legally-binding investment protection code. The current code emerged after both these claims were dropped. In other areas, the OECD has created instruments with "strict substantive commitments" which are legally binding - its Code of Liberalisation of Capital Movements and its Code of Liberalisation of Current Invisibles.
Of course, laws do not necessarily create compliance, nor does the absence of legally binding instruments necessarily indicate non-compliance with certain standards. The voluntary WHO/UNICEF baby milk code, for example, appears to have altered corporate behaviour, some claim as effectively as legislation. This outcome resulted from intense political pressure, internationally and at grass-roots level. This is now the challenge to communities and workers affected by the operations of MNEs, given the nature of the international rules discussed above.
2.5.3 Failure to Create International Rules
The third feature of these codes is the crucial reliance they place on national laws and practices: they do not establish a truly international system. There are two aspects to this - one is the requirement that MNEs obey national laws, and the other is that host countries treat MNEs no less favourably than domestic firms.
The requirement for MNEs to obey national laws is an absolute duty in the codes. There is no exception made for national laws which breach the fundamental tenets of the codes themselves or those of the other relevant international instruments, for example in the case of national regimes which legitimate discrimination against women. Thus the Sullivan Principles (see Section 4), with their exhortation to lobby for the change of national laws and, as de George claims, their actual requirement that apartheid laws be broken by MNEs, were arguably in breach of this requirement. While some have argued that it is proper to superimpose international standards upon countries which are "undemocratic", this possibility is not allowed for in the codes.
The OECD requires governments to treat MNEs no less favourably than domestic firms "in like situations". If "less favourable treatment" is taken to include requirements for MNEs to meet higher labour standards than domestic firms, then MNEs cannot be called on to do more than the national regulatory system. The OECD code does not clarify what is meant by "in like situations", nor what standards should apply when the MNE is not in a like situation with any domestic firm. Although the ILO code places more explicit weight on international Conventions and obligations, it limits all its key requirements by reference to national practice. Thus the ILO states that workers' right to be represented by a representative organization of their own choosing is to be exercised "in accordance with national law and practice". If national law and, more significantly, national practice undermines freedom of association, the ILO provisions are essentially nullified.
This is a particular problem in relation to MNEs, because national law and practice may treat MNEs differently (but not "less favourably") by allowing them to ignore domestic laws on worker rights. This may be a feature of the early days of industrialisation for example where export processing zones are set up or where the state seeks to foster certain industries in their infancy. For example, Wangel reports that the Malaysian Industrial Relations Act (1976) gave "protection of pioneer industries during the initial years of its establishment against any unreasonable demands from a trade union". In the Republic of Korea prior to 1987 "trade unions and industrial action were prohibited by law in order to attract potential investors".
In fact the deference to national systems can strip workers of their rights in many ways. It is not clear what the codes require where there is a gulf between national law and national practice, that is where entrenched patterns of non-compliance create their own illegal norms. If a country has no relevant legislation its national practice may be an absence of rights, as is the case with union recognition in the UK. Or what if national legislation and practice allows rights to be removed by agreement? In the absence of authoritative, independent interpretative and enforcement institutions to determine the proper balance between international and national requirements, we have to conclude that the international standards may be undermined.
2.5.4 The Myth of Symmetrical Interests
Another important feature of both codes is that they construct balanced schemes which purport to deal even-handedly with MNEs and their employees. This is seen in the 'symmetry' of some of the elements. But if it is accepted that there is an imbalance in economic, social and/or political power of employers and employees, and if one accepts that this might be exacerbated in the case of MNEs and their employees, the symmetry of rights is spurious.
The problem is compounded by the primacy given to the financial interests of MNEs. So the ILO Declaration says that workers may meet to consult each other "provided that the functioning of the operations of the enterprise...are not prejudiced"; host country management should be given certain opportunities "to the extent consistent with the efficient operation of the enterprise". The OECD's disclosure provisions require firms to provide information "having due regard to their nature and relative size in the economic context of their operations and to requirements of business confidentiality and to cost". It is difficult to see how these aspects of the codes, the ILO's in particular, are consistent with the Declaration of Philadelphia, which requires that "all national and international policies and measures, in particular those of an economic and financial character" should be judged "and accepted only in so far as they may be held to promote and not to hinder the achievement" of the fundamental objective of social justice.
2.5.5 Double Standards
The ICC Code on Bribery shows a higher level of self-regulation than in the labour codes. The ICC code is made up of "Rules", not principles or recommendations as in the case of the ILO and OECD codes, and is couched in stricter language than the labour guidelines. It is backed by a more active implementation strategy which involves attempts to obtain public acceptance of the standards at national level. And of course the Rules are only the tip of the regulatory iceberg - member companies are exhorted to obey national laws, but if these are incompatible with the codes companies should lobby states to improve their regulatory regime.
In conclusion, although the international codes cover some of the critical areas affecting the international interests of employees of MNEs, the standards they set are general and in some cases vague. More importantly, the codes cannot be enforced, and the supervisory bodies have tended to support MNE discretion in the application of the standards. Firms should not be able to benefit from unfair competition based on inadequate labour standards, and the concerns of business to police an issue like bribery should be extended to all labour matters, including a system of penalties for breach of standards.
3. ANATOMY OF A CODE OF CONDUCT
3.1 What Is the Code For?
There are many possible reasons for a firm to create a code of conduct. One set of reasons may be to do with internal management, particular across geographically far-flung sites: a code of conduct provides a framework for consistent managerial conduct and is thus a useful tool of internal governance. Another set of reasons may have to do with the benefits which would accrue to an organization (or industry grouping) through the proper application of fair labour practices. Ethical reasons may form the basis of the code, reflecting management's desire to be, or portray itself as, a good corporate citizen. A firm may want to use a code of conduct to extend or protect its reputation, which in some industries will be an important asset.
In reality, different reasons may co-exist. We will see in the case of the Sullivan Principles two primary purposes predominate: to move towards racial equality in US workplaces in South Africa and more generally in South African society, and to avoid financial losses which would occur from the anti-apartheid activism in the US ranging from bad publicity to compulsory disinvestment mandated by government. In the case of South Africa, the judgement on whether the two aims were compatible depends on whether continued investment by US firms in South Africa, even with complete compliance with the Principles, was supportive of apartheid or not. This in turn raises a critical question: who should make judgements such as this? What should guide a company with vested interests in its own survival and profitability in deciding social and policy issues in both home and host countries? The general principle must be that the businesses should consult widely in the areas where the code will impact before finalising their plans.
In the South African context, corporate decisions were shaped by the general international condemnation of apartheid. Such clear-cut international consensus is not common. (In the recent case of Myanmar, while the US and the EUhas instituted a policy of trade sanctions, other nations in the regional economic bloc have granted Myanmar admission to their group.) In the case of China, considered below in relation to the Toy Industry Code, the US government has 'de-linked' the issues of trade and human rights, creating a very different political climate in which business must make its ethical decisions.
Whatever their ostensible purpose, it is possible that codes will have unintended effects. For example, some schemes which establish central manufacturing points as an alternative to exploitative home work have had the unintended effect of reducing women's access to employment, and therefore income and opportunity. It is possible to introduce prohibitions on child labour which drive children into more dangerous and marginalised economic activity. In some cases it may be necessary to actively counteract potential negative consequences, and to recognise that not all of these can be addressed by the individual company acting alone. A multi-agency, multi-disciplinary approach may be required. UNICEF stresses that "any programme of elimination (of child labour) that does not provide reasonable alternatives for child workers - which from high moral ground simply casts them out of a workplace they had only entered due to extreme poverty - would trigger an avalanche of negative consequences." The development of all codes should involve a careful assessment of positive and negative likely outcomes. Companies acting in good faith would not introduce significant social change without a risk assessment, conducted with the full involvement of those most likely to be affected and their representatives.
But who are these people? How is the bargaining unit to be identified? What if they are not organised or represented in any formal way which allows ready consultation? What if there is a divergence of opinion amongst the affected community? What if the code will only bring long-term positive benefits while harming the current constituents in the short-term? There are no easy answers to these questions, and the advice of national and international expert practitioners should be sought as necessary.
This kind of planning exercise is not the same thing as inventing reasons not to act. It is to recognise the simple fact that labour standards imposed from afar without a sensitive understanding of the dynamics of the local community and workplaces may make people worse off.
3.2 Scope
Who does the code apply to? Kline states that a code of conduct should address "the corporation's functional relationship and responsibilities toward other constituency groups", not just the narrow focus of internal employees. This has been a critical issue in the 1990s codes, some of which extend to the employees of suppliers and sub-contractors. The second element is whose behaviour does it seek to influence: for example, the Sullivan Principles are not limited to the direct employees, but also to their families and the local community. It also calls for direct action (although very vaguely expressed) to change Government policy and the actions of other firms. It is important that MNEs adopting codes accept responsibility for the implementation of minimum labour standards across the whole production chain, if necessary through the use of production contracts which specify labour standards.
3.3 How Strictly Is the Code to Be Applied?
The first element is whether or not adoption of the code is compulsory for a company or group of companies. In the case of the Sullivan Principles, adoption (and subsequent opting out) was a matter for individual companies with operations in South Africa. No formal external pressure was applied except the threats and opportunities arising from the whole situation as the companies themselves perceived it. Companies could withdraw from the scheme if they wished (as an increasing number did over the years). By contrast, the Toy Code discussed below was adopted by the industry body, not individual companies, thus removing the companies one step from their obligations. (The Code's key provisions state that the industry body will "recommend" to its members that they undertake certain actions.)
The second element is whether, once the code is adopted, its terms are mandatory. This has two aspects. Are signatories bound by mandatory requirements once they have signed, and are these obligations expressed precisely enough to influence (and then judge) actual behaviour? In relation to the first aspect, the Sullivan Principles are expressed as mandatory: "Each signator will proceed immediately to..". In some cases, this element is matched by a clear and verifiable goal - for example to remove all race designation signs. In other more significant areas, the mandatory requirement is directed towards less clear goals (e.g. where important terms are not defined, as in the case of "equitable wage. (which is) well above the appropriate local minimum economic living level"). Again, the Toy Code sets a lower requirement - companies will "seek to ensure" the code's outcomes.
The third element is the way in which the code tests compliance to its own provisions. Even a voluntary code may be constructed in such a way as to allow for a definitive measure of compliance. However, the Sullivan Principles create a sliding scale of compliance which allows for different behaviours to be judged as in accord with the principles; and performance is aggregated across the Principles as a whole, so that failure to meet an individual requirement does not lead to a failing grade. Further, some signatories were exempt from measurement altogether.
3.4 Content of the Code
Choosing the content of a code is not a simple matter of selecting bits of international conventions and codes and cobbling together a document from disparate sources, as some commentators claim. The selection from these subtly, and not so subtly, different regimes should be made carefully, as these are statements which are going to bind company behaviour in important ways and must be appropriately authoritative. The selection would only be easy if this were not the case.
The ILO has identified a number of its Conventions which it regards as core minima, and these should also be regarded as the minimum provisions of any corporate code. The ILO's core of standards has a high degree of international legitimacy amongst the ILO's tripartite membership, as it is directly derived from the founding principles of the Organization which bind all members. Further, use of the ILO's core standards by way of explicit reference to them allows parties to a code to have access to the decades of ILO experience in implementing and interpreting these complex industrial rights.
The very great difference which the wording of codes will make to real world entitlements can be seen in the Table 3.1 below:
