AustLII Home | Databases | WorldLII | Search | Feedback

Melbourne University Law Review

Melbourne Law School
You are here:  AustLII >> Databases >> Melbourne University Law Review >> 2004 >> [2004] MelbULawRw 10

Database Search | Name Search | Recent Articles | Noteup | LawCite | Help

Cooney, Sean --- "A broader role for the commonwealth in eradicating foreign sweatshops?" [2004] MelbULawRw 10; (2004) 28(2) Melbourne University Law Review 290



[In this article I explore existing and potential forms of regulation that may prompt Australian firms to improve the conditions of workers in foreign sweatshops. I first note that workers experiencing sweatshop conditions are more likely to be linked to Australian firms through ‘buyer-driven supply chains’ than via direct employment relations. I then consider the current regulatory framework relevant to the relationship between Australian firms and foreign workers, surveying the common law, statute, self-regulatory initiatives in the private sector, and multi-stakeholder initiatives involving non-governmental organisations and international agencies. Each of these has only limited effectiveness. In the final part of the article, I investigate the potential for new regulatory strategies. I highlight the limitations of an overly prescriptive approach and examine the feasibility of various ‘meta-regulatory’ measures aimed at improving firms’ internal processes for dealing with foreign sweatshop-related issues.]



One of the major debates about economic globalisation concerns the extent to which firms based in developed countries have a responsibility to prevent labour abuses occurring outside their borders. Attention has focused on abuses in factories in developing countries that are linked through international supply chains to high profile multinational enterprises (‘MNEs’). One consequence of this attention has been the search for forms of regulation that track transnational production networks.

This article argues that existing forms of regulation in Australia have limited capacity to prompt Australian firms to improve the conditions of workers outside this country, especially those employed in sweatshops. I consider that additional, novel forms of government intervention may be appropriate.

Part II explores the connections between Australian firms and workers in developing countries, taking China as an example. I surmise that workers experiencing sweatshop conditions are more likely to be linked to Australian firms through ‘buyer-driven supply chains’ than via direct employment relations.

Part III analyses the current regulatory framework relevant to the relationship between Australian firms and foreign workers, focusing on four areas: the common law, statute, self-regulatory initiatives in the private sector, and multi-stakeholder initiatives involving non-governmental organisations (‘NGOs’) and international agencies. Each of these areas holds some promise for addressing sweatshop issues, but present evidence suggests that their impact is likely to be marginal.

Part IV investigates further governmental strategies for prompting Australian firms to be more responsive to the conditions faced by foreign workers. I highlight the limitations of an overly prescriptive approach and explore the feasibility of various meta-regulatory measures aimed at improving firms’ internal processes for dealing with sweatshop issues. The analysis is preliminary and tentative, but I nonetheless hope that it will contribute to the development of viable regulatory means for securing measurable improvements in some of the world’s worst workplaces.[1]


The globalisation debates of recent decades have highlighted the relationships between firms based in the developed world, including Australia, and workers in developing countries. These relationships take a variety of organisational forms. A connection between a firm and a foreign worker is most clearly established in the case of direct employment. However, this article will focus on the relationship between Australian firms and foreign workers mediated through supply chains, as it is those workers who generally experience the worst working conditions.

A Direct Employment[2]

It would appear that when Australian businesses are direct employers of overseas workers, they tend to operate in the developed world.[3] Since the operation of legal, economic and industrial relations systems tends to be broadly similar across industrialised countries, in contrast to developing countries, it may be presumed that most foreign direct employees of Australian firms experience working conditions roughly similar to Australian employees, and have access to comparable remedies where the conditions violate their contractual, industrial or statutory entitlements.[4] Where labour abuses occur, they are likely to be of a different order from those occurring in the sweatshops of developing countries.[5]

There is, however, an important exception to this generalisation. The Australian mining industry is a significant direct employer of labour in developing countries. Well-known companies such as BHP Billiton and the Western Mining Corporation (‘WMC’), and their subsidiaries and joint ventures, have attracted attention in Australia for labour abuses in their mines overseas.[6] The extractive industries are capital and technology-intensive (with large costs involved in exploration, mining and transportation) and are bound to fixed areas in which the minerals are located. Labour costs constitute a relatively low proportion of expenditure.[7] The problematic nature of these MNEs’ mining operations

therefore arises not so much in the creation of sweatshops, but rather because of the impact on indigenous communities and their land holdings, the effect on the environment, and the relationship between the MNEs and law enforcement agencies. Labour abuses are likely to be closely linked to these issues.[8]

B Indirect Relations: Buyer-Driven Commodity Chains

The sweatshop controversy arises primarily in the context of the indirect relationship between MNEs and workers engaged in labour-intensive industries in developing countries. Indirect relationships flow from the ‘factory-less’ production system adopted by corporations such as Nike.[9] In this form of flexible production,[10] MNEs employ ever fewer people in the developed world to manufacture their products, instead subcontracting the work out to factories in industrialising states. Gary Gereffi has described this production system as consisting of ‘buyer-driven commodity chains’.[11] In these chains, common in the clothing, footwear, apparel and toy industries, MNEs are merchandisers rather than manufacturers.

[They] design and/or market, but do not make, the branded products they sell. [They] rely on complex tiered networks of contractors that perform almost all their specialized tasks. Branded manufacturers may farm out part or all of their product development activities, manufacturing, packaging, shipping and even accounts receivables to different agents around the world.
The main job of the core company in buyer-driven commodity chains is to manage these production and trade networks and make sure all the pieces of business come together as an integrated whole. Profits in buyer-driven chains thus derive ... from unique combinations of high-value research, design, sales, marketing, and financial services that allow the buyers and branded merchandisers to act as strategic brokers in linking overseas factories and traders with evolving product niches in their main consumer markets.[12]

Buyer-driven production chains are generally labour-intensive at the manufacturing stage, involving highly competitive, decentralised production units, very frequently located in developing countries.[13] These production units are not usually owned or operated by MNEs or their subsidiaries. The legal relationships between the MNEs and local producers are based primarily on a succession of contracts.

Many factories in the buyer-driven commodity chains have very harsh labour conditions.[14] While most developing countries now possess, on paper at least, sophisticated legal provisions dealing with employment relations, the implementation of these provisions is severely flawed.[15] Employees encounter great obstacles to enforcing both their individual contractual entitlements and the labour standards determined by the state. Many factors contribute to this. The developing countries in which the factories are located frequently have weak legal infrastructures. There are structural flaws in the regulatory framework (sometimes deriving from the failure to adapt a foreign legal model to a local context). Limited resources are afforded to enforcement agencies. Regulatory agencies lack training and expertise and the regulatory culture often tolerates non-compliance and corruption. There is widespread ignorance of the law among firms and employees, as well as fraudulent behaviour (such as false record-keeping) by those firms that do understand their legal obligations.

These legal deficiencies are compounded by the law’s political, economic and social context. At a political level, authoritarian regimes provide little or no space for labour monitoring organisations such as autonomous trade unions, NGOs or an independent media. They may also subordinate courts to political imperatives. From an economic perspective, there are powerful incentives to disregard worker entitlements. As labour costs tend to be a very significant factor in the competition between factories to secure production contracts with the MNEs controlling supply chains, or with intermediate firms, there is great pressure to reduce labour expenditure. Since much of the production process in the relevant industries requires relatively unskilled labour, and since unskilled labour is plentiful in most developing countries, there is little countervailing pressure on firms to invest in their workers’ welfare.[16] In addition, as local firms are required by the MNEs to respond rapidly to the demands of just-in-time production and shifting consumer preferences, firms need to engage in bursts of high-intensity work. Finally, at a social level, managers in many firms are often influenced by societal norms that encourage militaristic and discriminatory work practices.

This combination of circumstances turns many factories into sweatshops. Workers face long working hours at low wages in dangerous environments under regimented authority. They are given little training and are afforded scant participation in decision-making, even regarding health and safety matters.

C The Example of China

China provides a paradigmatic illustration of the way these processes lead to the proliferation of sweatshops.[17] In many ways, China’s economic reform program has been resoundingly successful, providing prosperity and a growing range of employment opportunities for many of its citizens. Indeed, the proliferation of labour-intensive factories has provided many workers with opportunities to increase their incomes and escape from wretched rural poverty.[18] Moreover, because many factories, especially in the export-oriented sector, have been newly built and incorporate the most recent technology, the working environment may sometimes be as good as, or indeed superior to, that in the equivalent industry in Australia.[19]

However, many workplaces forming part of buyer-driven commodity chains certainly merit the ‘sweatshop’ label. Serious labour abuses are common.[20] Basic safety standards are violated, often resulting in severe injury or death.[21] It is typical for workers, even in many well-equipped factories, to be denied their accrued contractual entitlements, to pay heavy bonds upon commencing employment, to be fined for petty violations of rules, to work extreme hours[22] and to be subjected to militaristic and humiliating disciplinary regimes.[23]

China now has a well-structured labour law framework,[24] which enables employers and employees to conclude voluntary contractual arrangements, subject to a floor of rights incorporating many international labour standards,[25] including anti-discrimination principles.[26] However, the protections apparently accorded by the law are undermined by structural features of the legal system. There are many gaps and inconsistencies in the text of the written law. For example, China still lacks a substantive framework for collective bargaining.[27] Legislative instruments from various organs of the government set conflicting standards on working hours.[28] The legal system, while improving in many respects, remains plagued by corruption, lack of training and other shortcomings.[29] Regulatory agencies lack the resources to police the law, and firms frustrate inspectors by producing false documentation and coaching workers to indicate compliance with the law.[30] Workers, especially migrant workers, have very limited access to the courts or other avenues of redress.

These problems are exacerbated by the interactions between the law and the Chinese political, economic and social systems. The Chinese political system is dominated by the Chinese Communist Party (‘CCP’). While workers have rights to form trade unions, these must be controlled by the official trade union organisation, the All China Federation of Trade Unions (‘ACFTU’), which is subordinate to the CCP and is bound to implement party policy that is sometimes favourable to business at the expense of workers, especially at the local level.[31] The media remains controlled, and, despite some moves towards greater institutional autonomy, regulatory agencies (including the judiciary)[32] lack independence from the party-state.

Economically, any local firm that attempts to comply with labour law without support from its customers faces being swiftly driven out of business. With reduced working hours, it would not be able to meet client orders quickly enough. It would lose its contracts to firms that satisfy orders more rapidly because they continue to evade the law. Given these circumstances, it is economically rational for firms to circumvent the law.[33] Even if all Chinese firms complied with the law, loss of business and consequent unemployment might still result, this time on a national scale, as supply chains would reorganise around other developing countries.

On the other hand, there is little supply-side pressure to improve working conditions. China has between 80 million and 100 million migrant workers, mainly from rural areas in which there is large-scale unemployment and underemployment. They constitute the vast majority of workers in the factories of China’s prosperous coastal cities, such as Shenzhen, Dongguan and Shanghai. Lacking residency status in the cities in which they work,[34] even though they may work in them for up to 20 years, these workers are treated as a huge casual workforce. Most of these workers are women, and almost all are below the age of 35, with an average age of around 23.[35] They are easily replaceable, so firms have no incentive to invest in their well-being.[36]

There are two important social factors further worsening conditions. The first is the militaristic management style of many businesses: Taiwanese and South Korean managers, who run many of the sweatshops, have been highly influenced by their experiences of the former authoritarian regimes in those countries.[37] The second is discrimination based on residency and gender. Most rural female workers are unlikely to be able to improve their lot (through, for example, being promoted), because management positions are kept in the hands of predominantly male non-nationals and locals, who receive much better training and treatment from the time they commence employment. Discrimination also weakens the possibility for remedial action through local unions, staff-management consultative congresses or local political institutions, as these bodies are generally dominated by local residents unsympathetic to the plight of rural female workers.[38]

D The Move to Regulate from the Top of the Supply Chain

The inability — or unwillingness — of developing states such as China to address labour issues adequately has prompted a search for other means of improving sweatshop conditions. The prospects for effective intervention through international agencies are slim: the World Trade Organization (‘WTO’) has rejected trade-labour linkages[39] and the International Labour Organization (‘ILO’) is unable to enforce its labour standards. Consequently, many of those concerned with improving sweatshop conditions are examining the other end of supply chains, in particular merchandisers and retailers, as a site of intervention. Cannot MNEs and other businesses in the developed world assist the workers that produce the goods they merchandise?

Some suggest that the most that firms in industrialised countries can do for the sweatshop workers is to trade with those workers’ employers or contractors. Sweatshops are perhaps a necessary evil on a country’s path to economic prosperity. According to this line of argument, China and other developing countries may follow the trajectory of East Asian export-oriented newly industrialising countries (‘NICs’), such as Taiwan, South Korea and Singapore, where rising productivity and educational levels (albeit under authoritarian conditions) led to better wages and working conditions.[40] However, even if this argument is correct — and there are good reasons to doubt it[41] — it does not necessarily follow that no strategies other than open trade can be adopted to reduce or prevent serious labour abuses in developing countries. Firms in the developed world can, of course, both engage in open trade and use their market power, their resources and their expertise in areas such as occupational health and safety, employee training, equal opportunity and employee participation, to improve the crude conditions suffered by disadvantaged workers.

The question of how firms should be required to intervene in enterprises they do not own does not have a straightforward answer. Supply chains take many complex forms. Where a firm exercises considerable control over the activities of its subcontractors, there is a good case for attributing some ethical, and arguably legal, responsibility for those activities to the firm.[42] However, as the Organisation for Economic Cooperation and Development’s (‘OECD’) Business and Industry Advisory Group points out, the degree of control a firm has over a supply chain is likely to vary greatly depending on ‘the industry in which it operates, the quantity of suppliers, the structure and complexity of the supply chain, and the market position of the enterprise’.[43] The last factor is likely to be the most significant. On the one hand, a firm could be the major customer of a supplier, so that the supplier is dependent on the firm for its economic survival and thus easily amenable to its influence.[44] Such a firm would be in a strong position to exercise that influence and use its resources to create better workplace conditions in that supplier. Failure to seriously attempt to do so would rightly attract opprobrium. On the other hand, a firm may be only one of a large number of customers. It may not be in a strong position to pressure the factories in its supply chain, although it may be able to make some interventions to improve working conditions, especially if it acts in concert with other firms.

This complexity does not mean that obligations cannot be imposed on corporations in relation to working conditions in their supply chain. It is, however, one reason why those obligations must be carefully formulated. Further reasons for caution appear later in this article.

E Australian Firms and Global Supply Chains

What is the position of Australian firms vis-a-vis the sweatshop workers connected to them through supply chains? The target of international (and indeed Australian) activism has been the major brand-name merchandisers that control commodity chains in the clothing industry. These firms, such as Nike, Reebok, Levi-Strauss and Gap, are based in the developed nations of the Northern Hemisphere. However, contracting chains also connect workers in developing countries with Australian trading and retail firms (and Australian consumers). While the precise nature of this connection is unclear,[45] developing countries supply a very significant — and increasing — portion of manufactured products to the Australian market. China, for example, is now overwhelmingly Australia’s largest source of ‘miscellaneous manufactured articles’, a trade category which includes clothing, footwear, toys, furniture and many plastic articles.[46] There is, therefore, a chain of contracts connecting Australian consumers of ordinary goods like apparel and toys, manufacturers, wholesalers and retailers who import these goods, and workers in China and other developing countries.

Now, clearly, Australian firms do not have the market power of the ‘big name’ Northern Hemisphere MNEs. It might, therefore, be concluded that there is little Australian firms could be expected to do to influence work practices in developing countries. Indeed, sweatshop campaigns involving Australian firms have been generally domestically-oriented, focusing (very successfully) on homeworkers in the Australian clothing industry.[47]

Some Australian businesses, have, however, attempted to ameliorate labour conditions in offshore supply chains. For example, I, together with my colleague Dr Jill Murray, have participated in the mapping of a buyer-driven commodity chain linking an Australian retailer of optical frames (ModStyle), which is owned by the Australian charity the Brotherhood of Saint Lawrence (‘BSL’), and 13 factories in China that produce the large majority of those frames.[48] Most of the factories are owned by Hong Kong entrepreneurs, and vary in size from 150 to 4000 employees. The factories manufacture for the world market, including for brands such as Gucci, Timberland, DKNY, Esprit, Disney and Calvin Klein. Consequently, the proportion of annual production for the Australian firm is very small — around two per cent. The conditions in the factories vary,[49] but they generally have some ‘sweatshop’ aspects, including the characteristic worker profile.[50]

The BSL has inspected the factories in its supply chain. Rather than insisting that all the factories comply with Chinese law, or with a particular code of conduct, it has entered into negotiations with several of the owners with a view to implementing practical measures to improve occupational health and safety.

As a charitable organisation, the BSL has an inherent commitment to ethical business practice. There is little evidence that large numbers of Australian firms that engage in trade with developing countries are taking similar steps to investigate or improve working conditions in their supply chains.[51] Should the law encourage them to do so? If so, how? The remainder of this article attempts to respond to these questions by reviewing the existing legal and self-imposed obligations placed on Australian firms in respect to foreign workers. It then considers the scope for further government intervention.


This part of the article assesses the impact of four forms of regulation that may influence the relationships between Australian firms and overseas workers. These are obligations imposed by the common law, extraterritorial federal legislation, firms’ internal policies relating to corporate social responsibility and voluntary initiatives involving a range of stakeholders and governmental agencies. Each impacts to some extent on the behaviour of Australian firms towards overseas workers, but none establishes a comprehensive regulatory framework.

A Common Law

In certain circumstances, Australian firms are legally obliged to avoid causing harm or loss to persons outside Australia.[52] If a firm breaches this obligation, non-Australians who claim to have suffered resulting harm or loss may bring an action against that firm in an Australian court. However, attempts to invoke the common law to require Australian firms to take greater action to improve the poor working conditions of foreign workers face numerous legal and practical difficulties.

First, the relevant obligations imposed on Australian firms by the common law are quite limited. Generally, contractual obligations arise only where there is a direct employment relationship between a firm and a foreign worker.[53] Australian firms will not usually have any responsibility in contract where a worker is employed by a non-Australian entity, such as a subsidiary, a joint venture or a subcontractor.[54]

Where the relationship between an Australian firm and a foreign worker is indirect, the relevant obligations derive from tort, in particular negligence.[55] Several cases suggest that it may be possible to sue parent corporations in Australia where their overseas subsidiaries have failed to adopt safe work practices, at least where the parent corporation exercises considerable control over the operations of those subsidiaries.[56] However, in relation to the substantive question of liability, in those cases involving diffuse corporate structures and complex supply chains, there are likely to be major obstacles to establishing a duty of care, especially where the Australian firm exercises little influence over the entities employing the relevant workers.

Another problem confronting plaintiffs suing in Australia is the principle of forum non conveniens.[57] This principle refers to the discretionary power of a court to decline jurisdiction and to grant a stay of proceedings brought before it on the basis that the court is an inappropriate forum for those proceedings. Common law countries differ on the test for granting a stay. In Voth v Manildra Flour Mills Pty Ltd, a majority of the High Court held that a stay may be granted only when a court is a ‘clearly inappropriate forum’.[58] Where the plaintiff is a foreign resident, but the defendant is an Australian resident, the application of this test has often resulted in a defendant’s application for a stay being refused. For example, in James Hardie Industries Pty Ltd v Grigor,[59] the New South Wales Court of Appeal dismissed an appeal against the refusal of a stay application made by the defendant Australian firm (that is, the Court confirmed that the Australian forum was appropriate). The case involved personal injury proceedings brought in the Dust Diseases Tribunal of New South Wales. The plaintiff alleged that he had contracted mesothelioma through inhaling asbestos dust while cutting and drilling building materials manufactured by the Australian corporation and its New Zealand subsidiary. The plaintiff resided in New Zealand and was unable to travel to Australia because of his illness. Much of the evidence was in New Zealand, although evidence of the state of corporate knowledge and of the nature of the manufacturing process was located in New South Wales. The applicable law was that of New Zealand. Despite these considerations, a majority of the Court held that the discretion to refuse a stay had not miscarried.[60]

The application of the Voth test in cases such as Grigor suggests that there is significant scope for foreign workers to maintain common law actions in Australian courts in respect to personal injuries they have suffered as a result of the actions of Australian-controlled firms. However, the split decision in Grigor means that such workers cannot be certain that Australian courts will hear their claims over forum non conveniens objections. In particular, where the cause of action arises in a foreign jurisdiction, the applicable law is that of the foreign jurisdiction,[61] and the plaintiffs, most of the witnesses and most of the documentary evidence are located there, a stay could still possibly be granted.

A third legal obstacle concerns the aforementioned fact that where a foreign worker sues an Australian firm in an Australian court, that court may be required, under the rules of private international law, to apply foreign law.[62] If the cause of action is in contract (for example, a dispute over the terms and conditions in an employment contract between an Australian firm and a foreign worker), an Australian court may be required to apply foreign contract law. In the absence of an express or inferred agreement on choice of law to the contrary, the applicable law will be that of the foreign jurisdiction in which the employee was engaged and works, since that will usually be the legal system with which the transaction has the ‘closest and most real connection’.[63] Where the cause of action is in tort, the applicable principle in Australia, in doubt for some time, has now been settled by the High Court in Regie Nationale des Usines Renault SA v Zhang.[64] The applicable law will be the law of the place where the tort was committed — the lex loci delicti.[65] In an action launched by a foreign worker injured in a foreign jurisdiction, then, an Australian court may need to apply foreign law.[66] There may, therefore, be little in the way of substantive legal advantage to be gained by foreign workers suing in Australia rather than in their own domestic courts. Indeed, foreign plaintiffs may be put to the expense and trouble of proving foreign law in an Australian court.

The incentives for foreign workers to sue in the Australian court system in order to force firms to comply with their common law obligations derive from the nature of the courts themselves. Australian judges are, on the whole, likely to be better trained and less corrupt than judges in a number of developing countries, including China.[67] Further, enforcement of civil judgments will sometimes be easier in Australia. This is owing to the systemic problems of enforcement often experienced in the legal systems of developing countries,[68] and the fact that an Australian firm will often have more significant assets in Australia than in the foreign jurisdiction.

However, even where these advantages exist, there are overwhelming practical obstacles facing most workers who might wish to pursue an action in Australia. With the possible exception of some senior executives, individual foreign workers are most unlikely to have the resources to maintain an action in Australia. The costs associated with locating Australian lawyers, paying for their services, transporting witnesses to Australia (assuming that they are willing or can be compelled to go), providing interpreters, translating documents and forwarding them to Australia and so on are utterly beyond what most workers (particularly in developing countries) could afford.[69] The only circumstances in which litigation by foreign workers in Australia is feasible is in the case of large class actions, such as asbestos litigation or a major industrial disaster.[70] Even in these categories of cases, defendant firms are likely to oppose the actions strenuously.[71]

We can conclude that there is only limited potential to use the common law of obligations to regulate the relationship between foreign workers and Australian firms. Its most significant effect may be to induce firms (through the threat of large damages awards) to create systems of work in their overseas operations which reduce the risk of large-scale personal injuries. There would appear to be little scope to invoke the potential for tort and contract actions as a vehicle for offshore regulation of issues such as oppressive working hours, the abuse and assault of individual workers, discriminatory conduct and denial of the right to organise. This is certainly the case where the relationship between the Australian firm and the overseas workers is indirect.

B Extraterritorial Legislation

Turning to legislative intervention, there are a number of federal laws that regulate employment conditions outside Australia. However, like the common law, this extraterritorial legislation has only a very circumscribed impact on sweatshop labour conditions.

Subject to the Constitution, the federal parliament is competent to legislate extraterritorially.[72] However, unless a contrary intention appears in a federal Act, that Act will apply only to ‘matters and things in and of the Commonwealth’.[73] The major federal laws regulating workplace practices have only narrow extraterritorial application. For example, the Workplace Relations Act 1996 (Cth) has extraterritorial operation only in relation to certain shipping, aircraft and public sector employees and to workers in the Timor Gap.[74] Apart from these exceptions, the Workplace Relations Act 1996 (Cth) does not, in general, regulate relationships between foreign workers and Australian firms.[75] The same may be said of other major federal employment statutes.[76]

There are, however, several Australian statutes that, although not primarily directed at labour, do target extreme forms of abuse. They also suggest various ways in which legislation could be devised to regulate overseas working conditions more extensively. Certain legislation imposes sanctions on Australian firms for specified offences committed offshore. For example, the Commonwealth Criminal Code imposes heavy penalties on persons who bribe foreign officials,[77] engage in slavery[78] or cause a person to enter or remain in sexual servitude.[79] The relevant provisions have wide extraterritorial effect; if certain conditions are fulfilled, they enable a person who has committed one or more of these offences overseas to be prosecuted in Australia.[80] A related extraterritorial statute is the Crimes (Torture) Act 1988 (Cth), which renders state-sanctioned torture[81] in a foreign country punishable under Australian law if the perpetrators are Australian or are in Australia at the time when the prosecution is initiated.[82] These provisions all give effect to Australia’s obligations under various international Conventions to which it is a party.[83] Some of these are potentially applicable to workplace relations. For instance, if an Australian resident engaged a worker on conditions amounting to slavery or conducted a business involving workers providing sexual services against their will,[84] the resident would be guilty of a serious offence, whether the conduct took place in Australia or offshore. An Australian resident could also be prosecuted in Australia for torturing or inflicting cruel and inhuman treatment upon a worker overseas under the Crimes (Torture) Act 1988 (Cth), but only if he or she were acting at the behest of a state.[85]

An important aspect of the provisions in the Criminal Code is that they apply to corporations.[86] Most offences under the Criminal Code, including those of slavery and sexual servitude, require ‘physical elements’ and ‘fault elements’.[87] In the case of a corporation, the ‘physical element’ is satisfied if an ‘employee, agent or officer’ of that corporation acts ‘within the actual or apparent scope of his or her employment’ or authority.[88] The ‘fault element’ is satisfied if the corporation ‘expressly, tacitly or impliedly authorised or permitted the commission of the offence’.[89] Thus, if an Australian corporation, through its employees, agents or officers, engaged workers in another country in such circumstances that they were effectively slaves for the purposes of the Criminal Code, that corporation could be prosecuted.[90] While a corporation cannot of course be imprisoned (although its officers and employees could be), it can be given a very substantial fine.[91]

Nevertheless, Australian corporations are not subject to any statutory obligations or sanctions with respect to most forms of labour abuse that take place offshore, including torture unrelated to the state.

C Private Sector Initiatives

The most important response to the global sweatshop phenomenon has not emerged from the formal legal system. Rather, it has been in the form of self-regulatory initiatives such as voluntary corporate codes of conduct and ‘triple bottom line’ performance measurement, both of which aim to ensure that the firms operate in a socially responsible manner. This section focuses on initiatives devised by firms through their own internal processes. The following section examines initiatives which include significant involvement from the public sector or NGOs.

Over the last decade, partly — perhaps largely — as a result of significant consumer and investor pressure, firms have increasingly adopted various ‘corporate social responsibility’ (‘CSR’) initiatives[92] to address the social and environmental impacts of their operations.[93] One major CSR tool has been the corporate code of conduct. Such codes set out labour or environmental standards with which the firm commits to comply.[94] These codes are sometimes linked to public reporting of labour and environmental performance or to monitoring by an external agency. Private sector codes of conduct have been produced by individual firms (such as Nike and Reebok), and on an industry basis (such as in the toy industry).[95] There have been significant private sector CSR initiatives in Australia relating to domestic labour standards.[96] However, with the exception of measures in the mining industry,[97] Australian firms appear to have paid relatively little attention to calls to employ CSR measures pertaining to labour practices in developing countries.[98]

The effectiveness of private sector initiatives (and of codes of conduct in particular) is the subject of claims and counterclaims by businesses, labour activists and academics.[99] Self-regulatory initiatives have several advantages over the traditional so-called ‘command and control’ regulation by the state. For example, since firms undertake to monitor their own compliance rather than simply depending on external supervision, codes of conduct internalise enforcement and thus minimise compliance costs for government. As an expression of a firm’s ethical commitments, codes also appeal to the reputation and self-regard of actors within firms, not simply to their economic self-interest calculations.[100] They potentially make regulation adaptive to the specific conditions of a firm or industry, which is often difficult to achieve through state-based law. They can also produce innovative solutions to compliance problems, and they can be flexible, allowing for ‘rolling standards’. Finally, unlike domestic law, codes of conduct can potentially cross international boundaries and reach into contracting chains. Firms from developed countries can endeavour to ensure that not only their direct employees, but also the employees of their subcontractors, enjoy the protections in their codes. Thus, many Chinese factory owners (including those from Hong Kong and Taiwan) have committed themselves to adhere to codes of conduct, usually under pressure from MNEs.[101]

However, self-regulatory initiatives suffer several deficiencies as a mechanism for promoting labour standards.[102] First, many firms fail to take meaningful action. As firms are often led to adopt measures on labour standards in response to public pressure, those firms not in the public eye have limited incentive to act. This is often the case with enterprises that produce intermediate rather than consumer goods.[103] Some firms may give little attention to the issue. Others may engage in public relations exercises that reflect no genuine commitment to improving working conditions, and that cannot be evaluated or enforced.[104]

Second, even in cases where there is a genuine attempt to adopt self-regulatory measures to improve labour standards, those measures may be selective. Thus, many codes do not extend to all matters that the fundamental international labour rights seek to protect.[105] For example, many codes do not promote collective bargaining,[106] including those in the Australian mining industry.[107] There is also evidence that some initiatives target those issues that have the greatest potential to harm corporate reputation (such as child labour), rather than adopting a comprehensive framework to protect the interests of the workforce.[108]

Third, many private sector initiatives constitute only a temporary and contingent commitment to workers.[109] Thus, once workers are dismissed (remembering that many of the industries concerned have a preference for young migrant women) or production moves to another location, the initiatives have no further application. This is despite the fact that it is at this point that many workers will be most adversely affected.

Fourth, it is difficult to compare the relative effectiveness of firms’ CSR measures. This difficulty arises both from the variation between the content of codes and the failure to disclose social and economic performance according to generally agreed criteria. Australian firms are less likely than their overseas counterparts to produce reports on their social and environmental practices.[110] This is particularly true of the trade and retail sectors.[111] The authors of a 2002 survey on ‘triple bottom line’ accounting, commissioned by the Commonwealth Government, reported that:

Social indicators are still at the early developmental stage with very few firms measuring such issues in a consistent and frequent manner. While it is common to find information on company websites with regard to the dollar amount allocated to various community groups, not-for-profit organisations, sponsorship programs and community involvement activities, there are very few quantitative measures on the outputs and outcomes of these initiatives.[112]

This problem can be addressed to some extent by codes formulated by industry associations. Association members can then sign up to a base code and, if the codes set out benchmarks and provide for appropriate public reporting and auditing, performance across firms can sometimes be evaluated.[113] For example, the MCA publicises ‘best practice’ in safety and health through its ‘MINEX Safety and Health Excellence Awards’. However, comparison is problematic where there is no agreed industry standard, where industry codes omit key labour rights, and where industry codes set out obligations in terms incapable of being empirically assessed.

The recent developments of evaluation and reporting frameworks by both domestic[114] and international agencies (especially the Global Reporting Initiative)[115] may lead to greater systematisation. However, it is not yet clear which, if

any, of these frameworks will become widely accepted internationally.[116]

Fifth, even where firms evaluate and report on performance, in the absence of effective independent auditing their reports may lack credibility. Many firms do not establish external verification systems.[117] Even where there are such systems, they may not operate well. The auditors may not be adequately trained or familiar with local conditions. Empirical studies have shown that even major accounting firms that carry out social audits are frequently deceived, relying excessively on management accounts.[118] In some instances, the independence of the ‘independent’ audits may be compromised by financial linkages with the firm being supervised.[119] Moreover, as auditing systems are diverse and fragmented, there may be no common basis for the evaluation methods of different auditors.

Sixth, codes and monitoring systems may conflict with a domestic legal system. This is discussed in relation to Australia in more detail in Part IV(A).

Finally, and most significantly, private sector initiatives are often devised without consultation of the relevant stakeholders, especially the workers in the developing countries to whom they are intended to apply.[120] As Kai-Ming Liu and Shen Tan write in relation to their empirical findings on corporate social responsibility in Guangdong Province in China:

Workers have only very limited participation in the factory and in the implementation of codes of conduct. When it comes to putting the codes into effect, more than 80 per cent of clients and factory managers view workers only as obedient and passive recipients of the codes. They have not considered letting workers participate in the processes of code monitoring and implementation.[121]

As Adelle Blackett points out,[122] this exclusion entrenches managerial unilateralism, at the expense of worker–management co-regulation through collective bargaining, collaborative committees and other vehicles for employee expression.[123] Not only does this foreclose the possibility of participative, deliberative problem-solving at the local level, but it precludes what is potentially the most effective code monitoring system, direct worker supervision.[124]

Further, worker exclusion undermines one of the potential strengths of codes: the capacity of codes to be context-sensitive and respond directly to the major labour issues in local workplaces. Indeed, as Jill Murray comments, exclusion of local workers may make codes counterproductive: ‘labour standards imposed from afar without a sensitive understanding of the dynamics of the local community and workplaces may make people worse off.’[125]

Codes of conduct drafted together with worker organisations and NGOs based in developed countries go some way to remedying this deficiency. Such codes are more likely to address issues such as protection from physical abuse and the notification of standards.[126] However, even here, the voices of local workers are not being heard.

Given all the shortcomings outlined in this section, the effectiveness of private sector initiatives as a means of addressing the problem of sweatshops is dubious.[127] Some firms are now attempting to tackle these problems through specific and context-based initiatives going beyond code formulation, monitoring and reporting. Thus, for example, in addition to implementing its own code of conduct, Reebok is experimenting with various innovative ways of educating workers about workplace safety and labour rights and of fostering worker participation.[128] These initiatives involve considerable investment by the firm.[129] Such context-specific measures — involving careful groundwork, ongoing evaluation and specific financial commitments — may prove to be the most effective and credible model for private sector initiatives. However, Reebok’s approach has not yet silenced the criticism that labour abuses are common in its supplier factories, in violation of its codes.[130]

It therefore seems that private sector initiatives dealing with foreign supply chain workers will not in the foreseeable future be sufficiently systematic and comprehensive to bring about a transformation of sweatshops, although they may lead to certain limited improvements in working conditions. This conclusion is particularly true for Australia, where there are, in any case, very few such initiatives, let alone initiatives going beyond codes of conduct and social reporting.[131]

D Multi-Stakeholder Initiatives

In response to these difficulties with private sector measures, a number of schemes that involve not just businesses, but also NGOs, trade unions, governments and international agencies, have recently become prominent.[132] These schemes adopt diverse methodologies and objectives, but all go some of the way to addressing the difficulties of purely private codes. Three of the most significant are the OECD Guidelines for Multinational Enterprises (‘OECD Guidelines’), Social Accountability International (‘SAI’) and the Ethical Trading Initiative (‘ETI’). These are all international in scope and all address labour issues in supply chains. They each incorporate core ILO Conventions, as well as covering issues such as occupational health and safety. However, they represent three different approaches to co-ordinating private sector initiatives.

1 The OECD Guidelines

The OECD Guidelines are voluntary CSR standards developed and coordinated through an international agency, the OECD.[133] By virtue of its OECD membership, the Australian government is committed to promoting the Guidelines. The Guidelines have been prepared in consultation with business and trade union representative bodies, as well as NGOs. They were originally issued in 1976, but were extensively revised in 2000.[134] They include recommendations on responsible business conduct, and cover matters such as bribery, environmental and consumer protection, taxation and, most relevantly for present purposes, employment and industrial relations. The employment and industrial relations section of the Guidelines reflects the core ILO Conventions,[135] as well as issues such as occupational health and safety, disclosure of and consultation on organisational change and negotiation with bona fide employee representatives.[136] The Guidelines apply to the operations of MNEs, even in non-OECD countries, and require MNEs to ‘encourage, where practicable, business partners, including suppliers and subcontractors, to apply principles of corporate conduct compatible with the Guidelines.’[137]

The Guidelines are not legally binding either on states or on MNEs but are promoted through National Contact Points (‘NCPs’) in OECD countries.[138] The NCPs are also responsible for conducting consultations with the relevant stakeholders and for dealing with ‘implementation in specific instances’, which may involve complaints that the Guidelines have been breached.[139] In such cases, the NCP can offer ‘good offices’ to help resolve disputes and can issue recommendations. Such good offices procedures are voluntary and confidential. As of June 2003, 64 ‘specific instance’ complaints have been made to NCPs since the 2000 revisions.[140] Most of these have dealt with labour issues and two thirds have concerned the operations of MNEs in non-OECD countries. There have been no specific instance complaints relating to operations in Australia since the 2000 revisions.[141]

There is no formal system for monitoring compliance with the Guidelines, or for requiring MNEs to conduct and publish audits. Nor does the NCP have the resources, or the government mandate, to attempt to coordinate or evaluate voluntary CSR initiatives. From this perspective, the Guidelines are not particularly effective. On the other hand, the Guidelines have significant normative force, constituting an agreed statement of principles by the OECD nations. They appear to be playing a significant role as a reference point for policy-making in relation to CSR.[142] Moreover, the Australian NCP is actively promoting the Guidelines with Australian business and seeking to diffuse information about the Guidelines and other CSR initiatives through a well-developed website.[143]

2 The SAI Scheme

A second form of multi-stakeholder initiative involves the establishment by a non-governmental agency of a code, together with monitoring and evaluation procedures, and sometimes a certification process. SAI is a sophisticated example.[144] SAI is an organisation[145] established by various businesses, trade unions and NGOs.[146] The SAI scheme is centred on its standard, Social Accountability 8000 (‘SA8000’), which is intended to apply worldwide to improve workers’ rights and conditions, particularly in the manufacturing sector.[147] This standard requires firms to implement sophisticated internal compliance systems. These include procedures to evaluate the compliance of suppliers and subcontractors with the substantive criteria in SA8000, and to extend the protection of the standard to homeworkers.[148] Firms cannot avoid their obligations through contracting out their operations.[149]

SAI has established a certification system that involves examination and certification by accredited independent auditors. SAI does not itself conduct certifications of its own standard. Certified factories are identified on the SAI website for the benefit of consumers, investors and other stakeholders.[150] The certification and audit processes also require corrective action to address areas in which a firm does not comply with SA8000. In the event of a system-wide failure to comply with SA8000, firms which fail to take prompt corrective action have been refused certification or have been decertified.[151]

At present, there is insufficient empirical evidence to assess the effectiveness of SA8000. However, it is difficult to see how this approach will have a major impact without additional incentives to secure certification. Only a tiny proportion of firms in the sweatshop industries have been certified.[152] Given the stringent nature of the standard, it is unclear how many more firms will be able to gain certification. For example, SA8000 limits the amount of overtime to 12 hours per week, producing an absolute upper limit of 52 working hours.[153] This is well below the working hours described above in the discussion of Chinese sweatshops. Even though SA8000 allows for a phase-in period, it is hard to see how a firm can comply with this and still remain competitive, unless the firm’s customers (particularly MNEs) are prepared to accept additional costs and delay. Thus, it would seem that until there is a fundamental change in the practice of MNEs, most factories in the sweatshop industries will still be awarded supply chain contracts on the basis of price, quality and efficiency, not certification.[154]

3 The ETI

The ETI is a United Kingdom-based alliance of major corporations, trade unions and NGOs that aims to ensure that the labour conditions of workers producing for the UK market meet or exceed international standards.[155] The ETI has some connection with the British government in that the Department for International Development provided start-up funding and appoints an observer to the ETI Board. Nonetheless, the policy and strategy of the ETI is determined by its member organisations. Although there is no Australian equivalent to the ETI, it provides a useful model for government-supported initiatives.

The ETI shares several elements with SAI. The ETI’s ‘base code’[156] requires ETI members to adhere to a range of standards generally derived from core ILO Conventions. In addition, the obligations of ETI members extend not only to their employees but also to workers in their supply chains, and the obligations cannot be evaded through contracting out.[157] However, rather than a certification system, the ETI adopts an incrementalist, experimental approach to the implementation of codes of conduct, operating as a clearing house and informing all its members on what particular companies have learnt from their attempts to implement the codes.[158] Some of this information is made available to the public. The success of this experimentalist approach is not yet known; the ETI aims to complete a comprehensive impact monitoring study by the end of 2005.

The ETI scheme is clearly less rigorous than the SA8000 accreditation system. It can thus avoid the potentially counterproductive consequences of setting standards at an unrealistic level. On the other hand, the incrementalist approach risks allowing firms to treat international standards as merely aspirational, thereby making only perfunctory improvements to conditions unless the incremental steps are critically assessed.[159]

While they each have flaws, these multi-stakeholder schemes address many of the difficulties with private firm or industry codes. The OECD, SA8000 and ETI codes are relatively comprehensive and reflect international labour standards. Their formulation and implementation involves a range of stakeholder groups and, while this does not include the supply chain workers themselves, the codes do protect collective bargaining, which would give workers a greater voice in their factories.[160] The ETI Base Code, and especially SA8000, enable compliance with the code to be evaluated systematically, and the SA8000 has a credible audit system. It may be that over time these and other multi-stakeholder initiatives produce the most effective voluntary response to sweatshops.

Nevertheless, it must be conceded that in Australia these three schemes, and other multi-stakeholder programs, have until now had little or no impact. Consequently, it would seem that none of the four regulatory approaches considered in this part has more than a marginal effect in encouraging Australian firms to take greater action to address labour problems in their supply chains. This invites an analysis of whether a systematic, state-based regulatory response is needed.


This part examines the possibility of governmental interventions to induce Australian firms to increase their efforts in improving foreign working conditions. The complexity of the connections between Australian firms and foreign workers, especially when mediated through supply chains, creates major obstacles to introducing new regulatory measures. Yet it does not follow that governmental intervention should be abandoned. As the BSL ModStyle example shows, Australian firms that sell products made by workers in developing countries can influence the working conditions of those workers. Firms need to be coaxed into and assisted in doing so. How can government do this?

While governmental intervention often takes the form of legislatively mandated standards coupled with an enforcement mechanism,[161] this does not exhaust the range of regulatory strategies.[162] As there are well-known limitations on the effectiveness of prescriptive or ‘command and control’ regulation, it is appropriate to consider alternatives. Consequently, after examining the feasibility of a prescriptive approach, I look at ‘meta-regulatory’ strategies, including mandating the reporting and coordinating the evaluation of private and multi-stakeholder initiatives.

A Constitutional Issues

Any new legal measures regulating the overseas labour practices of Australian firms must of course be constitutional. While the constitutionality issue cannot be explored in detail here, an initial analysis suggests that well-crafted measures need not be constitutionally problematic. The external affairs power, in particular, confers broad authority on the Commonwealth to enact extraterritorial legislation.[163] The High Court has indicated that the power is not simply confined to the implementation of treaties: ‘If a place, person, matter or thing lies outside the geographical limits of the country, then it is external to it and falls within the meaning of the phrase “external affairs”.’[164] This interpretation of the power would, on its face, permit regulation of the overseas activities of Australian firms in relation to direct employment.[165]

Less certain is whether the external affairs power would support other measures, such as the regulation of the indirect supply chain relationship between Australian firms and foreign workers, CSR initiatives devised in Australia or financial incentives to Australian firms related to CSR measures.[166] However, these measures could probably rely on one or more of the other heads of constitutional power,[167] especially the corporations power,[168] the trade and commerce power[169] and the taxation power.[170] The arbitration power, while historically central to the federal regulation of employment, is probably too limited to support the regulation contemplated here.[171]

B Prescriptive Regulation

When governmental regulation is proposed, many people assume a command and control approach. Such an approach is typified by the Australian Democrats’ Corporate Code of Conduct Bill 2000 (Cth) (‘CCC Bill’).[172] This Bill sets out detailed requirements for corporations in the areas of environmental protection, labour and human rights. Corporations that fail to comply with the requirements would be penalised. The ‘code of conduct’ would thus be compulsory, in contrast to the many voluntary codes.[173] The CCC Bill has been the subject of an inquiry by the Parliamentary Joint Standing Committee on Corporations and Securities (‘PJSCCS’).[174] As the government members of the PJSCCS rejected the Bill, it has no chance of being passed in the current Parliament.[175] In any case, the analysis below makes it plain that the CCC Bill, although created with good intentions, is seriously flawed.

The Bill sets out to impose environmental and social standards on the extraterritorial operations of Australian corporations, and their overseas subsidiaries, that ‘employ or engage the services of’[176] more than 100 people in a foreign country.[177] These corporations are required to report on their compliance with the imposed standards[178] and to establish mechanisms to enforce them.[179] The specified labour standards cover health and safety, forced labour, compulsory labour, child labour (labour of any child under 14 years), the provision of a living wage,[180] unfair dismissal, anti-discrimination, and the rights to organise independent unions, to bargain collectively and to access independent enforcement authorities.[181] Somewhat redundantly, there is also a catch-all reference requiring firms to comply with core ILO Conventions.[182]

Part 4 of the CCC Bill enables civil penalties to be imposed by the Federal Court on an Australian corporation or its executive officers, or both, in relation to a contravention of the prescriptive standards.[183] Further, it enables ‘any person’ (wherever he or she resides) who suffers loss or damage, or who is reasonably likely to suffer loss or damage, as a result of an overseas corporation’s contravention of the Part 2 standards, to bring an action in the Federal Court.[184] As the person who suffers or who is reasonably likely to suffer loss or damage will usually be outside Australia, an action can be maintained on her or his behalf by a corporation or association ‘whose principal objects include protection of the public interest’ (such as an NGO).[185]

The prescriptive approach taken in the CCC Bill has the merit of setting up ‘bright line’ rules based on international standards that are clearly identifiable by firms, stakeholders and the wider public, thereby addressing the lack of consistency and substantive content in many of the private initiatives. It also provides for a clear enforcement mechanism, creating a powerful incentive for firms to implement the standards. Unfortunately, however, with the exception of ‘egregious’ labour abuses, discussed below, this approach is likely to be unworkable.

Even where it has domestic rather than extraterritorial application, command and control regulation may produce rules that are insensitive to context, and that create unintended and sometimes adverse consequences.[186] This problem is likely to be compounded where rules are intended to apply in economic, industrial and cultural circumstances with which regulators and judicial agencies have little or no familiarity and which are undergoing rapid change. Thus, the CCC Bill requires an Australian corporation to pay all its workers a ‘living wage’, defined as a ‘wage sufficient to meet the basic needs of a family of two adults and three children in the country or region they are resident in’.[187] It is not apparent why a worker without a family should be paid a wage determined on this basis. The main problem, though, is that if such a wage were imposed on a limited number of firms (those connected to Australia) in a developing country, there may well be prejudicial side effects. This is especially the case where the wage is imposed without any detailed economic analysis or consultation with local stakeholders. For example, the implementation of a much higher wage than firms had previously been paying might lead to firms dismissing workers or subcontracting work out into the informal economy.

While it would obviously be possible to draft a more appropriate minimum wage provision than that in the CCC Bill, it is unlikely that a standard could be devised which would be capable of application by Australian judges to any country outside Australia.[188] The same applies to many of the other labour standards in the Bill.

One response to this difficulty might be to draw on formulations in the major international codes, such as the OECD Guidelines, the ETI Base Code and SA8000. However, strict compliance with these standards may also be counterproductive, as was suggested in relation to the SA8000 stipulation on working hours. In any event, these codes were not designed to create an absolute and immediate legal obligation, and their use for this purpose would be inappropriate. Thus, even the demanding SA8000 code does not contain a bare prohibition of the use of child labour, but, in those instances where child labour is in fact used by a firm, it requires that firm to take remedial measures such as enabling a child to attend school.[189]

A heavily prescriptive approach can be flawed because of under-inclusion as well as over-inclusion. This is reflected in the CCC Bill. The labour standards in the CCC Bill are drafted to apply to the direct employees of Australian corporations.[190] However, as we have seen, labour abuses are less likely to occur in this context than where corporations indirectly engage foreign labour, for example, because they source products offshore. This means that the Bill fails to assist the most vulnerable workers. Indeed, the limitation is potentially counterproductive in that the Australian corporations might seek to offset any increased direct labour costs resulting from the legislation by pressuring subcontractors to lower prices. Subcontractors might do this by driving their labour costs down, worsening labour conditions. On the other hand, if subcontractors were to be included, it would be difficult to devise a formulation that would make it clear under what circumstances, and to what extent, a corporation will be held responsible for the wide range of labour law transgressions committed in its supply chain.[191] This is particularly so where the supply chain includes many workers in informal sectors.

An overly prescriptive approach is also likely to lead to conflict with foreign legal systems. It risks rendering firms subject to inconsistent legal obligations, namely, those specified in Australian law and those in the foreign jurisdiction in which the corporation operates.[192] Consider the case of Australian firms operating in China. International labour standards require states to respect freedom of association and the right of workers to bargain collectively.[193] However, China has not ratified the relevant ILO Conventions and its domestic law provides that workers’ local workplace organisations must be ‘led’ (lingdao) by higher level unions under the Leninist principle of ‘democratic centralism’.[194] In practice, this means that workers cannot establish unions independent of the communist-controlled ACFTU. If an Australian firm permitted its workers to form an independent union and refused to allow a higher level ACFTU union representative to ‘lead’ that union (that is, to interfere with it), it would be committing an offence under Chinese law.[195] It would be possible to avoid this problem by requiring a firm to adopt a ‘parallel’ substitute which did comply with the law,[196] but it would then be necessary, in the context of a penal provision, to define what that would be.

A further limitation of an overly prescriptive approach is that it risks generating a culture of hostility and resistance from regulated firms, thereby blocking the development of innovative self-regulatory measures, such as we have seen above.[197] Over-reliance on rules may preclude a goal-oriented (rather than a rule-oriented) approach that encourages firms to develop their own internal processes to respond ethically to the contexts in which they operate.[198]

Worse, excessive governmental prescription may interfere with the capacity of stakeholders to participate in the formation of local responses to labour abuses: it mandates obligations without consultation with those stakeholders.[199]

If the difficulties of formulating standards for a prescriptive corporate code of conduct Bill are severe, the enforcement of such standards poses just as great a problem. What interpretational reference point would an Australian court or regulatory authority use to determine whether or not a contravention has occurred? Judges might be inclined to base their interpretation on analogous Australian labour legislation, but that legislation was never designed to apply in the context of a developing country. Assessing what constitutes a safe and healthy working environment in another country based on Australian experience alone is fraught with danger. On the other hand, if the reference point is the country in which the contravention occurred, then the parties will be presented with serious evidentiary problems. Moreover, the court will have to reinterpret the provisions for each nation.

An additional problem with the uncertainty surrounding the meaning of standards is that, until some basis is found for defining their content more clearly, corporations will not know, in many instances, how to comply with them. While this is of course true of any regulatory provision, to some extent, the degree of uncertainty is far greater in this situation than is the case with legislation that is confined to Australia and will be interpreted according to well-established principles of Australian statutory interpretation.

In any case, the practical difficulties confronting an attempt to invoke legal proceedings for contravention of labour standards are, in most situations, overwhelming. As was pointed out in Part III(A) in relation to common law actions, except in the case of class actions concerning major incidents, workers will not have the resources or the connections with Australian lawyers and NGOs to commence and maintain legal proceedings. Moreover, Australia clearly cannot send phalanxes of government officers to foreign jurisdictions in a bid to police the standards offshore.

For these reasons, the prescriptive approach to labour standards taken in the CCC Bill is largely inappropriate. However, there may be a place for carefully targeted prescription. It is useful in this context to distinguish between two categories of unsatisfactory working conditions (albeit with porous boundaries). The first conceptual category, which I will call ‘egregious abuses’, concerns abuses that attract near universal condemnation and that should be immediately prevented. The labour abuses that do not fall into this category I will call ‘poor conditions’. These include two kinds of practices: those which, while in conflict with international standards, are acceptable or even required in particular countries (such as restrictions on independent trade unions in China, or discrimination based on race in Malaysia); and those which are generally seen as undesirable, but in respect to which it is not clear that immediate prohibition is the appropriate response to the problem (such as long working hours or the employment of children in safe and non-coercive conditions,[200] where there is no immediate economic alternative). Note that these categories do not correspond with current international labour standards: some violations of even core Conventions will not constitute ‘egregious’ labour abuses.

I suggest that legislative prohibitions coupled with sanctions would avoid many of the problems encountered in the CCC Bill if they were confined to ‘egregious’ labour abuses. There are three matters, at least, which constitute egregious abuses. The first concerns certain forms of forced labour: namely, slavery and the related condition of sexual servitude. These satisfy the requirements of egregious abuses in that they are universally condemned[201] and their continuation cannot be justified. We have already seen that Australia has extraterritorial legislation explicitly prohibiting slavery and sexual servitude. It may also be appropriate to extend these absolute prohibitions to certain further forms of forced labour that are opposed throughout the international community.[202]

The second is the torture of workers. Like slavery, torture is very widely recognised as internationally intolerable,[203] and, certainly in the context of a work relationship, cannot be justified. As was pointed out above, a person present in Australia can be prosecuted for an act of torture[204] committed outside Australia, but only where there has been state instigation, consent or acquiescence towards the action.[205] This limitation exists in order to adhere to the wording of the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment,[206] thus protecting the legislation from potential constitutional challenge. However, assuming any constitutional hurdles can be otherwise overcome,[207] I see no persuasive argument for confining extraterritorial prosecution of torture to state-related contexts.[208]

The third proposed egregious abuse is the abuse of children, as defined in art 3 of the Convention concerning the Prohibition and Immediate Action for the Elimination of the Worst Forms of Child Labour.[209] This directs states to prohibit specific practices such as trafficking in children, bonded child labour, forced child labour, the use of children for pornography, the use of children in drug trafficking, and the engagement of children in work likely to harm their health, safety, or morals. This Convention has widespread support. It was unanimously adopted by the tripartite International Labour Conference in its 87th session, and has already been ratified by 150 states[210] — the fastest ratification rate in ILO history.[211] Again (and in contrast to controversial cases such as the employment of children in agriculture), no purpose is served by permitting children to continue working under these conditions.

An important question is how all these prohibitions would operate in the context of a supply chain. Where an Australian firm directly enslaves a person, employs bonded child labour or tortures workers, sanctions can be applied in a straightforward manner. However, what of the far more likely scenario of a corporation sourcing products from a factory that engages in such practices? Difficulties arise concerning the knowledge the corporation may have of the egregious abuses — the corporation may be unaware or wilfully ignorant — and the degree of control the corporation may have over the commodity chain. As I have already pointed out, the limited market power Australian firms have in the context of international commodity chains may mean that the firm is unable to force the offending factory to prevent the abuse.

In light of these difficulties, it would not seem appropriate to impose severe criminal sanctions, such as those that currently exist for slavery, on corporations for the actions of their suppliers, where the corporations do not demonstrably control those suppliers. A sanction should nevertheless be imposed on those firms that are aware, or ought reasonably to be aware, of egregious abuses taking place in their commodity chains, and that fail to take measures such as reporting the abuses to relevant local authorities or, if the firms are unable to prevent the abuses, ceasing to trade with the suppliers.[212]

Furthermore, in constructing appropriate sanctions for corporations that are implicated in the commission of egregious labour abuses, emphasis should be placed on compelling those corporations to take remedial measures.[213] Such measures should have at least two components. First, firms need to ensure as far as possible that workers who have been abused have access to both medical care and to decent work in future. Second, firms must revise their own internal processes to ensure that there is no recurrence of abuses.

One objection to extending Australia’s legislation in this way is that it could be viewed as paternalistic and offensive to other nations.[214] However, most extraterritorial legislation impinges on the sovereignty of other states, and, while it is controversial, it is increasingly common.[215] The important issue is not the fact, but the nature, of extraterritorial legislation. Prohibiting Australian corporations from engaging in what are recognised by the overwhelming majority of states as egregious labour abuses is a restrained form of extraterritoriality.[216] Although it may have discernible effects on those foreign labour markets where Australian corporations have significant operations, such regulation does not seek to overturn foreign legislation, but to complement it.[217] It recognises that foreign legal systems may not always be able to enforce their own labour laws against Australian firms (for the reasons considered above).[218]

C ‘Meta-Regulation’

Given that a command and control approach to sweatshop labour abuses may be feasible only in relation to egregious labour abuses, alternative regulatory approaches need to be considered to deal with the category of poor conditions. An array of strategies might be devised with a view to steering the self-regulatory initiatives examined earlier. More specifically, rather than pre-empting, through prescription, the initiatives that firms and stakeholders have been devising to address sweatshop issues, state legislative and administrative resources could be deployed to promote and facilitate their more effective formulation, implementation and evaluation.

This suggested reorientation of government intervention away from the proposals focused on command and control is prompted by the literature on ‘reflexive’,[219] ‘responsive’[220] and ‘meta-’[221] regulation and ‘democratic experimentalism’.[222] This literature, while often recognising an important continuing role for some forms of command and control regulation, encourages a shift in state action away from direct regulation towards overseeing a decentralised reconfiguration of governance and management systems directed at addressing problems of environmental protection, occupational health and safety and other areas of social concern.[223] However, an important difficulty with drawing on this literature is that it is generally directed at addressing regulatory challenges within the national boundaries of a developed country. In developed countries, the reconfiguration can be constructed upon extensive experience in the formulation of environmental and safety standards and systems, and in the context of comparatively strong civil societies in which there is space for stakeholder concerns to be articulated and heard (especially via trade unions and NGOs).

In cases of corporate responsibility for foreign workers, on the other hand, there is little information about the nature of relationships between Australian firms and the foreign workers (especially where the relationships are mediated through supply chains). Nor is there knowledge about the kinds of standards and systems that Australian firms could reasonably be expected to introduce. Moreover, while the literature on new forms of regulation, as well as the specific empirical evidence on private sector initiatives,[224] suggests that stakeholder involvement must be a central objective, implementing this is problematic when the crucial stakeholders are often unorganised and disempowered, speak a different language and live under authoritarian regimes.

D The Ratcheting Labour Standards Proposal

Despite these difficulties, scholars have conceived meta-regulatory strategies to tackle sweatshop issues. One of the most prominent of these has been developed by three Americans, Charles Sabel, Dara O’Rourke and Archon Fung. They have proposed a scheme that they call ‘ratcheting labour standards’ (‘RLS’).[225] Rejecting both command and control and voluntarist approaches, they argue that labour standards can be improved (‘ratcheted up’) if firms are permitted to develop their own CSR codes and monitoring systems on the condition that they make their results transparent and subject to public evaluation, in particular, through independent, accredited monitoring organisations.[226] These monitoring organisations would systematically compare and rank the performance of firms. This process, they argue, would create an incentive structure in which firms would compete with each other to be market leaders in social responsibility, thereby generating continuous improvement in working conditions.

The core idea of ratcheting labor regulation ... is to compel firms to compete publicly with one another to improve their social performance. In its fullest version, every firm regulated under this regime would report wages, workforce profiles, environmental and labor management systems, and other elements of social performance to a certified monitor. Each monitor then unifies reports and ranks the overall performance of firms under its purview. Monitors would then make these rankings and the methods used to derive them publicly available. The reputation and credibility of monitors would be built upon public evaluations of their capacities for evaluating and improving their member firms. The information provided by this system allows the public acting as consumers, in interest groups, and through their national governments to put firms and monitors under complementary competitive pressures. Firms thus seek highly regarded monitors to elevate their public standing. Monitors would seek out the most sincere and capable firms and encourage their member firms to adopt best practices in order [to] build their reputations and expand their influence. Consumers and others would be able to distinguish leaders from laggards in social performance. Under this system, responsible firms could assume that their behavior would be rewarded and irresponsible ones would fear embarrassment, pressure campaigns, and official sanctions.[227]

The RLS framework responds to three key deficiencies in voluntary measures: the refusal of some firms to produce their own transparent and comprehensive processes for addressing sweatshop issues, the lack of credible external auditing and the absence of mechanisms to compare firm performance.

The RLS scheme also has several advantages over ‘command and control’ regulation.[228] RLS can generate more context-sensitive responses to labour abuses. Firms committed to improving working conditions will be able to use their knowledge of local economic, political and industrial conditions to devise specific, incremental, feasible remedies, rather than being forced to comply with legislatively mandated requirements, which may not be appropriate or practicable. Moreover, RLS engages firms’ own compliance and performance management systems to develop and assess standards, rather than relying on scant state enforcement resources:

a rolling rule regime makes compliance costs bearable by building upon the capacities of reorganization and learning that firms have already developed in the course of competition. It begins by asking them to extend their ingenuity, honed for the purposes of improving product quality and variety, to the problems of labor conditions and social performance generally. ... Because it builds upon these familiar disciplines and demands demonstrably feasible levels of performance, compliance with a ratcheting regime is manifestly less costly than adherence to abstract and uniform standards formulated by distant agencies.[229]

Furthermore, rather than threatening to generate a hostile response by firms, as may occur with excessive prescription, the RLS scheme creates a space for those managers with a strong ethical commitment to decent work practices to express that commitment creatively.[230] This encourages continuous improvement rather than compliance with static minima.

Finally, and most significantly, RLS is arguably a more effective means of reaching into supply chains than prescriptive regulation. Firms themselves assess when they are able to influence suppliers effectively, but this assessment will be based on the experience of what other firms have been able to achieve. This avoids difficult legal arguments about what constitutes control in a particular instance.[231]

There are, nonetheless, several areas of doubt about the potential of the RLS proposal. First, although the RLS authors emphasise local participation, as Jill Murray and Alexandra Owens have pointed out, they do not make clear how the voices of local workers can be heard.[232] As I have stated, if local workers are not involved in devising, monitoring and evaluating measures to reduce labour abuses, such measures may be unresponsive to worker needs, disempowering and patronising. This suggests that the involvement or creation of local representative bodies, such as independent trade unions (or, where these are prohibited, parallel organisations), and, in the informal sector, worker-oriented NGOs, needs to be better integrated into the scheme.

Second, although the RLS authors insist that RLS is not a voluntary system, as firms are required to disclose information about labour standards,[233] they do not elaborate on how this requirement should be framed. This elaboration is essential, since without an appropriate mandatory disclosure regime, RLS is unlikely to operate effectively. Social and market-based incentives for greater transparency are not enough, especially given Australian experience. Even if increasing consumer, investor and NGO pressure leads some corporations to greater disclosure, it is difficult to see how many firms, especially those in industries away from the public spotlight, have sufficient incentives to reveal what they are (not) doing. Australian firms outside the mining sector have little experience in social reporting.[234] This inexperience, in addition to the considerable costs of data collection and analysis, may lead to perfunctory reporting. Even if they do report in detail, in the absence of clearly established methodologies and content guidelines, the publication by firms of statements about their policies towards offshore workers may lead to a profusion of documentation which cannot be readily compared.

A related problem, as Ayres points out[235] and the RLS authors recognise,[236] is that unless all firms report adequately, a firm that makes a genuine attempt to be transparent about its efforts to remedy labour abuses in its supply chain may, perversely, be punished by media emphasis on the abuses rather than the remedy. Ayres comments: ‘If an employer who hires poor people is going to be vilified as running a sweatshop regardless of what she does, then what incentive does she have to do the right thing?’[237]

A third, compounding difficulty concerns the evaluation of disclosed information.[238] Australia does not yet have a well-functioning ‘market’ in monitoring and ranking firm performance on offshore labour initiatives and, in any case, the work of the few organisations that rank firms on a general CSR basis has been quite controversial.[239]

E Implementing a Meta-Regulatory Strategy in Australia

What could the Australian government do to prompt the virtuous cycle that RLS envisages? In the absence of regulatory experimentation and evaluation, any suggestions must be speculative and open to revision. Nonetheless, it may be feasible to design a scheme for government intervention with the objectives of:

This scheme would directly address the second and, to some extent, the first of the difficulties associated with implementing the RLS scheme. While it does not directly tackle the rudimentary state of labour monitoring and ranking in Australia, the desire of some disclosing firms to verify their initiatives (and thereby gain greater public credibility) may create more opportunities for businesses and not-for-profit organisations to participate in evaluative processes.[240]

1 Compulsory Aspect: Reporting Requirements

The extent to which any government intervention should involve imposing mandatory requirements on firms cannot be easily determined until there is greater information about what firms can be realistically expected to do. Initially, emphasis might be best placed on facilitative measures. Nonetheless, in order to generate this information, realisation of the first and second objectives should involve an element of compulsion. This might be in the form of a statutory requirement on a firm to assess its position, and then report,[241] both publicly and to an appropriate governmental authority:[242]

  1. (a) whether it engages workers overseas; and
    (b) if so, whether it has implemented policies to address labour issues
    concerning those workers; and
  2. (a) whether a significant part of the firm’s operations involves the supply of goods produced or services supplied offshore; and

This reporting obligation should be clarified in guidelines issued by the relevant authority, which should be regularly revised after assessments are made of how firms comply with the statutory requirement.[244] The guidelines would require careful drafting and updating. If the reporting obligation is too loose, it will reproduce the problems associated with voluntary codes. Firms will be tempted to produce florid statements full of terms such as ‘best practice’, ‘sustainable development’ and ‘continuous improvement’ that are so vague that they defy meaningful comparison with other endeavours.[245] On the other hand, if the reporting is too prescriptive,[246] the problems associated with command and control are likely to re-emerge. A middle path may be found if the guidelines are directed to eliciting specific information about the nature of overseas-related operations (such as mapping the supply chain) and the firm’s labour-related policies. Relevant to the latter would be the methodology adopted in formulating and implementing the policies, the extent of any stakeholder participation in those processes and the nature of any internal or external evaluation or verification system.

The authority would not mandate reporting against any specific labour standards, engage in evaluating or comparing firms or, at least initially, accredit external monitors. This is because there is not at present any locally or internationally agreed framework for social reporting. While one, such as the GRI, may emerge over time, this and other schemes are still at an experimental stage. The reporting legislation should aim to accelerate these experiments by increasing participation in them, and by facilitating coherence (through the measures outlined in the next section), rather than pre-empting the results. Evaluation should be left to private and not-for-profit organisations, at least until definitive standards emerge or the market for evaluation and accreditation fails.

However, the authority should be able to impose sanctions on firms that either fail to report in accordance with the guidelines or engage in false, misleading or deceptive conduct in relation to a report.[247] These should include an ability to order a firm to repeat its report. Penalising token reporting (technical, rather than illuminating, compliance) may be difficult in the absence of a convergence in agreed reporting principles and standards. Nonetheless, experience with the environmental reporting requirement in the Corporations Act 2001 (Cth)[248] suggests that even minimalist mandatory reporting can have some effect in improving the quality of information made available to the public.[249] In any case, if the facilitative measures discussed below lead to greater harmonisation, token reporting will become easier to identify and target.

There are two qualifications that may need to be made to a general public reporting requirement. One concerns whether all reports should be made public. In some cases, consumers and investors may not assess the information correctly.[250] As has been previously pointed out, firms may be punished, rather than rewarded, for revealing the extent of labour abuses within their supply chains. In this context, public disclosure might constitute a disincentive to investigate supply chains closely.[251] Consideration should therefore be given to assessing whether a limited form of confidentiality is appropriate.[252] For example, firms could be permitted to restrict disclosure of sensitive parts of their reports to the authority, provided that they could demonstrate that the restriction promoted rather than undermined the objectives of the government scheme.[253]

A second issue is whether the reporting obligation should extend to all firms. This would, in principle, be desirable as reports from innovative small enterprises, such as BSL’s ModStyle, could assist in establishing goals for other firms with limited resources and market power. However, it is the very problem of limited resources that suggests that smaller firms should not be required to meet the same standards as large enterprises equipped with elaborate and well-staffed internal planning systems. Firms are already struggling under a profusion of ‘triple bottom line’ surveys.[254] Accordingly, the reporting requirements could replicate, at least initially, the provisions in the Corporations Act 2001 (Cth) that exempt small proprietary companies from the obligation of preparing directors’ reports in most circumstances.[255]

2 Facilitative Aspect: Encouraging Convergence on CSR Principles

A reporting requirement must be complemented by governmental efforts to assist firms to adopt successful CSR initiatives and to produce greater coherence and consistency in reporting, monitoring and evaluation. Some government departments are already facilitating CSR initiatives in certain areas,[256] but there is no agency that coordinates initiatives involving overseas labour conditions.

It may therefore be useful to entrust the facilitation role to a new agency.[257] Such an agency might be similar to the Office of International Labor Affairs in the US Department of State, which, among its various responsibilities, ‘[f]unds and assists in the development of programs to eliminate abusive sweatshop labor conditions in foreign factories that produce consumer goods for the American market’ and ‘[e]ncourages corporate responsibility with regards to worker rights’.[258] This agency would liaise closely with other government bodies involved with CSR initiatives, but would maintain its distinct focus. It would also furnish advice to, and liaise with, the authority responsible for mandated reporting.

The facilitative measures should involve at least three elements.[259] First, given the comparatively low level of knowledge in Australia about CSR measures to assist foreign workers, exchange of experience and good practice should be coordinated. The pooling of information (from both Australian and overseas firms) and the publicising of success stories are two keys to this. They will identify feasible measures that firms can take to begin to address complex issues, such as working conditions in supply chains. The approach taken by the UK government-backed ETI, which is focused on incremental learning, offers a workable model for this first element.

Second, the government should collaborate with businesses, NGOs and trade unions to develop common approaches, principles and metrics for establishing codes, reporting and evaluation. The European Union provides a useful illustration of what might be put into effect.[260] In 2002, the European Commission adopted a strategy that, while preserving the voluntarist character of CSR initiatives, created an active role for the Commission in promoting transparency, comparability and more extensive stakeholder involvement in those initiatives.[261] A key plank in this strategy is a ‘multi-stakeholder forum on CSR’.[262] The forum is coordinated by the Commission, and includes business, labour and NGO representatives. It is seeking greater convergence in CSR measures, based, in labour-related matters, on the ILO Conventions and the OECD Guidelines.[263]

Third, the agency should liaise with relevant governmental bodies in developing countries to elicit information about the impact of the Australian initiatives. The agency should also invite Australian stakeholders to seek independent feedback from NGOs and (where possible) autonomous trade unions in those countries.

In conducting its facilitative work, the agency needs to consider the distinct circumstances of small and medium-sized enterprises. Although they are not obliged to report, many may wish to participate in CSR initiatives. Again, the European Commission provides a useful model. Part of the work of its forum is to develop tools (such as mentoring by larger firms) to assist smaller firms in implementing CSR initiatives, including supply chain initiatives.[264]

One potential obstacle to facilitating CSR initiatives, and convergence in particular, is the Australian federal government’s controversial attitude towards what the major multi-stakeholder initiatives deem to be a core CSR principle. The OECD Guidelines,[265] SA8000[266] and the ETI Base Code[267] (as well as the Global Compact)[268] all uphold the right of employees to bargain collectively, in accordance with the ILO Right to Organise and Collective Bargaining Convention.[269] Australia is a party to that Convention and has also endorsed its expression in the OECD Guidelines. However, according to the ILO, the federal government’s approach to collective bargaining, as reflected in the Workplace Relations Act 1996 (Cth), is inconsistent with that Convention, as it gives primacy to individual contracting rather than collective bargaining.[270] Thus, the Workplace Relations Act 1996 (Cth) has been interpreted to enable an employer, in certain circumstances, to refuse to negotiate collectively with employees.[271] The divergence between the key multi-stakeholder initiatives and the Australian legislative framework is exacerbated by the active promotion of individualised employment relations by the federal government and many corporations.[272] Consequently, it will be difficult for the government to reach agreement with unions and NGOs on common principles and metrics. While this does not prevent government facilitation of exchange of experience, it is likely to inhibit stakeholder commitment to, and endorsement of, the facilitative process.

3 Incentives

The proposals for government action that this article has put forward involve the application of sanctions only for egregious labour abuses and for failure to observe a fairly loose reporting requirement. This may not be enough to prompt many firms to give serious consideration to addressing poor working conditions offshore. This is so even in combination with the type of market and civil society-based penalties and rewards contemplated in the RLS scheme. The embryonic state of evaluation and ranking mechanisms in Australia mean that the effectiveness of those penalties and rewards is likely to be quite limited.

Accordingly, it may be necessary to provide additional incentives to firms to participate actively in the search for effective initiatives.[273] Such positive and negative incentives might include preferential status for government procurement,[274] provision of ‘start-up’ subsidies to assist firms in implementing relevant CSR strategies, listing of poor performers and rewards to good performers (such as public governmental recognition).

However, appropriate incentives are hard to structure. In the absence of sophisticated agency information systems, they may be ineffective in dealing with firms uninterested in improving the welfare of overseas workers. These firms may manufacture the paperwork necessary to obtain a benefit or avoid a detriment, but have no real commitment to changing their procedures.[275] Moreover, determining entitlements to incentives can become a contentious matter given that no agreed metrics for assessing performance have been determined.[276] For instance, if one firm takes steps to bolster freedom of association in its offshore suppliers while another focuses on occupational health and safety, how is it to be determined which is more meritorious? Is the fact that a firm is SA8000-accredited relevant to an incentive scheme? If so, what weighting should the accreditation be given?

There are two forms of incentives that may nevertheless be practicable. First, competitive start-up subsidies could be granted to firms (individually or jointly) and to multi-stakeholder initiatives, to implement specific and well-constructed plans for improving the conditions of foreign workers connected to the Australian market. The plans should include credible self-evaluation and external evaluation systems. Second, a ‘negative screening’ system might be introduced, rendering firms that fail to comply with the reporting requirement, or that have been sanctioned for egregious labour abuses, ineligible for certain government benefits such as export credits.[277] These, like the other measures proposed here, would need to be worked out in detail and put into practice in an experimental and revisable manner.


More and more goods consumed by Australians are manufactured by workers in developing countries. Most Australians, though, never have any personal contact with these workers or see the factories in which they work. Australian firms, even those that import the products the workers make, usually have no direct connection with them. Many Australians are aware that some of these workers are often engaged in sweatshop conditions. However, it is tempting to believe that there is little or nothing that we can do to remedy the situation. The sweatshops derive from factors such as widespread rural poverty and underemployment, weak legal systems and authoritarian governments over which we have no control. If sweatshops are also the responsibility of international corporations, then perhaps the blame should be laid at the feet of major Northern Hemisphere brand-name companies, whose market power enables them to dictate terms to the firms frantically competing for business lower down the supply chain. Australian firms do not enjoy that kind of dominance. Even if they did, attempts to improve conditions from afar might actually make matters worse. Perhaps, we might conclude, the best that can be done is to contribute to a gradual increase in global prosperity through greater trade.

This article has argued against resigning ourselves to inaction. Increasingly sophisticated initiatives around the world, including some in this country, are revealing ways in which enterprises can adopt practical, constructive policies that assist foreign workers. In general, Australian firms have shown comparatively little awareness of, or interest in, these initiatives. There is little in our present regulatory framework to encourage them to do so. Yet government can have a pivotal role in inducing change. This is not through legislating a raft of broad prohibitions and obligations that can be neither properly defined nor enforced, and that would risk yielding little more than grudging compliance (although there is a place for certain tightly drafted sanctions). Rather, it is through prompting and assisting firms to assess what they might feasibly do, and, where there is potential for constructive intervention, to take action. I have suggested that this might be achieved through measures that require firms to report on their self-assessments, facilitate institutional learning through pooling information about what has worked and what has failed, and assist firms, unions and NGOs to seek greater agreement on principles for formulating initiatives and on metrics for evaluating them. Other governments, particularly in Europe, are already putting some of these measures in place.

I reiterate that, given the complexity of the sweatshop issue, the proposals here are tentative. They need to be revised and reworked in the light of further information, thought and experience. Yet their indefinite nature should not lead firms — and even more so, government — to revert to the belief that the status quo is the best that can be achieved. This would foreclose the real possibility of making working life more bearable for the people who make the tokens of our affluence.

[*] BA (Hons), LLB, LLM (Melb), LLM (Columbia); Senior Lecturer, Faculty of Law, The University of Melbourne. I would like to thank John Braithwaite and the scholars at RegNet at the Australian National University, Anita Chan, Richard Garnett, Serena Lillywhite, Richard Mitchell, Ian Ramsay, the anonymous referees and in particular Jill Murray for their comments on, and generous assistance with, this piece. The errors are, of course, mine. This research was partially funded by an Australian Research Council large grant.

[1] The issues discussed in this article connect with wider regulatory debates about the human rights obligations of corporations and corporate social responsibility: see generally Stephen Bottomley and David Kinley, Commercial Law and Human Rights (2002); Christine Parker, The Open Corporation: Effective Self-Regulation and Democracy (2002); Bryan Horrigan, ‘Fault Lines in the Intersection between Corporate Governance and Social Responsibility’ [2002] UNSWLawJl 31; (2002) 25 University of New South Wales Law Journal 515. The focus of this article remains, however, quite specific.

[2] The insight in this section is Jill Murray’s.

[3] By far the largest recipient of Australian investment abroad in 2001–02 was the United States, which accounted for almost half of all direct investment in that period. More than 80 per cent of direct investment was in Organisation for Economic Cooperation and Development (‘OECD’) countries: Australian Bureau of Statistics, International Investment Position, Australia: Supplementary Country Statistics 2001–2002, ABS Catalogue No 5352.0 (2003) 3–4.

[4] Of course, developed countries do breach labour rights and most have informal sectors which fall outside the scope of national rules on social protection and labour regulation.

[5] In any case, there is evidence that where firms from developed economies do directly employ workers in developing countries, such as China, conditions in those firms are superior to those in domestic firms: Kit-Chun Lam, ‘A Study of the Ethical Performance of Foreign-Investment Enterprises in the China Labor Market’ (2002) 37 Journal of Business Ethics 349.

[6] Consider, for example, the role of WMC at the Tampakan mine in the Philippines and the role of (then) BHP at the Ok Tedi mine in Papua New Guinea.

[7] The Australian Mines and Metals Association estimates that labour costs as a proportion of total costs in mining were between 20 and 30 per cent in 2000: Jill Murray, Interview with Ian Masson, Victorian and Special Services Manager, Australian Mines and Metals Association (Melbourne, 20 October 2000).

[8] One form of labour abuse for which there is evidence is sex discrimination and harassment: see generally Ingrid Macdonald and Claire Rowland (eds), Oxfam Community Aid Abroad, Tunnel Vision: Women, Mining and Communities (2002).

[9] See, eg, Miguel Korzeniewicz, ‘Commodity Chains and Marketing Strategies: Nike and the Global Athletic Footwear Industry’ in Gary Gereffi and Miguel Korzeniewicz (eds), Commodity Chains and Global Capitalism (1994) 247.

[10] On the impact of flexible production on labour law generally, see Patrick Macklem, ‘Labour Law beyond Borders’ (2002) 5 Journal of International Economic Law 605, 60914. For an extensive analysis of the position in the United States, see Mark Barenberg, ‘Democracy and Domination in the Law of Workplace Cooperation: From Bureaucratic to Flexible Production’ (1994) 94 Columbia Law Review 753, 881–90.

[11] Gary Gereffi, ‘The Organization of Buyer-Driven Global Commodity Chains: How US Retailers Shape Overseas Production Networks’ in Gary Gereffi and Miguel Korzeniewicz (eds), Commodity Chains and Global Capitalism (1994) 93, 97. Gereffi contrasts buyer-driven production chains with capital and technology-intensive ‘producer-driven production chains’: at 96–100. In these chains, which are characteristic of the vehicle and aircraft manufacturing industries, the MNE exercises much greater direct control over the production process: at 97. See generally David Held et al, Global Transformations: Politics, Economics and Culture (1999) 236–82.

[12] Gereffi, above n 11, 99.

[13] Ibid 104.

[14] See generally Ivanka Mamic, Business and Code of Conduct Implementation: How Firms Use Management Systems for Social Performance (2003) 49. While Mamic focuses on chains operating offshore, these are also present in Australia: see below nn 47, 193. Note also that, while this article focuses on manufacturing industries, poor working conditions also occur in the service sector: see Alexandra Owens, ‘Testing the Ratcheting Labor Standards Proposal: Indonesia and the Shangri-La Workers’ [2004] MelbJlIntLaw 7; (2004) 5 Melbourne Journal of International Law 169.

[15] The following discussion derives from a much more extensive analysis in Sean Cooney and Richard Mitchell, ‘What Is Labour Law Doing in East Asia?’ in Sean Cooney et al (eds), Law and Labour Market Regulation in East Asia (2002) 246.

[16] The extent to which this is the case naturally varies within and between industries depending on the product to be produced. Where the production process requires a greater degree of skill, for example because of the need to operate sophisticated or new technology, greater investment in the workforce may be required.

[17] For illustrations from other Asian countries, see the studies collected in Sean Cooney et al (eds), Law and Labour Market Regulation in East Asia (2002); Stephen Frost, Omana George and Ed Shepherd (eds), Workers’ Rights for the New Century (2003).

[18] Kai-Ming Liu, Bianyuan Ren (2003) 5–7 [Migrant Labour in South China].

[19] See, eg, the account of the electroplating industry in China in Stephen Frost, Labour Standards in China: The Business and Investment Challenge (2002) 1819.

[20] I expect it to be uncontroversial to describe these practices as abuses. This article does not examine low wages as an example of abuse. While it is often argued that workers should be accorded ‘living wages’, whether a pay rate is too low is a matter for complex economic debate.

[21] Chinese statistics on occupational health and safety are inaccurate and are inadequately systematised: see generally Asian-Pacific Regional Network on Occupational Safety and Health Information, International Labour Organization, China: Promoting Safety and Health in Township and Village Enterprises (1998). In particular, deaths, injuries and occupational diseases may be very significantly under-reported. According to the State Administration of Work Safety, there were 4952 non-mining industrial fatalities in 2001: 2001 Nian Quanguo Shangwang Shigu Qingkuang Fenxi (2002) Guo Jia An Quan Sheng Chan Jian Du Guan Li Jü [State Administration on Work Safety] <> [An Analysis of the National Casualty Incidents 2001] (Chinese only). This is startling when compared to the 5745 non-mining fatalities over the same period in the United States, a developed country with a much smaller industrial workforce. The State Administration on Work Safety statistics nevertheless indicate that these non-mining industrial accidents have been on the increase: there was a 19 per cent rise in non-mining industrial accidents in the first nine months of 2003, to 5203: see British Broadcasting Corporation, ‘Workplace Deaths Rise in China’, BBC News, 23 October 2003 <> . For non-governmental accounts of workplace injuries, see below n 23.

[22] For example, in the clothing industry in Guangdong Province, because of the high intensity in client orders during the ‘peak’ season, the (migrant women) workers are, in a large number of enterprises, required to work for several months without a day’s rest. Overtime reaches 150–200 hours per month, greatly in excess of the legal limit. These working hours are obviously injurious to health: Kai-Ming Liu and Shen Tan, Kuaguo Gongsi De Shehui Zeren Yu Zhongguo Shehui (2003) 85 [Corporate Social Responsibility in China]; Minghua Zhao and Jackie West, ‘Adjusting to Urban Capital: Rural Female Labour in State Cotton Mills’ in Dong-Sook Gills and Nicola Piper (eds), Women and Work in Globalising Asia (2002) 169, 175–9; Anita Chan, ‘Labor Standards and Human Rights: The Case of Chinese Workers under Market Socialism’ (1998) 20 Human Rights Quarterly 886. It is sometimes claimed that workers themselves wish to work very long hours: for a critical evaluation of this claim, see Frost, above n 19, 30. Short bursts of long working hours, followed by adequate rest, may not be injurious. However, the evidence suggests that long working days may continue for successive weeks or even months without rest. Note that in the ‘off-season’, on the other hand, workers are terminated in large numbers.

[23] All of these conditions have been well-documented as a result of the extensive empirical work conducted by the Chinese NGO, the Institute for Contemporary Observation (ICO), based in Shenzhen: see Liu and Tan, above n 22, 22. For a powerful English-language collection of stories of labour abuses, drawn from sources including the All China Federation of Trade Unions and the All China Women’s Federation, see Anita Chan, China’s Workers under Assault: The Exploitation of Labor in a Globalizing Economy (2001). See also Frost, above n 19, 16–36.

[24] See, eg, Ying Zhu, ‘Economic Reform and Labour Market Regulation in China’ in Sean Cooney et al (eds), Law and Labour Market Regulation in East Asia (2002) 157.

[25] See generally Ann Kent, China, the United Nations and Human Rights: The Limits of Compliance (1999).

[26] Labour Law 1994 (People’s Republic of China) arts 12–13. These articles prohibit gender discrimination, but not discrimination based on regional origin. However, art 12 prohibits discrimination based on the employee’s ethnic community.

[27] Article 20 of the Trade Union Law 1992 (as amended in 2001) (People’s Republic of China) contains only vague references to collective contracts negotiated by trade unions on behalf of workers and staff members. An important step towards improving the regulatory framework for collective contracts was taken with the promulgation by the Ministry of Labor and Social Security of the Provisions on Collective Contracts (passed 30 December 2003, effective 1 May 2004).

[28] See below n 153.

[29] For a comprehensive and nuanced analysis of the state of the Chinese legal system, see generally Randall Peerenboom, China’s Long March toward Rule of Law (2002).

[30] Liu and Tan, above n 22, 76–9; Chan, China’s Workers under Assault, above n 23, 123–5.

[31] See below n 194 and accompanying text.

[32] See generally Peerenboom, China’s Long March toward Rule of Law, above n 29, 280–342.

[33] Liu and Tan, above n 22, at 83–5.

[34] The effects of the hukou residency system are examined extensively in Liu, above n 18; on its effect on labour conditions, see esp at 139–67, 209–44. See also Frost, above n 19, 1921. The residency permit system is being slowly reformed: see ‘Reforms Make Life and Travel Much Easier’, China Daily (Beijing), 8 August 2003 <> .

[35] On the conditions facing Asian migrant women workers, see generally Susan Horton (ed), Women and Industrialization in Asia (1996); Zhao and West, above n 22, 175–9.

[36] While the coastal cities benefit from the labour of these workers, the associated costs (such as payments for sickness, injury or unemployment) are borne by their rural communities, which have far fewer resources to meet them. Thus, when the workers outlive their usefulness by reaching their thirties or by suffering injury, they may have little alternative but to return to their local communities, where they often must also take responsibility for children and elderly people: Liu, above n 18, 12–13.

[37] See, eg, Anita Chan and Hong-Zen Wang, ‘Raising Labor Standards, Corporate Social Responsibility and Missing Links — Vietnam and China Compared’ (Paper presented at The Labor of Reform: Employment, Workers’ Rights, and Labor Law in China, University of Michigan, 21–22 March 2003) <> . The authors comment that these practices are much less tolerated in Vietnam than in China.

[38] Liu, above n 18, 222–44. Resentment is likely to be compounded by the fact that some low-skilled local workers have been replaced by much cheaper migrant labour: at 5.

[39] See, eg, Jagdish Bhagwati, ‘Break the Link between Trade and Labour’, Financial Times (London), 29 August 2001, 13. Bhagwati argues that trade should not be linked to labour standards. For a proposal for imposing labour-related trade sanctions that may be WTO compliant, see Macklem, above n 10, 63845.

[40] Nicholas Kristof and Sheryl WuDunn, ‘Two Cheers for Sweatshops’, New York Times (New York), 24 September 2000, 70. The growth history of East Asian economies is documented in World Bank, The East Asian Miracle: Economic Growth and Public Policy (1993). This was written prior to the East Asian economic crisis, which saw worsening conditions in countries such as Indonesia. The description of the transition from low to high wages in Taiwan and South Korea, however, remains essentially correct.

[41] There are major political, legal, demographic and educational distinctions between present-day China and the NICs at a similar stage of development. Accordingly, it may be doubted whether the Chinese workforce will, as a whole, enjoy an equivalent improvement in living standards over a similar time-scale.

[42] For an analysis of three different modes of legal control (ownership, authority and contract), and of mechanisms to impose liability in the context of each mode, see Hugh Collins, ‘Ascription of Legal Responsibility to Groups in Complex Patterns of Economic Integration’ (1990) 53 Modern Law Review 731. See also Steven Ratner, ‘Corporations and Human Rights: A Theory of Legal Responsibility’ [2001] YaleLawJl 32; (2001) 111 Yale Law Journal 443, 518–22.

[43] Business and Industry Advisory Committee to the OECD, ‘Supply Chains and the OECD Guidelines for Multinational Enterprises’ (Paper presented at the OECD Roundtable on Corporate Responsibility: Supply Chains and the OECD Guidelines for Multinational Enterprises, Paris, 19 June 2002) 4.

[44] Multinational firms are continually improving supply chain performance management, including monitoring of product quality and design, through the use of sophisticated information technology and organisational systems: see, eg, Stanford Global Supply Chain Management Forum <> . See also Charles Sabel, Dara O’Rourke and Archon Fung, ‘Ratcheting Labor Standards: Regulation for Continuous Improvement in the Global Workplace’ (Social Protection Discussion Paper No 11, Social Protection Unit, World Bank, 2000) 9–11.

[45] Held et al note that, at an international level, systematic quantitative data on global production networks is unavailable (although certain studies of specific industries have been conducted): Held et al, above n 11, 237. This appears to be true of Australia.

[46] Australian Bureau of Statistics, Year Book Australia 2003, ABS Catalogue No 1301.0 (2003) 919 (tables 30.27 (‘Merchandise Exports by Country’) and 30.28 (‘Merchandise Imports by Country’)). Data is classified according to the United Nations Standard International Trade Classification, Revision 3. In the financial year 2001–02, these imports totalled more than A$5 billion. See also the data on Chinese imports in Australia in Kirsty Needham, ‘The Price Is Right’, Sydney Morning Herald (Sydney), 18 October 2003, 41. This article cites an ABN-Amro report in 2003 indicating that 68 per cent of all imported clothing and footwear sold in Australia is made in China, as well as half of imported toys and sporting goods, a third of imported furniture, 61 per cent of travel goods, 23 per cent of household equipment and 15 per cent of computers.

[47] The establishment of retailer and manufacturer ethical codes, monitored by an effective industry union, supported by a range of NGOs, and backed up by state government legislation, represents a very significant response to the phenomenon of Australian sweatshops. The Textile, Clothing and Footwear Union of Australia (‘TCFUA’) has entered into agreements protecting homeworkers with major retailers and manufacturers: see the National Retailers/TCFUA Ethical Clothing Code of Practice (2002) Australian Retailers Association <> . There is state legislation providing, inter alia, for the implementation of mandatory codes if these voluntary codes prove unsuccessful: see Industrial Relations (Ethical Clothing Trades) Act 2001 (NSW) pt 3; Outworkers (Improved Protection) Act 2003 (Vic) pt 3 div 2. These mandatory codes include independent monitoring procedures by the Ethical Clothing Trades Council: see Industrial Relations (Ethical Clothing Trades) Act 2001 (NSW) ss 9, 12; Outworkers (Improved Protection) Act 2003 (Vic) ss 20, 22. These arrangements do not apply to workers outside Australia as state governments lack jurisdiction and the TCFUA is unable to monitor firms outside Australia.

[48] Serena Lillywhite, ‘Pursuing Corporate Responsibility in China — Experiences of a Small Enterprise in the Optical Industry’ (Paper presented at the OECD Roundtable on Corporate Responsibility: Supply Chains and the OECD Guidelines for Multinational Enterprises, Paris, 19 June 2002) <> .

[49] The larger, newer firms may provide a reasonably safe working environment (indeed sometimes superior to equivalent Australian firms) and higher pay, but they require extremely high amounts of overtime work. Employees at other firms work in a more dangerous environment, but with reduced working hours. Overall, though, the firms are characterised by harsh disciplinary regimes, involving numerous fines and penalties, little employee participation in management, ineffective grievance procedures, and regimented dormitories, where employees’ lives outside working hours are closely regulated. Moreover, while safety equipment is often provided, workers generally do not receive systematic training in safe working procedures. For example, I have noticed on factory visits that safety instructions for equipment are often printed solely in English. See generally ibid.

[50] Serena Lillywhite, who has regularly visited the firms as part of an ethical business project, describes the workers in the firms in her OECD report (ibid 3):

The majority (80 per cent) of workers at factories visited are young women aged 17–25 years who have migrated from surrounding rural provinces to work in the burgeoning industrial districts. Generally speaking, they appear to have a relatively low level of education. Most are employed in repetitive, low-skilled jobs with only about 10 per cent of the production process requiring technical expertise. Staff turnover is high, with most workers staying with a factory for one to two years.

[51] See below nn 110–11 and accompanying text.

[52] For an examination of the position in the United States and the United Kingdom, see generally Beth Stephens, ‘Corporate Accountability: International Human Rights Litigation against Corporations in US Courts’; Michael Byers, ‘English Courts and Serious Human Rights Violations Aboard: A Preliminary Assessment’; Richard Meeran, ‘Liability of Multinational Corporations: A Critical Stage in the UK’ in Menno Kamminga and Saman Zia-Zarifi (eds), Liability of Multinational Corporations under International Law (2000) 209–64.

[53] For an analysis of the conflicts of laws principles in this context, see Ron McCallum, ‘Conflicts of Laws and Labour Law in the New Economy’ (2003) 16 Australian Journal of Labour Law 50, 60–4.

[54] See, eg, Sheldrick v WT Partnership (Aust) Pty Ltd (1998) 89 IR 206. See also ibid regarding conflict of laws issues.

[55] Note that there is no Australian equivalent to the United States Alien Tort Claims Act, 28 USC §1350 (1948), which enables foreign nationals to file civil suits in the United States for torts ‘committed in violation of the law of nations or a treaty of the United States’. Thus, for example, a foreign national can sue in an American court in respect to torture committed outside the United States.

[56] See, eg, CSR Ltd v Wren (1997) 44 NSWLR 463; James Hardie Industries Pty Ltd v Grigor (1998) 45 NSWLR 20 (‘Grigor’). See also the decision of the House of Lords in Lubbe v Cape plc [2000] UKHL 41; [2000] 1 WLR 1545. This decision went to procedural rather than substantive questions and the case was subsequently settled: see generally ‘Cape Plc Finally Pays Out’, Business Day (South Africa), 30 June 2003 <,3523,1377427-6078-0,00.html> .

[57] There is a preliminary issue concerning service. However, since, in the context of this article, the place of business of the defendant will be in Australia, service will not be an obstacle: Corporations Act 2001 (Cth) s 109X.

[58] [1990] HCA 55; (1990) 171 CLR 538, 557 (Mason CJ, Deane, Dawson and Gaudron JJ) (‘Voth’). Considerations relevant to determining this include: the availability of relief in a foreign forum; the presence of ‘connecting factors’ with the local forum (such as applicability of local law to the transaction, location of the parties and availability of witnesses); any ‘legitimate personal or juridical advantage’ enjoyed by the plaintiff in the local forum; and the plaintiff’s right to choose the forum (although this is of little weight): see esp at 558, 564–5, 571 (Mason CJ, Deane, Dawson and Gaudron JJ). ‘[A] court is not an inappropriate forum merely because another is more appropriate’: Regie Nationale des Usines Renault SA v Zhang [2002] HCA 10; (2002) 210 CLR 491, 503 (Gleeson CJ, Gaudron, McHugh, Gummow and Hayne JJ). Overall, this test may be more favourable to plaintiffs than those in other common law jurisdictions. See the views of Andrew Bell, ‘Human Rights and Transnational Litigation — Interesting Points of Intersection’ in Stephen Bottomley and David Kinley (eds), Commercial Law and Human Rights (2002); Peter Prince, ‘Bhopal, Bougainville and Ok Tedi: Why Australia’s Forum Non Conveniens Approach Is Better’ (1998) 47 International and Comparative Law Quarterly 573; Richard Garnett, ‘Stay of Proceedings in Australia: A “Clearly Inappropriate” Test?’ [1999] MelbULawRw 2; (1999) 23 Melbourne University Law Review 30. See also Spiliada Maritime Corporation v Consulex Ltd [1986] UKHL 10; [1987] 1 AC 460, 476–8 (Lord Goff), 464 (Lord Keith, agreeing with Lord Goff), 464 (Lord Templeman, largely agreeing with Lord Goff), 465 (Lord Griffiths, agreeing with Lord Goff), 466 (Lord Mackay, agreeing with Lord Goff); applied in Connelly v RTZ Corp plc [1997] UKHL 30; [1998] AC 854; Lubbe v Cape plc [2000] UKHL 41; [2000] 4 All ER 268, 277 (Lord Bingham, all other judges agreeing). In the United States, the predominant rule is that courts will dismiss an action where the ‘plaintiff’s chosen forum imposes a heavy burden on the defendant or the court, and where the plaintiff is unable to offer any specific reasons of convenience supporting his choice’: Piper Aircraft Co v Reyno, [1982] USSC 28; 454 US 235, 249 (1981). Further, according to the US Supreme Court, a foreign plaintiff’s choice of forum will be accorded ‘less deference’ than that of a domestic plaintiff: at 256–7.

[59] (1998) 45 NSWLR 20. An application for special leave to appeal to the High Court was refused: see Transcript of Proceedings, James Hardie & Coy Pty Ltd v Grigor (High Court of Australia, Gleeson CJ and Gaudron J, 7 August 1998).

[60] The majority comprised of Mason P and Beazley JA; Spigelman CJ dissented.

[61] The applicable law is a relevant but not decisive factor: Regie Nationale des Usines Renault SA v Zhang [2002] HCA 10; (2002) 210 CLR 491, 504, 521 (Gleeson CJ, Gaudron, McHugh, Gummow and Hayne JJ).

[62] Foreign law is only applicable in an Australian court if pleaded by a party: see, eg, Damberg v Damberg [2001] NSWCA 87; (2001) 52 NSWLR 492, 505.

[63] Bonython v Commonwealth [1951] AC 201, 219.

[64] [2002] HCA 10; (2002) 210 CLR 491.

[65] Ibid 520 (Gleeson CJ, Gaudron, McHugh, Gummow and Hayne JJ).

[66] An Australian court might refuse to enforce contractual terms that purport to apply foreign law on public policy grounds, especially if the matter appeared to be regulated by an inconsistent Australian statute: see Akai Pty Ltd v People’s Insurance Co Ltd (1996) 188 CLR 418, 447. Thus, if the foreign law permitted slavery, it would not be applied: see below n 78 and accompanying text. However, generally speaking Australian labour legislation does not have extraterritorial application: see below nn 75–6 and accompanying text.

[67] See, eg, John Gillespie, ‘Law and Development “In the Market Place”: An East Asian Perspective’ in Kanishka Jayasuriya (ed), Law, Capitalism and Power in Asia (1999) 118, 129–30; Penelope Nicholson, ‘Judicial Independence and the Rule of Law: The Vietnam Court Experience’ (2001) 3 Australian Journal of Asian Law 37, 43–4, 48–9; Tim Lindsey, ‘Black Letter, Black Market and Bad Faith: Corruption and the Failure of Law Reform’ in Chris Manning and Peter van Dierman (eds), Indonesia in Transition: Social Aspects of Reformasi and Crisis (2000) 278, 296; Randall Peerenboom, ‘Globalization, Path Dependency and the Limits of Law: Administrative Law Reform and Rule of Law in the People’s Republic of China’ (2001) 19 Berkeley Journal of International Law 161, 227.

[68] See, eg, Donald Clarke, ‘Power and Politics in the Chinese Court System: The Enforcement of Civil Judgments’ (1996) 10 Columbia Journal of Asian Law 1.

[69] Persons and associations in Australia that are interested in deploying the common law against Australian corporations on behalf of overseas workers, such as NGOs and trade unions, will not normally be able to initiate an action directly because they will not have suffered any loss compensable at common law. On the other hand, if they seek to sponsor a test case, they will face considerable logistical difficulties in identifying a plaintiff.

[70] See, eg, Grigor (1998) 45 NSWLR 20; Lubbe v Cape plc [2000] UKHL 41; [2000] 4 All ER 268.

[71] See generally Richard Meeran, ‘Liability of Multinational Corporations: A Critical Stage in the UK’ in Menno Kamminga and Saman Zia-Zarifi (eds), Liability of Multinational Corporations under International Law (2000) 251.

[72] Statute of Westminster 1931 (UK), adopted in Statute of Westminster Adoption Act 1942 (Cth) s 3. State legislation is not considered here, but see Australia Act 1986 (Cth) s 2(1).

[73] Acts Interpretation Act 1901 (Cth) s 21.

[74] Workplace Relations Act 1996 (Cth) ss 5–5AA. The constitutionality of s 5(3) (which, inter alia, extends the application of the Act to the relationship between employers (including non-Australian employers) and maritime workers (including non-Australian workers) where matters relate to trade and commerce between Australia and a place outside Australia, and between Australian states) was upheld in Re Maritime Union of Australia; Ex parte CSL Pacific Inc (2003) 200 ALR 39, 49 (Gleeson CJ, McHugh, Gummow, Kirby, Hayne, Callinan and Heydon JJ). See also McCallum, above n 53.

[75] Except perhaps where those relationships are based on contracts of employment governed by Australian law.

[76] Federal employment legislation possessing extraterritorial effect includes the Public Service Act 1999 (Cth) s 5, the Occupational Health and Safety (Commonwealth Employment) Act 1991 (Cth) s 13, and the National Occupational Health and Safety Commission Act 1985 (Cth) s 4. The operation of the first two Acts is limited to Australian employees, while the latter Act concerns Australian-controlled areas or craft. Anti-discrimination legislation has no extraterritorial effect: Sex Discrimination Act 1984 (Cth) s 9; Brannigan v Commonwealth [2000] FCA 1591; (2000) 110 FCR 566, 573 (O’Loughlin J).

[77] Criminal Code s 70.2. The Criminal Code is contained in the Schedule to the Criminal Code Act 1995 (Cth).

[78] Criminal Code s 270.3. Slavery is defined as ‘the condition of a person over whom any or all of the powers attaching to the right of ownership are exercised, including where such a condition results from a debt or contract made by the person’: s 270.1.

[79] Criminal Code s 270.6.

[80] For example, for committing offences like slavery (Criminal Code s 270.3) or sexual servitude (Criminal Code s 270.6) and deceptive recruiting for sexual services (Criminal Code s 270.5).

[81] A perpetrator must be a public official, a person acting in an official capacity, or a person acting at the instigation, or with the consent or acquiescence, of a public official or person acting in an official capacity: Crimes (Torture) Act 1988 (Cth) s 6(1)(a).

[82] Crimes (Torture) Act 1988 (Cth) ss 4, 6(1)(b).

[83] See Convention on Combating Bribery of Foreign Public Officials in International Business Transactions, opened for signature 17 December 1997, [1999] ATS 21 (entered into force 15 February 1999); International Convention to Suppress the Slave Trade and Slavery, opened for signature 25 September 1926, [1927] ATS 11 (entered into force 9 March 1927); Protocol Amending the Convention to Suppress the Slave Trade and Slavery, opened for signature 7 December 1953, 182 UNTS 51 (entered into force 7 December 1953); Supplementary Convention on the Abolition of Slavery, the Slave Trade, and Institutions and Practices Similar to Slavery, opened for signature 7 September 1956, 226 UNTS 3 (entered into force 30 April 1957); Convention on the Elimination of All Forms of Discrimination against Women, opened for signature 18 December 1979, 1249 UNTS 13 (entered into force 3 September 1981); Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, opened for signature 10 December 1984, 1465 UNTS 85 (entered into force 26 June 1987).

[84] Criminal Code ss 270.6, 270.3.

[85] Crimes (Torture) Act 1988 (Cth) s 6(1).

[86] Criminal Code s 12.1.

[87] Criminal Code s 3.1.

[88] Criminal Code s 12.2.

[89] Criminal Code s 12.3(1). This expression is further elaborated in s 12.3(2). The Criminal Code treats negligent fault elements separately: ss 5.5, 12.4.

[90] For a discussion of the involvement of MNEs in slavery and forced labour and potential legal remedies from an American perspective, see Tobias Wolff, ‘The Thirteenth Amendment and Slavery in the Global Economy’ (2002) 102 Columbia Law Review 973. Wolff points out that, in the context of the United States Constitution amend XIII, ‘slavery’ has extended to various forms of forced labour: at 979–85. It is not clear whether an Australian court would interpret the definition of slavery to extend to forced labour: see Criminal Code s 270.1 for the definition of slavery.

[91] Crimes Act 1914 (Cth) s 4B.

[92] For an analysis of the ‘business case’ for these measures, including evidence that most Australian consumers (and employees) take social and environmental issues into account, see the report prepared for Environment Australia by KPMG Sustainability Advisory Services, The State of Sustainability Reporting in the Trade and Retail Sector (2002) 8–12; Dahle Suggett and Ben Goodsir, Triple Bottom Line Measurement and Reporting in Australia (2002) 85–101. See also Ruth Pearson and Gill Seyfang, ‘New Hope or False Dawn? Voluntary Codes of Conduct, Labour Regulation and Social Policy in a Globalizing World’ (2001) 1 Global Social Policy 49. Pearson and Seyfang write (at 53) that consumer campaigns

have intelligently utilized media and internet modes of communication and linked these to increasing market niche segmentation of global brands in the fashion and sportswear industry, making leading retailers and producers sensitive to the threat of adverse publicity or consumer boycott against particular products.

[93] For an analysis of the situation in Australia, see Suggett and Goodsir, above n 92, 9–16.

[94] The ILO has produced comprehensive analyses of codes of conduct: see especially Mamic, above n 14, 946. The ILO defines a code of conduct as a ‘statement of business principles defining a set of relationships on a range of topics between an entity and its stakeholders’: at

19–20. See also the ILO’s database of private sector initiatives addressing labour and social conditions in the workplace: ILO, BASI: Business and Social Initiatives Database <> . For a discussion of the history of this development, see Jill Murray, ‘Corporate Codes of Conduct and Labour Standards’ in Robert Kyloh (ed), Mastering the Challenge of Globalisation: Towards a Trade Union Agenda (1997) 47. See also Helle Jorgensen et al, World Bank, Strengthening Implementation of Corporate Social Responsibility in Global Supply Chains (2003).

[95] Such as the toy industry code: Murray, ‘Corporate Codes of Conduct and Labour Standards’, above n 94.

[96] See the discussion of the homeworker codes in above n 47. For discussions of self-regulation proposals that involve ‘opting out’ of Australia’s formal industrial relations systems, see Richard Naughton, ‘Self-Regulation of Australian Workplaces’ (1999) 12 Australian Journal of Labour Law 131; Jill Murray, ‘Opting Out: A New Proposal for Self-Regulation’ (2000) 13 Australian Journal of Labour Law 315.

[97] See the processes established by the Minerals Council of Australia (‘MCA’) <http://> . The processes regulate matters such as the environmental and occupational health and safety practices of MCA members.

[98] KPMG Sustainability Advisory Services, above n 92, 14–15. An early example of an Australian retailer adopting a code of conduct in relation to imported goods was (then) Myer Grace Brothers, which adopted a ‘Rugmark’ code designed to ensure that its imported carpets were not made by bonded child labour: Michael Duffy et al, Report on Labour Standards in the Asia-Pacific Region (1996) [88]–[90]. However, it may no longer be possible to purchase ‘Rugmark’ rugs in Australia, for there are no Australian stores listed as Rugmark stores: see Rugmark <> .

[99] See, eg, Adelle Blackett, ‘Global Governance, Legal Pluralism and the Decentered State: A Labor Law Critique of Codes of Corporate Conduct’ (2001) 8 Indiana Journal of Global Legal Studies 401; Robert Liubicic, ‘Corporate Codes of Conduct and Product Labeling Schemes: The Limits and Possibilities of Promoting International Labor Rights through Private Initiatives’ (1998) 30 Law and Policy in International Business 111.

[100] See generally Ian Ayres and John Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (1992) 19–20.

[101] The best account of the implementation of codes of conduct in China, containing very extensive empirical information, is Liu and Tan, above n 22.

[102] The following discussion draws extensively on Mamic, above n 14, 23–39; Liubicic, above n 99; Blackett, above n 99; Sabel, O’Rourke and Fung, above n 44; Macklem, above n 10, 634–8; Jorgensen et al, above n 94, 15–31.

[103] Liubicic, above n 99, 141–2.

[104] See generally Mamic, above n 14, 23–4.

[105] The fundamental labour rights are those set out in the ILO Declaration on Fundamental Principles and Rights at Work, adopted 18 June 1998, 37 ILM 1233 (1998). The ILO Declaration identifies four key rights, set out in eight Conventions, namely:

The ILO Declaration commits ILO members to respect and promote the rights protected in these Conventions, regardless of whether states have ratified them.

[106] For example, one review of the codes of major United Kingdom corporations found that none referred to collective bargaining: Clare Ferguson, A Review of UK Company Codes of Conduct (1998) [8.25] <> . However, the 2002 ILO Business and Code of Conduct Implementation Study of major MNEs found that there was a ‘relatively high incidence of reference to the freedom of association and collective bargaining’: Mamic, above n 14, 30.

[107] See the reports referred to in below n 113. The Construction, Forestry, Mining and Energy Union (‘CFMEU’) has complained to both the Global Reporting Initiative (‘GRI’) and the Secretary-General of the United Nations (in relation to the UN Global Compact) alleging that BHP Billiton fails to accord some of its Australian workers the right to bargain collectively. See also BHP Billiton’s response to the Secretary-General: Letter from C W Goodyear (Chief Executive Officer, BHP Billiton) to Kofi Annan (Secretary-General, United Nations), 17 June 2003 <> .

[108] Pearson and Seyfang, above n 92, 66.

[109] Ibid 66–8.

[110] Suggett and Goodsir, above n 92, 19–21. A 2002 KPMG report for the Commonwealth Government indicated that only 14 per cent of Australia’s 100 top listed companies surveyed in 2002 published separate environment, social or sustainability reports compared with 23 per cent of the top 100 companies in each of 19 other countries and 45 per cent of GFT 250 companies: KPMG Sustainability Advisory Services, above n 92, 12–13.

[111] Suggett and Goodsir, above n 92. According to a KPMG survey of the 45 largest trade and retail companies in the Business Review Weekly list of the Top 1000 Australian companies, the only Australian companies to provide detailed information on environmental and social initiatives were Coles Myer, Woolworths and The Warehouse: see KPMG Sustainability Advisory Services, above n 92, 6, 14–15. On the question of labour issues in supply chains, this information is not nearly as extensive as that provided by, for example, Sainsbury in the United Kingdom: see J Sainsbury plc <> .

[112] Suggett and Goodsir, above n 92, 63. This report examined the views and practices of 29 Australian companies and seven overseas companies.

[113] In response to public concerns, Australian mining corporations have taken considerable steps to establish CSR systems, which include self-evaluation and disclosure in areas such as indigenous rights, environmental sustainability and labour standards: see, eg, WMC Resources Limited, Sustainability Report (2002) <> BHP Billiton, BHP Health Safety Environment and Community Report (2003) <

hsecReport/2003/home/home.html>. For a critical account of the activities of these corporations, see Ingrid MacDonald and Brendan Ross, Mining Ombudsman Annual Report (2003) Oxfam Community Aid Abroad <> .

[114] Standards Australia, the peak non-government standards body in Australia, has produced a series of new corporate governance standards: AS 8000–8004 (June 2003). The most relevant are AS 8002 (Organisational Codes of Conduct) and AS 8003 (Corporate and Social Responsibility). See Standards Australia <> . These standards are vague in comparison to the Global Reporting Initiative (‘GRI’) system referred to in below n 115. They focus mainly on the process of implementing codes and CSR programs. While they denote social issues that these processes should address, they do not prescribe indicators against which performance could be measured. Certain internationally recognised rights are not mentioned. For example, neither AS 8002 nor AS 8003 makes reference to collective bargaining. In this they contrast with the GRI, the OECD Guidelines (see below Part III(D)(1)), the Ethical Trading Initiative (‘ETI’) Base Code (see below Part III(D)(3)) and Social Accountability 8000 (‘SA 8000’) (see below Part III(D)(2)), as well as the fundamental ILO Conventions. Australian rating systems for CSR have also emerged, the most prominent of which is probably RepuTex, which formerly produced the Good Reputation Guides published in The Age and The Sydney Morning Herald. This ranking system does not appear to address supply chain issues in a systematic way: see the current RepuTex Ratings Results <> . The work of RepuTex is critically reviewed in Colleen Ryan, ‘The Reputation Wars’, Australian Financial Review, 24 November 2003, Boss Magazine 24; see also Mike Nahan, ‘Social Responsibility a Threat to Society’, The Australian (Sydney), 15 October 2003, 15. Nahan, executive director of the Institute of Public Affairs, argues that RepuTex and similar systems are distorted by single minded NGOs. He writes that ‘the fundamental flaw of corporate social responsibility ... is, in the end, just another attempt by the great and good to usurp the rights of ordinary people.’

[115] See GRI, Sustainability Reporting Guidelines (2002) <

guidelines/2002.asp>. The GRI is an independent international institution whose Board is composed of leading figures from business, trade unions, NGOs, government and academia. The GRI is a collaborating centre of the United National Environment Program. The GRI encourages organisations using its framework to report on ‘policies and procedures to evaluate and address human rights performance within the supply chain and contractors, including monitoring systems and results of monitoring’: at 61. In this context, ‘human rights’ refers, inter alia, to the fundamental ILO Conventions: see above n 105.

[116] As of 10 November 2003, 336 companies worldwide used the GRI Framework, 21 of them in Australia, including BHP Billiton, WMC, Ford, Toyota, Telstra and several utility companies: GRI, Organisations Using the Guidelines (2004) <


[117] Suggett and Goodsir, above n 92, 83.

[118] Liu and Tan, above n 22, 76–9; Dara O’Rourke, ‘Monitoring the Monitors: A Critique of Corporate Third-Party Labour Monitoring’ in Gill Seyfang, Rhys Jenkins and Ruth Pearson (eds), Corporate Responsibility and Labour Rights: Codes of Conduct in the Global Economy (2002) 196. See also Sasha Courville, ‘Social Accountability Audits: Challenging or Defending Democratic Governance?’ (2003) 25 Law and Policy 269.

[119] The 2002 ILO Business and Code of Conduct Implementation Study reported that its ‘research has revealed that auditors may hesitate to present truly damaging information, based on a desire to maintain good business relations and receive future work from the client’: Mamic, above n 14, 39.

[120] See, eg, Jorgensen et al, above n 94, 22–6, 48–50. Initiatives that include a commitment to collective bargaining and worker participation as understood in ILO Conventions are less amenable to this criticism. However, as we have seen, these commitments are often excluded from private codes.

[121] Liu and Tan, above n 22 (translated by author).

[122] Blackett, above n 99, 418–23.

[123] For a comprehensive analysis of employee empowerment in firms under conditions of flexible production, see Barenberg, above n 10, 879–92.

[124] For example, empirical evidence strongly suggests that active worker involvement in occupational health and safety programs increases their effectiveness: see, eg, the research findings reviewed in David Walters, ‘Workplace Arrangements for the 21st Century’ (Working Paper No 10, National Research Centre for Health and Safety Regulation, Australian National University, 2003).

[125] Murray, ‘Corporate Codes of Conduct and Labour Standards’, above n 94, 75.

[126] Pearson and Seyfang, above n 92, 64–5. The 2002 ILO Business and Code of Conduct Implementation Study strongly supported systematic consultation with local stakeholders: Mamic, above n 14, 20.

[127] Private sector codes of conduct, which are now quite widespread in Guangdong, seem to have done little to stop excessive hours of work or remedy other abuses. For example, Liu and Tan write of a company with over 6000 employees that has adopted a code of conduct limiting working hours, as required by its major client, an international footwear company, and has stationed a permanent auditor. However, the workers informed investigators outside the factory that they nevertheless worked 12 to 14 hour days: Liu and Tan, above n 22, 75. This failure to implement the code would appear to be a widespread phenomenon.

[128] See Reebok Business Practices <

improving.html>. For example, in 2002, Reebok arranged for open, competitive elections for union representatives, involving secret ballots, to be conducted in two of its supplier firms in China. The elections were observed by Hong Kong labour rights NGOs and academics: see Alison Maitland, ‘Sewing a Seam of Worker Democracy in China’, Financial Times (London), 12 December 2002, 10. While firm unions must remain affiliated with the ACFTU under Chinese law, the workers in one firm were able to vote out the ACFTU-sponsored union chair. Subsequent opposition by the ACFTU hierarchy may mean that this particular initiative cannot be readily replicated. However, Reebok is continuing to explore what it can feasibly do in the Chinese context to increase workers’ capacity to improve their conditions.

[129] For an Australian example, in the BSL initiative (see Lillywhite, above n 48, 7), the adoption of a prescriptive code of conduct has been rejected in favour of an ‘engagement’ approach in which the BSL focused on what practical, measurable steps could be taken by the BSL (given its small market power) in the workplaces in question. An occupational health and safety program involving worker participation is therefore being negotiated between the BSL and the suppliers. The program is being funded out of BSL profits.

[130] See Reebok’s Human Rights Standard and Chinese Workers’ Working Conditions (2002) China Labor Watch <> .

[131] The BSL project being a notable exception: see above n 129.

[132] See generally Mamic, above n 14, 25, 40–5. In the late 1990s, the Australian Council of Trade Unions (‘ACTU’) devised a Sydney Olympic Games Code of Conduct, which sought to ensure that certain overseas suppliers of Games merchandise adhered to core labour standards and a number of other standards in the areas of pay and hours of work. In the analysis by myself and Jill Murray, this was unsuccessful. The supervisory procedures were entirely ineffective. The implementation of the Code consisted simply of a questionnaire sent to supplier firms in six countries. The questionnaire asked whether the firms complied with certain specified labour standards. The questionnaire was sent after the supply contracts had been concluded. There was no monitoring or verification procedure, although the ACTU did attempt (unsuccessfully) to request that peak union bodies in some of the relevant countries inspect the firms. With the assistance of colleagues in Taiwan, we attempted several times to interview managers of the Taiwanese firms in relation to their commitments under the Code. These attempts were all rebuffed. This experience suggests that unilateral attempts by non-governmental bodies to set up local equivalents to the major international schemes considered below are likely to fail in the absence of the coordination of, and cooperation from, a wide range of stakeholders and firms.

[133] Two other well-known examples of voluntary CSR initiatives coordinated through international agencies are the ILO Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy, adopted 17 November 2000, 41 ILM 186 (2002) and the United Nations Global Compact (launched at UN Headquarters, New York, 26 July 2000) <> . These differ from the OECD Guidelines in that they do not require the establishment of a specific governmental agency, such as the National Contact Point: see below nn 138–43 and accompanying text.

[134] For a discussion of the revisions, see Jill Murray, ‘A New Phase in the Regulation of Multinational Enterprises: The Role of the OECD’ (2001) 30 Industrial Law Journal 255.

[135] See above n 105.

[136] See OECD Guidelines (2000) pt IV (‘Employment and Industrial Relations’).

[137] OECD Guidelines (2000) pt II s 10.

[138] The Australian NCP is the General Manager of the Foreign Investment Policy Division of the Department of Treasury. The NCP conducts biannual consultations with academics, NGOs, trade unions and business representatives: see Australian NCP for the OECD Guidelines for Multinational Enterprises <> .

[139] OECD Guidelines (2000) pt 1C (‘National Contact Points: Implementation in Specific Instances’).

[140] OECD, OECD Guidelines for Multinational Enterprises: 2003 Meeting of the National Contact Points: Report by the Chair, 23–24 June 2003, 9–11.

[141] A specific instance complaint was made on 19 November 2003 in Canberra, under the previous OECD Guidelines concerning industrial relations in the mining sector.

[142] The Chair’s Report of the 2003 meeting of NCPs advised that:

The Australian NCP increased its efforts to incorporate the Guidelines into various domestic corporate governance and social responsibility reporting frameworks (eg Standards Australia’s Corporate Governance, Corporate Social Responsibility and Bribery papers, Australia’s Triple Bottom Line Reporting Guidelines, Australia’s Environmental Reporting Guidelines and the Australian Securities and Investment Commission’s Socially Responsible Investing Disclosure Guidelines).

OECD, OECD Guidelines for Multinational Enterprises: 2003 Meeting of the National Contact Points, above n 140, 4.

[143] See above n 138.

[144] Schemes established on a similar basis include the European-based Clean Clothes Campaign <> , the rival US-based clothing certification schemes the Worker Rights Consortium <> and the Fair Labor Association <> , active in college and university apparel labelling, and Fairtrade Labelling Organizations International <> , focusing on labelling the products of primary producers, which is active across the Northern Hemisphere.

[145] SAI <> . See generally Sasha Courville, ‘Social Accountability Audits: Challenging or Defending Democratic Governance?’ (2003) 25 Law and Policy 269. SAI is an offshoot of the Council of Economic Priorities and is a member of the International Social and Environmental Accreditation and Labelling (ISEAL) Alliance <> .

[146] The organisations involved include CARE International, Amnesty International, the International Textile, Garment and Leather Workers Federation, Toys R Us, and the Dole Food Company. The SAI Advisory Board includes members of these organisations: SAI, Boards at SAI <> .

[147] The standard is available at SAI, SA8000 Documents (2001) <

Document%20Center/Documents.htm>. It is supplemented by a detailed Guidance Document, also available at that website.

[148] SA8000 cll 9.6–9.9.

[149] SA8000 cl 8.3.

[150] The SAI scheme aims to provide credible information to the market on firms’ compliance. By advertising through measures such as product labelling, firms can use certification to differentiate themselves from competitors not only on price or quality, but on the basis of consumer preferences for socially-responsible manufacturing. As of May 2004, 400 facilities (employing around 260 000 people) had been certified SA8000 compliant. No Australian firm has been certified: SAI, SA8000 Certified Facilities As of May 20, 2004 <


[151] There are extensive processes for complaints and appeals. There is also a ‘Corporate Involvement Program’, which assists firms to gain SA8000 certification: see SAI, Corporate Involvement Program (2004) <> .

[152] As of May 2004, out of the hundreds of thousands of firms in China participating in supply chains, there were only 53 certified under the SAI procedures. Even among certified firms, there have been instances of non-compliance and false reporting: see the criticisms of SA8000 in Ed Shepherd, ‘SA8000 — An External Code of Conduct, Accreditation and Monitoring System’ (2001) 37 Asian Labour Update <> .

[153] Labour Law 1994 (People’s Republic of China) art 36 stipulates a normal working week of 44 hours. However the State Council has reduced this to 40 hours: Regulation of the State Council Governing Working Hours for Workers 1995 (People’s Republic of China). This Regulation has some flexibility in its application.

[154] See also Mamic, above n 14, 44–5. I understand that proposals to modify the SA8000 working hours clause (cl 7) in relation to agriculture have been formulated as part of the Social Accountability in Sustainable Agriculture project, in which SAI is collaborating: Interview with Sasha Courville (November 2003).

[155] See ETI <> . Members include The Body Shop International, Levi Strauss, Marks and Spencer and Safeway Stores, as well as the Trades Union Congress, the International Confederation of Free Trade Unions, Oxfam and Save the Children: ETI, Members of the Ethical Trading Initiative (2004) <


[156] ETI, The Base Code (2004) <> (‘ETI Base Code’).

[157] Thus cl 8.2 of the ETI Base Code provides:

Obligations to employees under labour or social security laws and regulations arising from the regular employment relationship shall not be avoided through the use of labour-only contracting, sub-contracting, or home-working arrangements, or through apprenticeship schemes where there is no real intent to impart skills or provide regular employment, nor shall any such obligations be avoided through the excessive use of fixed-term contracts of employment.

[158] ETI, Purpose, Principles, Programme and Membership Information <http://www.ethicaltrade.


[159] Mamic, above n 14, 29.

[160] The OECD Guidelines would appear to provide more extensive support for worker participation: see OECD Guidelines pt IV. Cf SA8000 cl 4; ETI Base Code cl 2.

[161] For an excellent statement of the case for this form of intervention, see Macklem, above n 10.

[162] See, eg, Robert Baldwin and Martin Cave, Understanding Regulation: Theory, Strategy and Practice (1999); Anthony Ogus, Regulation: Legal Form and Economic Theory (1994); Neil Gunningham and Peter Grabosky, Smart Regulation: Designing Environmental Policy (1998).

[163] Australian Constitution s 51(xxix).

[164] Polyukhovich v Commonwealth [1991] HCA 32; (1991) 172 CLR 501, 632 (Dawson J), approved by a majority of the Court in Victoria v Commonwealth (1995) 187 CLR 416, 485 (Brennan CJ, Toohey, Gaudron, McHugh and Gummow JJ) (‘Industrial Relations Act Case’).

[165] See, eg, George Williams, Labour Law and the Constitution (1998) 90–1.

[166] If the regulatory strategies implemented an international treaty, then the relevant legislation might be supported by the external affairs power. That power, for example, supports those parts of the Workplace Relations Act 1996 (Cth) implementing Conventions and recommendations of the ILO. For a law to be valid under the power, it ‘must be reasonably capable of being considered appropriate and adapted to implementing the treaty’: Industrial Relations Act Case (1995) 187 CLR 416, 487 (Brennan CJ, Toohey, Gaudron, McHugh and Gummow JJ). Further, although it is possible to implement just some of the obligations of a treaty, where the treaty specifies an implementation regime, the Commonwealth is not free to legislate without regard to that regime: at 486. Whether the most obvious treaties — the ILO Conventions — are pertinent here may turn on whether they can be construed as imposing obligations on states which travel beyond the labour conditions within their territory.

[167] For an analysis of the alternative bases for federal employment regulation, see generally Williams, above n 165; Andrew Stewart, ‘Federal Labour Law and New Uses for the Corporations Power’ (2001) 14 Australian Journal of Labour Law 145.

[168] Australian Constitution s 51(xx). Although the precise scope of the power is uncertain, it is reasonably clear that a law specifically directed at regulating the conditions of the employees of a corporation is a valid exercise: Re Dingjan; Ex parte Wagner [1995] HCA 16; (1994) 183 CLR 323. The power would perhaps support broader regulatory measures, such as requiring corporations to be to some extent accountable for the employment policies of their subcontractors: see generally Williams, above n 165, 115–19; Stewart, above n 167, 154–60.

[169] Australian Constitution s 51(i). See Re Maritime Union of Australia; Ex parte CSL Pacific Inc (2003) 200 ALR 39. The High Court held that ‘[i]t is ... well settled that, in the exercise of the trade and commerce power, the Parliament can validly regulate the conduct of persons employed in those activities which form part of trade and commerce with other countries and among the States’: at 48 (Gleeson CJ, McHugh, Gummow, Kirby, Hayne, Callinan and Heydon JJ). The Court rejected arguments that the trade and commerce power was available only if an employer or employee were located in Australia: at 49.

[170] Australian Constitution s 51(ii). This might be relied upon to support financial incentives for a corporation to direct resources to improving the conditions of the overseas workers connected to it, for example by imposing a charge on corporations who failed to expend a prescribed minimum amount on such improvements. Compare the use of the taxation power in relation to the former Training Guarantee Act 1990 (Cth), upheld in Northern Suburb General Cemetery Reserve Trust v Commonwealth [1993] HCA 12; (1992) 176 CLR 555.

[171] The Australian Constitution s 51(xxxv) empowers the Commonwealth to make laws for ‘conciliation and arbitration for the prevention and settlement of industrial disputes extending beyond the limits of any one State’. Generally speaking, the arbitration power cannot be used to make laws establishing labour standards; it enables the federal Parliament to regulate working conditions only in so far as this is a consequence of resolving interstate industrial disputes through conciliation and arbitration: see, eg, Re Pacific Coal Pty Ltd; Ex parte Construction, Forestry, Mining and Energy Union (2000) 203 CLR 346, 359–60 (Gleeson J), 368–70 (Gaudron J), 383–4 (McHugh J), 404–5 (Gummow and Hayne JJ), 430–6 (Kirby J). The power would not, therefore, support extraterritorial regulatory strategies, such as the imposition of minimum standards or reporting requirements on Australian firms operating overseas, that are not premised on a system of conciliation and arbitration. It may support legislation allowing the Australian Industrial Relations Commission to conciliate and arbitrate disputes arising from an Australian corporation’s employment activities outside Australia where the dispute had a ‘distinct industrial connexion with Australia’: R v Foster; Ex parte Eastern & Australian Steamship Co Ltd [1959] HCA 10; (1958) 103 CLR 256, 278. See Williams, above n 165, 43–85.

[172] Originally introduced into the Senate in September 2000, the Bill was restored to the Notice Paper on 13 February 2002. See Commonwealth, Parliamentary Debates, Senate, 13 February 2002, 200 (Vicki Bourne).

[173] In 2000 and 2001, former United States Congress member Cynthia McKinney introduced two similar Bills into the US House of Representatives. The first, the ‘TRUTH’ (Transparency and Responsibility for US Trade Health) Bill, H R 5492, 106th Cong (2000), reintroduced as H R 460, 107th Cong (2001), required American corporations trading overseas to disclose information about their operations, including their location, the age and gender of employees and labour practices. Unlike the CCC Bill, the TRUTH Bill applied not only to the direct foreign employees of US corporations, but also to persons employed by their ‘subsidiaries, subcontractors, affiliates, joint ventures, partners, or licensees’: cl 2(a). A second Bill, the Corporate Code of Conduct Bill, H R 2782, 107th Cong (2001), required US nationals employing more than 20 persons in a foreign country to implement a code of conduct with respect to the employment of those persons. This Bill would have required government agencies to give preference to entities that adopted and enforced such codes. Cynthia McKinney is no longer in Congress and the Bills have now lapsed.

[174] The PJSCCS is now the Parliamentary Joint Committee on Corporations and Financial Services. The Committee handed down its report on 28 June 2001: PJSCCS, Parliament of Australia, Report on the Code of Conduct Bill 2000 (2001) (‘PJSCCS Report’).

[175] The five government members of the PJSCCS, who constituted half its membership, recommended that the Bill not be passed ‘because it is unnecessary and unworkable’: ibid [4.53]. The four Labor members of the Committee, on the other hand, determined that the ‘intent’ of the Bill was ‘right and proper’ and opted for a mandatory reporting and auditing scheme: PJSCCS, Parliament of Australia, Comments by Labor Members on the Corporate Code of Conduct Bill 2000 (2001). Senator Andrew Murray, the sole Australian Democrat member of the Committee, in a lengthy minority report, argued that there was a demonstrated need for the Bill, but accepted that it required amendment: PJSCCS, Parliament of Australia, Minority Report on the Corporate Code of Conduct Bill 2000 (2001) [89].

[176] CCC Bill cl 4.

[177] The CCC Bill refers to these as ‘overseas corporations’.

[178] CCC Bill cl 3(1)(b).

[179] CCC Bill cl 3(1)(c). The Bill does not require overseas corporations to comply with standards to which they are not subject in Australia: cl 3(2).

[180] Defined as ‘a wage sufficient to meet the basic needs of a family of two adults and three children in the country or region they are resident in’: CCC Bill cl 6.

[181] CCC Bill cll 8–10.

[182] This provision is redundant because the content of the other substantive provisions is largely drawn from those Conventions. Moreover, this provision misconstrues the nature of ILO Conventions by requiring corporations to comply with standards contained in them. ILO Conventions apply to states, not corporations. Drafters of any future proposed legislation on this subject should distinguish between the obligations Conventions impose on a state and those which would be imposed on a corporation if a state ratified the Convention.

[183] It would appear that proceedings may be instituted only by the Treasurer, the Attorney-General or the Chairperson of the Australian Securities and Investments Commission (‘ASIC’): CCC Bill cl 16.

[184] CCC Bill cl 17. This provision appears to be modelled on provisions in pt VI of the Trade Practices Act 1974 (Cth). If loss or damage has been sustained, the Court can grant an injunction or make an order for ‘compensation’. If the Court is satisfied only that a person is reasonably likely to suffer loss or damage, it can grant an injunction only.

[185] CCC Bill cl 17(6).

[186] For a classic analysis, see generally Eugene Bardach and Robert Kagan, Going by the Book: The Problem of Regulatory Unreasonableness (1982).

[187] CCC Bill cl 9(3)(a).

[188] SA8000 takes a more workable approach: ‘The company shall ensure that wages paid for a standard working week shall always meet at least legal or industry minimum standards and shall be sufficient to meet basic needs of personnel and to provide some discretionary income’: pt IV cl 8.1. However, it would be very difficult for an Australian court to interpret and enforce this, as assessment of whether it had been breached would require extensive evidence from the relevant foreign country.

[189] SA8000 pt IV cl 1. As pointed out above, the SA8000 scheme also establishes categories of non-compliance, each with different consequences.

[190] In the case of forced child labour, however, the obligation is broader. ‘An overseas corporation must not use or obtain the benefit of’ such labour: CCC Bill cll 9(1), (2).

[191] One promising attempt to do so within the Australian context can be found in state legislation regulating outworkers in the clothing industry: see, eg, Outworkers (Improved Protection) Act 2003 (Vic) s 12. This legislation imposes liability upon principal manufacturers and potentially upon retailers (if they fail to comply with existing voluntary codes) at the head of the supply chain for outworker entitlements. It may well prove effective in Australia because it harnesses the clear market power conferred on retailers by the oligopolistic structure of the top of the supply chains. I do not think it could be readily extended offshore in its current form. Apart from the multiplicity of legal difficulties discussed in this section, Australian firms do not exercise the same degree of control over offshore supply chains as they do over suppliers in this country. As Igor Nossar, Richard Johnstone and Michael Quinlan remark in the context of considering the prospect of extending this form of legislation offshore: ‘this ... depends entirely on the effective business controllers of supply chains, the large retailers, being located within the relevant geographical jurisdiction in which the legislation is enacted’: Igor Nossar, Richard Johnstone and Michael Quinlan, ‘Regulating Supply-Chains to Address the Occupational Health and Safety Problems Associated with Precarious Employment: The Case of Home-Based Clothing Workers in Australia’ (Working Paper No 21, National Research Centre for OHS Regulation, Australian National University, 2003) 16 <> .

[192] This is compounded in the case of subsidiaries incorporated in a foreign jurisdiction. The CCC Bill applies to the overseas subsidiaries of Australian corporations: cl 4. As Robert McCorquodale points out, under international law, these subsidiaries are the nationals of the states in which they are incorporated, not the state to which their parent company belongs, even if it controls the subsidiary: Robert McCorquodale, ‘Human Rights and Global Business’ in Stephen Bottomley and David Kinley (eds), Commercial Law and Human Rights (2002) 89, 94–5, 100–2 (citing Case concerning the Barcelona Traction, Light and Power Company Ltd (Belgium v Spain) (Second Phase) [1970] ICJ Rep 3, 18–19). Consequently, there is potential for conflict not only with foreign labour law, but also with foreign corporations law.

[193] See Freedom of Association and Protection of the Right to Organise Convention (ILO No 87), opened for signature 9 July 1948, 68 UNTS 17 (entered into force 4 July 1950); Right to Organise and Collective Bargaining Convention (ILO No 98), opened for signature 1 July 1949, 96 UNTS 257 (entered into force 18 July 1951).

[194] The current version of this principle was introduced in 2001 in Trade Unions Law 1992 (People’s Republic of China) art 9.

[195] Trade Unions Law 1992 (People’s Republic of China) art 51. A similar difficulty arises in the context of the obligation not to discriminate on the basis of race or sexuality. For example, the Malaysian Constitution provides for positive discrimination in favour of Malays, and Malaysian law criminalises consensual male sexual relations: Penal Code 1976 (M’sia) ss 377A–377B. Persons who violate this provision may be subject to up to 20 years in prison, and to whipping.

[196] Cf SA8000 pt IV cl 4.2.

[197] See, eg, Ayres and Braithwaite, above n 100, 19–53; Gunningham and Grabosky, above n 162, 45–6; Neil Gunningham and Richard Johnstone, Regulating Workplace Safety: Systems and Sanctions (1999) 34–6.

[198] See, eg, Parker, above n 1, 26–7. See also Clifford Shearing, ‘A Constitutive Concept of Regulation’ in Peter Grabosky and John Braithwaite (eds), Business Regulation and Australia’s Future (1993) 67, 67–79.

[199] See above nn 120–6 and accompanying text.

[200] That is, in circumstances other than those defined in Convention concerning the Prohibition and Immediate Action for the Elimination of the Worst Forms of Child Labour, opened for signature 17 June 1999, 2133 UNTS 163, art 3 (entered into force 19 November 2000): see below n 209.

[201] The Conventions against slavery and sexual servitude are almost universally ratified: see, eg, International Convention to Suppress the Slave Trade and Slavery, opened for signature 25 September 1926, [1927] ATS 11 (entered into force 9 March 1927); Protocol Amending the Convention to Suppress the Slave Trade and Slavery, opened for signature 7 December 1953, [1953] ATS 8 (entered into force 7 December 1953); Supplementary Convention on the Abolition of Slavery, the Slave Trade, and Institutions and Practices Similar to Slavery, opened for signature 7 September 1956, [1958] ATS 3 (entered into force 30 April 1957); Convention on the Elimination of All Forms of Discrimination against Women, opened for signature 18 December 1979, 1249 UNTS 13 (entered into force 3 September 1981). China has not ratified them (they were originally ratified by the former Nationalist regime under the title ‘Republic of China’). However, the Criminal Law 1979 (People’s Republic of China) severely punishes slavery and sexual servitude: arts 238–42.

[202] See Wolff, above n 90.

[203] See the United Nations Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, opened for signature 10 December 1984, 1465 UNTS 112 (entered into force 26 June 1987). States party to this Convention include countries as diverse as the United States, Indonesia and China. Whether they comply with it is another question.

[204] Torture is defined in Australian law as ‘any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person’: Crimes (Torture) Act 1988 (Cth) s 3(1). This reflects the wording of the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, opened for signature 10 December 1984, 1465 UNTS 112 (entered into force 26 June 1987) art 1.

[205] Crimes (Torture) Act 1988 (Cth) s 6.

[206] Opened for signature 10 December 1984, 1465 UNTS 85 (entered into force 26 June 1987).

[207] See above Part IV(A).

[208] It may be possible to formulate a provision that covers a related egregious labour abuse — deprivation of liberty endangering life. This would apply to firms which lock workers into factories so that they are unable to escape in the event of fires and other disasters.

[209] Opened for signature 17 June 1999, 2133 UNTS 163 (entered into force 19 November 2000). Article 3 provides:

For the purposes of this Convention, the term the worst forms of child labour comprises:

Article 3(d) is further defined in ILO, Recommendation concerning the Prohibition and Immediate Action for the Elimination of the Worst Forms of Child Labour, adopted 17 June 1999, 38 ILM 1211 (1999). However, this recommendation does not clearly explain the extent of the obligation and it may require further clarification in any statutory scheme.

[210] ILO, Convention No C182 Was Ratified by 150 Countries (2004) <


[211] ILO, Has Your Country Ratified? (2004) <


[212] In formulating obligations and sanctions in the subcontracting context, it would be very helpful to consider how occupational health and safety regulation pertaining to subcontractors might be adapted to offshore supply chains: cf Gunningham and Johnstone, above n 197, 129–33; Nossar, Johnstone and Quinlan, above n 191.

[213] Cf Gunningham and Johnstone, above n 197, 192–5, 267–70.

[214] The majority report of the PJSCCS claimed that the imposition of labour standards on Australian corporations operating offshore could offend foreign nations, commenting that Australians have in the past resented foreign nations applying their laws extraterritorially within their jurisdictions: PJSCCS Report, above n 174, [3.61]–[3.67], [4.49]. This is unconvincing since it is not suggested that Australian law should apply directly to Chinese workers or Chinese companies.

[215] See, eg, Deborah Senz and Hilary Charlesworth, ‘Building Blocks: Australia’s Response to Foreign Extraterritorial Legislation’ [2001] MelbJlIntLaw 3; (2001) 2 Melbourne Journal of International Law 69, 70–2.

[216] It is notable that some of the submissions from business groups which raised this objection before the PJSCCS were less opposed to ‘very targeted and very focused’ legislation: PJSCCS Report, above n 174, [3.77]–[3.82].

[217] McCorquodale, above n 192, 102 comments that:

the prohibition of slave labour by a home state in relation to all the activities of all corporations within a TNC [trans-national corporation] is likely to be consented to by a host state as being in the general interests of the international community. On the other hand, if a home state sought to impose strict labour regulation, such as limits on working hours, rates of pay and benefits on all such corporations, it is likely to be objected to by the home state as an uwaranted intrusion into its sovereignty.

[218] See above n 15 and accompanying text.

[219] Gunther Teubner, ‘Substantive and Reflexive Elements in Modern Law’ (1983) 17 Law and Society Review 239.

[220] See generally Ayres and Braithwaite, above n 100.

[221] See generally Parker, above n 1, 245–91.

[222] Michael Dorf and Charles Sabel, ‘A Constitution of Democratic Experimentalism’ (1998) 98 Columbia Law Review 267.

[223] See, eg, Gunningham and Johnstone, above n 197; Parker, above n 1, 26–9.

[224] See, eg, Mamic, above n 14.

[225] Sabel, O’Rourke and Fung, above n 44.

[226] Ibid.

[227] Ibid 15.

[228] Ibid 17–18.

[229] Ibid 13.

[230] Ayres and Braithwaite, above n 100, 50.

[231] While RLS and prescriptive approaches are contrasted here, it should be noted that they are not mutually incompatible. Indeed the evaluation process can, over time, identify practices which could form a floor suitable for legislative mandate, as noted by Sabel, O’Rourke and Fung, above n 44, 16:

the system also provides the basis for elaborating a framework of baseline social performance levels and procedures over time. In the first instance, there will be many diverse and contending practices and standards among monitors and firms. Over time, however, some routines will become widespread and accepted as so basic that all sincere and capable firm[s] should adopt them. These routines would then become minimal in the sense that all firms participating in the rolling rule regime would be required to adopt them.

[232] Owens, above n 14, 191–4 (noting in particular the failure of the RLS model to promote collective bargaining); Jill Murray, ‘The Sound of One Hand Clapping? The “Ratcheting Labour Standards” Proposal and International Labour Law’ (2001) 14 Australian Journal of Labour Law 306. Murray also notes that the relationship between racheting labour standards and ILO standards has not been properly clarified.

[233] Sabel, O’Rourke and Fung, ‘Ratcheting Labour Standards’, above n 44, 32.

[234] See above n 98 and accompanying text.

[235] Ian Ayres, ‘Monetize Labor Practices’ in Joshua Cohen and Joel Rogers (eds), Can We Put an End to Sweatshops? (2001) 80, 80.

[236] Archon Fung, Dara O’Rourke and Charles Sabel, ‘Realizing Labor Standards’ in Joshua Cohen and Joel Rogers (eds), Can We Put an End to Sweatshops? (2001) 3, 47.

[237] Ayres, above n 235, 80.

[238] See generally Parker, above n 1, 216–21.

[239] See above n 114.

[240] This is especially the case as regards those already involved in other forms of social auditing.

[241] This is a form of ‘meta-regulation’ or reflexive regulation in that it induces firms to improve their own internal processes for addressing labour standards, but does not direct what the outcomes of those processes should be: see Parker, above n 1, 279–83; David Hess, ‘Social Reporting: A Reflexive Law Approach to Corporate Social Responsiveness’ (1999) 25 Journal of Corporation Law 41. The proposals here are much less sophisticated than those discussed by Parker and Hess.

[242] This could be, but need not be, an existing authority such as ASIC. Although the proposal does not entail that legislative provisions be located in the Corporations Act 2001 (Cth), or that the relevant authority be ASIC, this may be convenient.

[243] It would make sense to incorporate reporting on foreign workers into a wider social and environmental framework. This was the approach taken in the the recent French amendments to the Code de Commerce 2002 (France) which provide for mandatory reporting on social and environmental practices:

The [Annual Report] shall also include information, a list of which shall be fixed by an Order approved by the Conseil d’Etat, on the way in which the company has taken the social and environmental consequences of its activities into account. This sub-paragraph shall not apply to companies whose shares are not admitted to trading on a regulated market ...

Article L 225-102-1. See also Décret no 2002-221 du 20 février 2002 pris pour l’application de l’article L 225-102-1 du code de commerce et modifiant le décret no 67-236 du 23 mars 1967 sur les sociétés commerciales. This detailed administrative order indicates that reporting must cover subcontracting activities, but does not specifically refer to overseas activities.

[244] The proposed legislative structure here is similar to that pertaining to ‘Product Disclosure Statements’ (‘PDS’): pt 7.9 div 2 of the Corporations Act 2001 (Cth) requires issuers and sellers of financial products and other regulated persons to issue PDSs and to lodge them with ASIC. In relation to products with investment components, such as superannuation products, managed investments and investment life insurance, a PDS must include certain stipulated ‘information ... a person would reasonably require for the purpose of making a decision, as a retail client, whether to acquire the financial product’: s 1013D(1). These products may require information regarding ‘the extent to which labour standards ... are taken into account in the selection, retention or realisation of the investment’: s 1013D(1)(l). If the PDS indicates that the product issuer or seller does take into account labour standards, the Corporations Regulations 2001 (Cth) provide that the PDS must also indicate the standards that the product issuer considers to be labour standards for that purpose (reg 7.9.14C(c)(i)) and the extent to which the product issuer takes those standards into account in the selection, retention or realisation of the investments (reg 7.9.14C(c)(ii)). ASIC has issued guidelines pursuant to s 1013DA of the Corporations Act 2001 (Cth) clarifying what this disclosure obligation entails: ASIC, Section 1013DA Disclosure Guidelines (2003). They are ‘designed to facilitate (a) transparency ... (b) accuracy ... (c) comprehensibility ... and (d) comparability ...’: at 2. They are non-prescriptive, but make clear that product issuers must indicate the methodology for taking standards into account, where one exists, and the weight given to the standard, where a weightings system is used, and should disclose the nature of any external ratings system used: at 9–10. A breach of the PDS obligations, including the production of false or misleading statements (Corporations Act 2001 (Cth) s 1041E) may lead to the imposition of penalties, and to civil liability: Corporations Act 2001 (Cth) pt 7.9 div 7; pt 7.10.

[245] For criticism of such language, see generally Don Watson, Death Sentence: The Decay of Public Language (2003); see esp at 48–54.

[246] Part 3 of the CCC Bill establishes very extensive reporting requirements which, however, cannot be readily applied to supply chains.

[247] By analogy with the Trade Practices Act 1974 (Cth), civil actions should be available for persons who suffer loss as a result of the false, misleading or deceptive reports.

[248] Section 299(1)(f). This provision was included in the Act as a result of a Democrats amendment. In a review of the provision, the government members of the PJSCCS recommended that it be deleted. This measure was opposed by the ALP and the Democrats: PJSCCS, Parliament of Australia, Matters Arising from the Company Law Review Act 1998 (1999). The provision has not yet been repealed.

[249] One analysis of reporting behaviour before and after the introduction of this provision concluded that the introduction of s 299(1)(f) significantly improved overall reporting by Australian companies on their environmental performance, contradicting arguments that voluntary and existing disclosure requirements were effective, and that companies would have difficulty in interpreting the provision: Geoffrey Frost and Linda English, ‘Mandatory Corporate Environmental Reporting in Australia: Contested Introduction Belies Effectiveness of Its Application’ (2002) 3 Drawing Board: An Australian Review of Public Affairs <


[250] Baldwin and Cave, above n 162, 49.

[251] See generally Parker, above n 1, 284–8. However, Parker suggests that certain immunities should be provided for self-disclosure and self-correction by companies.

[252] Cf Equal Opportunity for Women in the Workplace Act 1999 (Cth) s 14.

[253] For example, a firm might do this by demonstrating that it was collaborating with NGOs in an ETI-style experimental measure.

[254] Suggett and Goodsir, above n 92, 94–101.

[255] Corporations Act 2001 (Cth) s 292. See s 45A(2) for the definition of ‘small proprietary company’.

[256] For example, the Department of the Environment and Heritage, the Department of Family and Community Services, the Department of Foreign Affairs and Trade and (in relation to the OECD Guidelines) the Treasury.

[257] Such initiatives could be integrated into an already established CSR program. However, the danger with this is that the sweatshops issue might be marginalised in the context of CSR initiatives that impact more immediately on the Australian population.

[258] US Department of State, Labor <> . See also United Kingdom Department for International Development, which played a key role in setting up the ETI: United Kingdom Department for International Development, How DFID Works in the UK <> .

[259] The discussion draws on a European example of facilitative government action. See Jorgensen et al, above n 94, 39, 55.

[260] For an empirical investigation of the current state of CSR in Europe, see generally Jean-Pierre Segal, André Sobczak and Claude-Emmanuel Triomphe, Corporate Social Responsibility and Working Conditions (2003). This report, while pointing to a number of positive measures, highlights the embryonic state of CSR in relation to subcontracting, and the inherent difficulties in comparing national CSR reports.

[261] European Commission, Communication from the Commission concerning Corporate Social Responsibility: A Business Contribution to Sustainable Development, COM(2002) 347 final. For a critique of the European approach, see Pall Davidsson, ‘Legal Enforcement of Corporate Social Responsibility within the EU’ (2002) 8 Columbia Journal of European Law 529. See also European Commission, Communication from the Commission to the Council, the European Parliament and the Economic and Social Committee: Promoting Core Labour Standards and Improving Social Governance in the Context of Globalisation, COM(2001) 541 final. The European Parliament, while generally supporting this voluntarist approach, has pressed for the introduction of binding measures to enable greater transparency and scrutiny of CSR reporting: European Parliament, Report on the Commission Green Paper on Promoting a European Framework for Corporate Social Responsibility, Eur Parl Doc (A5-0159/2002 final); European Parliament, Report on the Communication from the Commission concerning Corporate Social Responsibility: A Business Contribution to Sustainable Development (A5-0133/2003 final). The first resolution, passed prior to the Commission’s Communication, proposed a Directive on CSR reporting. The second resolution was less specific, but also contemplated a form of mandatory reporting.

[262] The forum was launched in October 2002: see EU Multi Stakeholder Forum on CSR <> .

[263] European Commission, Communication concerning Corporate Social Responsibility, above n 261, 18.

[264] Ibid.

[265] See especially pt IV cll 1, 2, 7–8.

[266] Part IV cl 4.

[267] Clause 2.1.

[268] See above nn 107, 133.

[269] (ILO No 98), opened for signature 1 July 1949, 96 UNTS 257 (entered into force 18 July 1951).

[270] See ILO Committee of Experts on the Application of Conventions and Recommendations, Individual Observation concerning Convention No 98, Right to Organise and Collective Bargaining, 1949 Australia (ratification: 1973) (1998) <


[271] See BHP Iron Ore Pty Ltd v Australian Workers’ Union [2000] FCA 430; (2000) 102 FCR 97; David Noakes and Andrew Cardell-Ree, ‘Individual Contracts and the Freedom to Associate’ (2001) 14 Australian Journal of Labour Law 89; Sarah Richardson, ‘Freedom of Association and the Meaning of Membership: An Analysis of the BHP Cases’ [2000] SydLawRw 20; (2000) 22 Sydney Law Review 435. Richardson observes that lack of a right to collectively bargain renders freedom of association illusory: ‘[a] union without the ability to bargain collectively is like a football club that is not allowed to play games’: at 446. Further, an employer can make an offer of employment conditional on that employment being covered by an Australian Workplace Agreement (‘AWA’) rather than a collective agreement: Burnie Port Corporation Pty Ltd v Maritime Union of Australia [2000] FCA 1768; (2000) 103 IR 153, 158.

[272] As regards the government, this is evidenced in its promotion of AWAs through the Office of the Employment Advocate. In the case of corporations, it is evidenced by the stance of BHP Billiton (and its subsidiary, BlueScope Steel): see above n 107.

[273] See, eg, Gunningham and Johnstone, above n 197, 79–92; Baldwin and Cave, above n 162,


[274] See, eg, Jorgensen et al, above n 94, 55. With A$16 billion per year of purchasing power, the government can significantly influence firm behaviour through its procurement policies. Those policies already require environmental considerations to be taken into account, but do not appear to cover foreign labour issues: see, eg, Department of the Environment and Heritage, Procurement <> . State governments can also have a very significant impact through their procurement policies.

[275] Cf Gunningham and Johnstone, above n 197, 90.

[276] A further problem is that incentives may spark even more qualifying rules — a return to the disadvantages of command and control: Baldwin and Cave, above n 162, 43.

[277] A related — highly controversial — measure was the requirement introduced by the former Dutch government that linked export credits and insurance guarantees to the OECD Guidelines. To qualify for the government benefit, companies were required to state that they were aware of the Guidelines and would endeavour to comply with them to the best of their ability: OECD, OECD Guidelines for Multinational Enterprises: Focus on Responsible Supply Chain Management, Annual Report 2002 (2002) 25, 61. The OECD and business groups oppose a linkage that would render the OECD Guidelines compulsory: at 22, 24.

AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback