ilologo2.gif (2827 bytes)International
Labour Organization

logactra.gif (6461 bytes)Bureau for
Workers' Activities


Hoe Lim



1. The Global Economy: Trends, Impact and Implications

1.1 Liberalisation of International Trade and Capital Flows

1.2 The Global Factory - TNCs, EPZs and Foreign Direct Investment

1.3 The Global Labour Market: Opportunity or Threat?

2. The Role of Labour Standards in a Global Economy

2.1 ILO Conventions and International Labour Standards

2.2 Workers’ Rights Are Human Rights

3. The Social Clause: What Is It All About?

3.1 What Is a Social Clause?

3.2 Which Labour Standards Are Being Proposed?

3.3 Refuting Some Popular Misconceptions about the Social Clause

4 The Social Clause: Is It Justified?

4.1 Economic Arguments For and Against the Social Clause

4.1.1 Conceptual Debates: The Neo-Classical vs. the Neo-Institutional

4.1.2 Economic Grounds for Core Labour Standards

4.2 Some Empirical Findings on Core Labour Standards and the Social Clause

4.2.1 Output, Productivity and Wages

4.2.2 Effects on Trade

4.2.3 Effects on Foreign Direct Investment

4.3 Core Labour Standards: Obstacles to or Catalysts of Development?

4.4 The Role of Trade Unions

4.5 From Human Rights and Political Economy Perspectives

5. How Would a Social Clause Work?

5.1 Some Social Clause Type Mechanisms

5.1.1. Labour Standards in US Trade and Investment Legislation

5.1.2 The NAFTA Side Agreement on Labour

5.1.3 Non-US Initiatives, the Lomé Convention and the EU GSP System

5.1.4 International Commodity Agreements

5.1.5 Voluntary Codes of Conduct and Social Labelling

5.2 Labour Standards and Trade Disciplines

5.2.1 The WTO: a New Multilateral Trading System

5.2.2 The Trade Policy Review Mechanism

5.2.3 The Dispute Settlement Procedures

5.2.4 Articles of GATT 1994 Pertinent to the Social Clause

5.2.5 Which Option Is the Best for Making the Link?

5.3 How Can the ILO and WTO Work Together?

5.3.1 The Trade Union Proposal

5.3.2 Joint ILO/WTO Implementation of a Social Clause

6. Conclusion




For many workers, globalization and flexible labour markets mean having to work under inhumane conditions, to forego their basic human rights, and to take whatever remuneration they are given no matter how unreasonable it might be.

The conclusion of the Uruguay Round in 1994 marked a new era in the multilateral trading system. As traditional barriers to trade come down, greater emphasis is being placed on developing safeguards against potentially negative economic and social repercussions of a global economy. Similarly, attention is also being drawn to the direct and indirect effects of non-trade related national legislation and policies on cross-border transactions in goods and services, and flows of foreign direct investment. It is within this context that the nineteenth-century debates on the relationship between international labour standards and trade are increasingly being revisited.

For some, this relationship is about ensuring that countries with lower labour standards do not gain an ‘unfair advantage’ in trade at the expense of countries with higher labour standards. Debates on this point have been focused on whether lower labour standards actually confer a real economic advantage in trade. Unfortunately, by concentrating on the perceived international trade and investment effects of labour standards, such debates miss the point. The relationship between international labour standards and trade should be viewed from a human rights perspective. Within this context, the issue is how international trade and investment are affecting labour standards, rather than vice versa.

For the most part, labour codes governing pay, working-time, health and safety, welfare and social security are positively related to economic development. Therefore, it is natural that countries at different levels of development will be able to afford different levels of pay and benefits. Indeed, the Conventions of the International Labour Organization have in-built flexibility clauses to accommodate such differences. The international labour standards which have been proposed for the social clause are not about setting international wage levels or benefits. Rather, the social clause will only contain core labour standards that embody human rights, as defined by the United Nations Universal Declaration and Covenants on Human Rights.

The social clause is not an abstract or fashionable idea. It is about ensuring that working people have their fundamental human rights honoured everywhere. It is about counteracting the ways in which the liberalisation of international trade and investment has undermined the bargaining power of labour and encouraged the exploitation of workers. The rights of workers to basic human rights of the type embodied by the social clause should not depend on a country’s wealth, any more than the rights to freedom of speech or freedom of religion. A social clause is not about fixing global minimum wages. No one is arguing that workers’ wages and benefits should not vary according to a country’s level of income. However, workers have the right to bargain on relatively equal terms for whatever conditions and benefits are appropriate to a given market situation.



1.1 Liberalisation of International Trade and Capital Flows

Globalization has been the dominant feature of the world economy over the past two decades. The globalization process started in the industrialised countries with the deregulation of financial markets in the 1970s. It gathered pace in the 1980s with the virtual abolition of exchange controls among most industrialised countries, resulting in the rapid growth of financial markets -circulating vast amounts of money - often beyond the regulatory control of public monetary authorities. According to a recent study by the United Kingdom Centre for the Study of Financial Innovation, the $215 billion of gold reserves held by the central banks of the world’s leading economies are of virtually no value in interventions in the foreign exchange markets.

In the 1990s, the conclusion of the Uruguay GATT Round and the continued expansion and liberalisation of financial markets have made the drive for global integration virtually unstoppable. In 1990, volumes of goods and services traded across borders accounted for about 45 % of world GDP, up from 25 % in 1970. These trade flows are likely to increase in the future with the implementation of the Uruguay GATT Round, the most ambitious multilateral trade agreement ever negotiated, resulting in a 40 % cut in industrialised countries' tariffs on manufactured goods, tariffication of all non-tariff barriers on agricultural products and a new agreement on trade in services..

At the same time flows of foreign direct investment (FDI) have also increased sharply, with average annual flows increasing more than threefold since the 1980s for the world as a whole.. Other financial flows have also increased with massive growth in the volume of trade in foreign exchange, new financial instruments such as derivatives, and cross-border trading in bond and equity markets.

The lifting of trade and capital barriers, technological advances, and continually falling communications and transportation costs are rapidly transforming the nature of competition and production. The homogenisation of consumer preferences, product standards and production methods; the explosion of national and international mergers and acquisitions; and the increasingly ‘footloose’ nature of large corporations, willing to locate anywhere to maximise profits, are important features of this transformation. At the same time, the removal of external barriers are blurring the boundaries between issues of domestic and international concern. In particular, there is a growing realisation that issues of social policy, especially labour standards, can both affect and be affected by international trade and investment flows.

A number of factors give rise to this concern. First, due to perceived cost advantages, trade and investment could shift in favour of countries with lower labour standards. Second, this creates an incentive for countries to compete in the global economy by either reducing or suppressing their labour standards. Third, such competition is economically, socially and politically undesirable as it could lead to a ‘race to the bottom’. In the following sections, globalization, the implications for international competitiveness, and the reasons why these trends highlight the importance of a link between labour standards and international trade are discussed further.

1.2 The Global Factory - TNCs, EPZs and Foreign Direct Investment

One of the driving forces behind market globalization has been the rapid growth of transnational companies (TNCs). The number of TNCs rose to 37,000 in 1992 and they collectively account for about one third of international trade and of the world's private productive assets. By the end of 1993, the sales by foreign affiliates of TNCs totalled $5.8 trillion. While the vast majority of TNC employment is located in the industrialised countries, developing countries are beginning to gain a larger share with 5 out of the 8 million jobs created by TNCs between 1985 and 1992 located in developing countries. While OECD countries still account for the majority share of world foreign direct investment, during the 1980s the stock of foreign direct investment (FDI) in the developing world rose threefold and FDI inflows have been even more rapid during the 1990s.

An important feature of a TNC is its ability to shift production where market and costs advantages are most favourable. Moreover, different parts of a TNC production process may be performed by subsidiaries or subcontractors in different countries, resulting in increased foreign investment, international specialisation and higher levels of intra-firm trade. Often, the final product is a complicated bundle of inputs, sourced and produced in a variety of locations, for sale in host or home country or anywhere else in the world.

Due to the scale of their operations, potential for employment generation and technological spin-offs, countries have sought to create conditions that are perceived to attract TNC investment. These conditions typically involve a combination of tax breaks, subsidies, availability of infrastructure and access to domestic markets. In addition, many countries have also - either by law or by practice - suppressed or lowered labour and other social standards that companies have to observe. Increasingly, the combination of flexible production processes, the economic strength of TNCs, the desire to attract FDI, and the view that what is good for TNCs is also good for national economies is resulting in ‘regulatory competition’ between countries.

The proliferation of economically, socially and physically separate enclaves of export processing zones (EPZs) is one of the most visible results of such competition. In 1995, there were more than 230 EPZs spread across nearly 70 countries, including more than 100 in Latin America and the Caribbean, 64 in Asia and 31 in Africa. Employment in these zones, the bulk of which is for TNCs, grew by 9 % between 1975 and 1986 and by 14 % between 1986 and 1990. Estimates suggest that there are presently about 1.2 million Latin American and Caribbean, 3 million Asian and 250,000 African workers employed in these zones. In addition, there are 14 to 40 million workers in China's Special Economic Zones. While EPZs are a global phenomenon, the bulk of investment in EPZs is concentrated in a few countries, with 14 out of a total of 45 countries accounting for 94.5 % of total employment in these zones globally.

The original idea behind the creation of EPZs was to allow enterprises to import materials for processing and to have the final product re-exported without having to pay duty. However, as the EPZ concept spread around the world more incentives were added, including guarantees of low-cost workers with few or no labour rights. While EPZs have created employment, these are usually badly-paid jobs often carried out under degrading and inhuman conditions. In some countries, the exploitation of workers in EPZs is supported by special legislation restricting freedom of association, collective bargaining and the right to strike (see Box 1).

Box 1: Some examples of special legal restrictions on labour rights in EPZs

Turkey: The Free Trade Zone Act forbids workers to strike during the first 10 years of operation of an EPZ, disagreements must be settled by binding arbitration.

Bangladesh: The 1969 Industrial Relations Ordinance does not allow workers in EPZs to organise in trade unions and to bargain collectively.

South Korea: Until the advent of democratisation in mid-1987, Korea legally restricted the right to form unions and bargain collectively in its two EPZs. While restrictions were lifted in 1987, the Korean government re-imposed restrictions on strikes in August 1989 by ruling that EPZ firms are public interest companies. This effectively restricts the right to strike and requires mandatory arbitration in almost all cases.

Pakistan: Section 25 of the Export Processing Zones Authority Ordinance, 1980, which provides that the Government may exempt EPZs from the application of labour laws. The Government has used this provision to exempt EPZs from the Industrial Relations Ordinance (IRO) of 1989 which includes the right of workers to establish and join trade unions. Furthermore, Section 4 of the Export Processing Zone (Control of Employment) rules of 1982 deprives workers employed in EPZs of the right to strike.

Source: US Department of Labour Bureau of International Labour Affairs (1990), Workers’ Rights in Export Processing Zones.

In the majority of cases, despite the applicability of national labour legislation, government officials, zone administrators and employers have been found to collude to prevent trade union activity (see Box 2).

Increasingly powerful and mobile TNCs are more than ever able to play one country against another in search of the least onerous regulatory environment. The long-term value of these ‘footloose’ investments is also questionable as the companies concerned are prone to leave at the slightest economic shock or change in the regulatory environment, and are unlikely to develop strong links with the rest of the economy. This problem is especially acute in low-skill industries such as garments and footwear, where firm-specific knowledge is slight and exit costs are low. Moreover, apart from a handful of Asian countries - namely Singapore, Hong Kong and Malaysia - the investments made in EPZs have generally failed to generate the expected benefits of economic growth, human resource development and technology transfer.

It is important to note that there are viable alternatives to the prevailing EPZ model of cheap labour and trade union repression. Mauritius, which has used its EPZs to develop an impressive industrial base, has emphasised the improvement of skills, working conditions and pay of its workforce. A director of a promotional agency for the zones has stated , "It is not our aim to attract foreign investors with cheap labour. In the long term, it is always counter-productive, because unskilled workers produce goods with no value added which do not sell well on the export market".

Box 2: Examples of violations of trade union rights in EPZs despite the existence of labour legislation

Philippines: Although labour legislation applies in the country’s EPZs, in practice a "non-union, no strike" policy exists in most of the zones. In order to prevent unionisation, employers intimidate workers with threats of dismissal and closure. Local government officials and zone administration prevent group meetings, union organisers are not allowed, and workers are closely monitored.

Sri Lanka: Failure to enforce labour laws and standards is particularly flagrant in the country’s EPZs which employ around 95,000 workers, mainly women. Union organisers encounter severe difficulties in gaining access to tightly guarded zones where there are few or no unions. Reports have emerged of police preventing workers from gathering, even on private premises. In addition, employers hire security officers to monitor workers. In the place of trade unions, wages and working conditions are set by Joint Consultative Councils, chaired by the Government’s Board of Investments (BOI). This effectively gives the Department of Labour full control of industrial relations and disputes in the zones. It has been reported that at the request of employers, BOI officials intervene in industrial disputes, and often collude with police if a strike takes place.

Dominican R: While the 1992 labour code guarantees the right to organise and protects against acts of anti-union discrimination in the country’s 32 EPZs, employers continue to ignore it. There were some improvements after 1994, when the US Government threatened to suspend trade privileges. Nevertheless, by early 1996 only eight of the 114 registered unions in the EPZs were able to function and only four collective agreements had been signed. Working conditions in the zones are reported as being particularly bad. Female workers were physically abused; there were two-way mirrors in bathrooms to spy on employees; tickets to control access to the toilets; forced overtime with no toilet facilities; unsafe work practices and conditions; and intimidation. Protests were met with threats to close factories and leave the country.

El Salvador: Although protection against anti-union discrimination exists in law, and employers are obliged to re-hire employees fired for trade union activity, this has not been enforced. Government action against employers for dismissing trade union members and assaulted workers is ineffective because of the inadequacies of the legal system, and a lack of will on the part of the authorities fearful of losing investors.

Honduras: Although the labour code applies to the country’s EPZs which employ around 40,000 workers, anti-union discrimination and assaults on workers are widespread. Degrading working conditions and abuse of mainly women workers in EPZs led to numerous strikes in 1994. In some cases, workers unable to keep up with the pace of work were forced to hold chairs above their heads for up to one hour. They were dismissed if they let their arms down during this period. It is also reported that some enterprises had given workers amphetamine injections to force them to work 48-hour shifts and of supervisors patrolling with heavy clubs, arbitrarily beating workers. On 18 August 1995, workers at King Star Garments in the Buffalo free trade zones were ordered to report to work six hours early. On arrival 25 workers, mostly leaders of the emerging union, were taken to the company’s offices. The remaining 500 workers were locked inside the factory. They were ordered to leave a few hours later but refused, as management gave no explanation about their 25 colleagues. During the night, 50 security guards attacked the workers with clubs and tear gas, and fired shots. Honduras and the US signed an agreement at the end of 1995 to improve trade union rights and working conditions for the 75,000 workers in the privately-owned free trade zones. The agreement set out measures to ensure that Honduras complied with the workers’ rights provisions in the US Generalised System of Preferences, which grants countries preferential trade access to the US.

Source: ICFTU, Annual Survey of Violations of Trade Union Rights, Brussels (various editions).

1.3 The Global Labour Market: Opportunity or Threat?

Although labour has remained relatively immobile, a global labour market is emerging with a growing proportion of the world labour force engaged in activities linked to international trade and capital flows. Globalization has also internationalised subcontracting arrangements with an ever increasing number of small and medium-sized enterprises involved in international production. In this sense, labour markets across the world are becoming more interlinked. Production involving the exchange of labour services of workers in a number of different countries is thus becoming increasingly common:

  • Consider, for example, a British entrepreneur who hires an Italian company to design a new line of clothing, then has those designs sent for production in Southern China, and has a shipping company in Hong Kong send the finished product for sale in the United States. Without the entrepreneur or any worker having to cross a national border.
  • One of the negative consequences of globalised labour markets is that the world has become a huge bazaar with nations peddling their workforces in competition with one another, offering the lowest prices for doing business. In addition to regulatory competition between governments, globalization has also generated pressures on enterprises to adopt cost-minimisation strategies with negative consequences for workers. Corporate restructuring and downsizing are examples of such strategies. At the same time, the bargaining power of labour has weakened, as employers can now more easily threaten domestic workers with either relocation or substitution by foreign workers.

    While conventional economic analysis maintains that the benefits of globalization exceed its costs, it is noticeable that employment conditions continue to deteriorate in many parts of the world. In most industrialised countries, unemployment rates have been on the rise since the early 1970s. In the former centrally planned economies, the transition to an open market economy has been accompanied by the rapid emergence of mass unemployment. Many developing countries in sub-Saharan Africa and Latin America are still struggling to emerge from the economic crisis of the 1980s and have so far been excluded from the benefits of globalization. Around the world, income and wage inequalities continue to expand, with societies polarised between those who have the wealth or skill to gain from global integration and those who remain trapped in poverty without productive employment or basic labour standards.

    It is time to recognise that while globalization may benefit the world economy as a whole, there are widespread negative consequences which cannot be ignored. Compromising the goals of social justice and full employment will ultimately undermine political support for a global economy and raise feelings of economic nationalism. As highlighted by Kapstein:

  • Populists and demagogues of various stripes will find "solutions" to contemporary economic problems in protectionism and xenophobia. Indeed, in every industrialised nation, such figures are on the campaign trail. Growing income inequality, job insecurity, and unemployment are widely seen as the flip side of globalization. That perception must be changed if Western leaders wish to maintain the international system their predecessors created. After all, the fate of the global economy ultimately rests on domestic politics in its constituent states.
  • Countermeasures and safeguards which can ‘civilise’ international competition, ensure a fairer distribution of wealth and ease the pressures on the ‘losers’ of globalization are not only socially desirable but also economically and politically necessary. International action to safeguard basic labour standards is central to maximising the gains of a global labour market, while minimising its costs. Lowering labour standards is neither the only nor the best way to deal with increased international competition. Enhancing labour productivity through investment in skills development, exploiting the productivity-raising potential of high labour standards and co-operative forms of work organization, and productivity-augmenting investments in infrastructure and research and development are both viable and desirable alternatives.



    2.1 ILO Conventions and International Labour Standards

    ILO Conventions constitute the most comprehensive set of international labour standards, covering virtually all aspects of working conditions and industrial relations including: fundamental human rights; employment; social policy; labour administration; industrial relations; conditions of work; social security; employment of women; employment of children and young persons; old workers; migrant workers; indigenous workers and tribal populations; and particular categories of workers (notably seamen and fishermen).

    Once ratified, these Conventions create binding obligations for member states, that is, they must be introduced into national legislation. The ILO has a well-respected supervisory mechanism for monitoring the application of ratified Conventions in national law and practice. Each member country is required to submit to the ILO reports on measures taken to implement, in law and in practice, the Conventions which it has ratified. This machinery also includes an adjudication process which permits complaints to be made by non-governmental organizations, namely trade unions and employers’ associations. A special procedure exists for investigating alleged infringements of freedom of association by governments regardless of whether the respective Conventions have been ratified or not. Member States are bound to respect the fundamental principle of freedom of association, by virtue of their membership to the ILO.

    While ILO Conventions are not ranked in terms of their order of importance, there is an underlying hierarchy which can be discerned. In the first category are Conventions dealing with freedom of association and collective bargaining (Conventions 87 and 98), forced labour (Conventions 29 and 105), non-discrimination in employment (Conventions 111) and child labour (Convention 138). These core Conventions were identified and given prominence in the conclusions of the World Summit for Social Development in 1995. In the second category are technical standards which establish norms to improve working conditions, for instance, on matters such as occupational safety and health.

    The ILO Constitution and its Annex (the Philadelphia Declaration of 1944) makes an important distinction between countries at different levels of economic development. The Declaration in Part V stipulates that while fundamental principles on which the ILO is based are fully applicable to all peoples everywhere,

  • the manner of their application must be determined with due regard to the stage of social and economic development reached by each people.
  • Subsequent reviews by ILO working parties have upheld the principle of ‘flexibility’, noting that international labour standards have to represent realistic targets with due regard given to countries at different levels of development and to ensuring adaptation to their special needs. However, it is emphasised that differences in economic development do not excuse violations of the fundamental conventions dealing with freedom of association and collective bargaining (Conventions 87 and 98), forced labour (Conventions 29 and 105), non-discrimination in employment and wages (Conventions 111 and 100) and child labour (Convention 138).

    2.2 Workers' Rights Are Human Rights

    These seven ILO Conventions are widely regarded as intrinsic elements of universal human rights as defined by the UN Universal Declaration of Human Rights (UDHR) (1948), the International Covenant on Civil and Political Rights (ICCPR)(1966) and the Covenant on Economic, Social and Cultural Rights (ICESCR) (1966). In the Covenant on Economic, Social and Cultural Rights, provisions concerned with labour include the right to:

    In addition, Article 8 of the Covenant on Civil and Political Rights prohibits slavery, servitude and other forms of forced and compulsory labour . The recent UN Convention on the Rights of the Child (1989) also contains provisions on labour, with Article 32 establishing the right of the child to be protected against economic exploitation and dangerous or unhealthy work, or work that would hamper education prospects. Article 32 also requires countries to determine a minimum age for child labour and to regulate working conditions, including hours of work.

    It is generally accepted that the principles established in these instruments are to be interpreted in the light of the corresponding and far more detailed ILO Conventions. The fact that the rights contained in ILO's Conventions on freedom of association, collective bargaining, forced labour, non-discrimination and child labour are also enshrined in the widely ratified UN human rights instruments supports the view that these core labour standards are synonymous with basic human rights. This relationship is important because the body of international law on human rights considers the principles of freedom of association, prohibition of forced labour, elimination of child labour and non-discrimination as being universal, transcending all political, economic, social and cultural situations. This notion of universality is supported by the fact that the UN Covenant on Economic, Social and Cultural Rights has been ratified by 131 countries, the Covenant on Civil and Political Rights by 129 and the Convention on the Rights of the Child by 168. In addition, Article 55 of the Charter of the United Nations of 1945 states that countries should promote respect for human rights and basic liberties for all, without distinction of race, gender, language or religion. All member States to the UN are bound to this provision by virtue of their membership to the organization.


    3.1 What Is a Social Clause?

    In the context of international trade, a social clause essentially refers to a legal provision in a trade agreement aimed at removing the most extreme forms of labour exploitation in exporting countries by allowing importing countries to take trade measures against exporting countries which fail to observe a set of internationally agreed minimum labour standards. The trade measures may include:

    While the focus is presently on trade measures, social clause provisions have been linked to non-trade arrangements. For instance, the US Overseas Private Investment Corporation (OPIC), a government agency which offers insurance to US companies operating in developing countries, will withdraw its services from projects in countries not taking steps to adopt and implement laws that extend a set of internationally recognised labour standards. It has also been suggested that social clauses should also be added to development aid and loan programmes.

    At present, while the ILO actively promotes the ratification and supervision of Conventions, it cannot force compliance or impose financial, commercial or other sanctions; rather it relies on persuasion and peer pressure to encourage States to meet their obligations. Nevertheless, despite not having a punitive enforcement mechanism, the ILO has in practice attained an influence that goes beyond legal formality.

    3.2 Which Labour Standards Are Being Proposed?

    Past discussions concerning the social clause, especially the terminology used, often lacked clarity. Terms such as ‘workers' rights’, ‘labour rights’, ‘social rights’ or even ‘social dumping’ were often not clearly defined. As a result, there was much confusion over the nature of the labour standards which would be covered by a social clause. A clearer understanding of what is to be contained in a social clause and how they will be interpreted and applied is important in allaying fears of abuse by protectionists.

    Most proposals for a social clause are based upon the aforementioned seven ILO Conventions. These are listed below:

    Table 2.1: Number of countries that have ratified core conventions

    ILO Conventions

    Number of countries (ratification)

    Freedom of Association and Protection of the Right to Organise, 1948, (No.87)


    Right to Organise and Collective Bargaining, 1949, (No. 98)


    Forced Labour Convention, 1930, (No.29)


    Prohibition of Forced Labour, 1957 (No.105)


    Equal Remuneration Convention, 1951 (No.100)


    Discrimination (Employment and Occupation), 1958, (No.111)


    Minimum Age Convention, 1973 (no.138)


    Source: ILO, Report III (Part 2), International Labour Conference 1997.

    The above table indicates the number of ratifications as at. 31 December 1996 These Conventions are included in virtually all proposals for a social clause for a number of reasons:

    The high level of ratification suggest that the underlying principles in these Conventions command widespread agreement. In the case of Conventions 29, 87, 98, 111 and 105 practically no country opposes ratification on grounds of principle. Reasons for non-ratification tend to refer to specific details of the Convention or their interpretation by ILO bodies:

    3.3 Refuting Some Popular Misconceptions about the Social Clause

    Although a number of studies have recently been published on the potential effects of linking trade and labour standards, there remain considerable misconceptions and misinterpretation of its scope and implications. In particular, there is a tendency for those who oppose the social clause to assert - wrongly - that there is no agreement on what constitutes core labour standards, that it will establish a global minimum wage, retard economic development, destroy the world trading system and undermine national sovereignty. Each of these and other misconceptions concerning the social clause are briefly examined:

    There is no agreement on what constitutes core labour standards: There is actually more agreement than there is disagreement on what constitutes core labour standards. As discussed in the previous section, core labour standards represent basic human rights and should not be confused with labour codes governing pay, working-time, health and safety, welfare and social security, etc. There is general acceptance among Member States of the UN, even those that have not acceded to the ICCSPR and ICESCR, of the universality of the human rights embodied in these core labour standards.

    A global minimum wage will be set: Core labour standards do not set minimum wages, but ensure that workers have a say in what their conditions of work and remuneration should be. Moreover, ILO Conventions do not attempt to impose a rigid global harmonisation of labour laws. As levels of productivity, price structures and the supply of labour vary so dramatically between countries, it is inconceivable that greater respect for core labour standards such as freedom of association would lead to a global minimum wage. In most developing countries that have low labour standards, there is virtually an "unlimited supply of labour. In these situations, trade unions have little or no opportunity to raise wages above the market clearing rate. If anything, globalization by bringing labour markets together will create an even larger labour pool and thus further strengthen the ability of employers to force wages down, potentially to a level below the optimal market clearing rate. Respect for core labour standards such as freedom of association and collective bargaining would be an important counterbalance and help achieve an efficient market outcome.

    Low wages are the key to economic success: If low wages automatically translate into economic success, the world’s poorest countries would be economic superpowers. They are not, because differences in wages reflect differences in productivity. Low wages tend to go hand in hand with low productivity. The task for countries is not simply to keep wages low but to raise productivity. Here, labour standards play an important role in enhancing productivity. A recent study by Aggarwal provides some evidence that the most successful exporters in the ten developing countries studied did not suppress core labour standards to reduce production costs.

    Table 2.2: Global labour costs - hourly compensation in manufacturing ($)

















    Other Asia




    Latin America




    Source: Financial Times 22 October 1996, p.12

    Similarly, a recent study by the World Bank and earlier work by Amsden and Lim all point to the general conclusion that the economic performance of the East Asian newly industrialising countries (NICs) has not been due either to the brutal exploitation of labour or to downward real wages . Of greater importance have been the productivity gains from education, infrastructure development and the restructuring of markets from low-wage labour-intensive to higher value-added and skill-intensive techniques. As expressed by a representative of Citicorp at the 88th Annual Meeting of the American Society of International Law:

  • In the long run, an educated and healthy workforce is necessary for continued economic development. It is logical for business to encourage government policies that support these goals. Employees whose basic nutritional and medical needs are met will be better and more productive workers. Company policies that assure adequate wages, workplace access to a good meal and - very importantly - superior local health care, will all support this objective. Employees who do not have to live in fear of a battered-in door in the middle of the night, or of speaking freely to one another at appropriate times during the work day, will be more focused on the profitable task at hand. Company policies that refuse to co-operate with the presence of political or government "monitors" in the workplace will help ensure this focus.
  • The comparative advantage of developing countries will be eroded: It is neither conceptually nor empirically clear that greater adherence to core labour standards result in higher labour costs. A recent OECD study found that improving adherence to labour standards does not necessarily result in a significant increase in total costs in low-standard countries and could in fact improve labour market efficiency and worker productivity. The study also found that most low- to medium-income countries exhibited export dynamism, irrespective of their labour standards. Greater adherence to core labour standards is likely to facilitate human capital development, which is one of the key factors in achieving long-term growth.

    Sanctions will be used indiscriminately: If a social clause is to work, it must follow a step-by-step procedure which is open, fair and multilateral, and which allows problems to be resolved through negotiation. Within such a framework, there would be many safeguards against the indiscriminate use of sanctions. One of the most important safeguards would be the need for multilateral agreement on the use of sanctions. It is likely that within such a framework sanctions would be used only as a final measure, when all other avenues have failed to persuade the offending party to address its violation of core labour standards. It should be noted that the US already has domestic legislation which could be used to impose sanctions on countries which fail to respect certain core labour standards. It is likely that support for such unilateral measures, which are more prone to misuse, will intensify unless agreement is reached to establish a social clause on a multilateral basis.

    Protectionist tendencies will be supported: Cheap manufactured goods from the developing countries are increasingly being blamed for job losses in the industrialised countries, hence the ever-growing support for populist demands for protection in one form or another. As unemployment worsens in the industrialised countries, there is likely to be increased public pressure to restrict the flow of what are perceived to be ‘unfairly’ produced goods. The US, and increasingly the EU, already have legislation which could be used to enforce such restrictions on a unilateral basis. Whether these allegations have an economic basis or not, they pose a serious threat to the world trading system, with greater use of unilateral measures leading to an eventual erosion of the rule of law in international trade and potentially to the establishment of even higher levels of protection. A social clause could rein in these unilateral measures and provide a more transparent and equitable means for settling disputes. A social clause could thus lower rather than raise protectionist tendencies and help keep markets open by sharing out the benefits of trade more fairly.

    The introduction of a social clause is a threat to national sovereignty: While there can be no compromise on the principles established by core labour standards, how they are translated into law and practice may vary according to the institutions and customs of the country concerned. Moreover, by virtue of being members to the United Nations, States have already committed themselves to uphold basic human rights. These are in most respects synonymous with core labour standards (see section 2.2). The social clause would not create any new human rights obligations but would strengthen the existing international enforcement mechanism. At the same time, the sovereignty of even major industrial countries on economic and social policy matters is being increasingly undermined by globalization. An internationally agreed social clause would strengthen national sovereignty by making countries less vulnerable to manipulation by powerful private economic actors, and would enhance their ability to pursue national economic and social objectives.

    The argument of sovereignty is clouded by the new global circumstances. The position that sovereignty should not be ceded on issues such as [domestic] labour policy regulation [of international processes] needs to be re-assessed. The reality is that sovereignty has already been ceded and policy is being shaped to an increasing extent by the demands of the global market. The issue is not whether but how policy will be determined internationally - through the market or through political negotiations. As stated by Langille, the problem is no longer avoiding a potential loss of sovereignty, but whether to take an opportunity to reclaim some measure of it.

    National culture will be threatened: The notion that the social clause threatens national culture is a myth created by vested interests; in fact it is the pervasive commercialism of the free market which is the real threat. National culture cannot survive where people have no power to control the market for their own and their community's benefit. The social clause by promoting respect for human rights would effectively empower people and their communities to maintain their cultural identity in a global economy.

    Human rights and labour standards are culturally specific: Numerous contemporary arguments against universal human rights, and by association international labour standards, hide behind the shield of cultural relativism but are often not supported by any discernible cultural basis. Arguments of cultural relativism tend to be made by economic and political elites. The very same elites who raise culture as a defence against external criticisms based on universal human rights often ruthlessly suppress inconvenient local customs, whether of the majority or the minority. All too often, "..leaders sing the praises of traditional communities - while they wield arbitrary power antithetical to traditional values, pursue development policies that systematically undermine traditional communities, and replace traditional leaders with corrupt cronies and party hacks. Such cynical manipulation of tradition occurs everywhere".

    For the most part, in undemocratic or closed political systems, it is only the views of the ruling elite which are given wide recognition. This should not be mistaken for an unchallenged consensus. For instance, prior to the Second World Conference on Human Rights in Vienna, 1993, NGOs from the Asia-Pacific Region gathered at Bangkok to demonstrate that the position of their governments on cultural relativism did not necessarily have popular support. In the Bangkok NGO Declaration, it is stated that

  • We can learn from different cultures in a pluralistic perspective....Universal human rights are rooted in many cultures. We affirm the basis of universality of human rights which afford protection to all of humanity....While advocating cultural pluralism, those cultural practices which derogate from universally accepted human rights, including women's rights, must not be tolerated. As human rights are of universal concern and are universal in value, the advocacy of human rights cannot be considered an encroachment upon national sovereignty.
  • This position is in line with the Vienna Declaration adopted by States at the Second World Conference on Human Rights, which provides in Sec.I, para.5 that

  • All human rights are universal, indivisible, and interdependent and interrelated.....While the significance of national and regional particularities and various historical, cultural and religious backgrounds must be borne in mind, it is the duty of States, regardless of their political, economic and cultural systems, to promote and protect all human rights and fundamental freedoms.
  • Child labour is irreplaceable: Common explanations given for using children are that they incur lower costs and have irreplaceable skills (‘nimble fingers’). Based on this, it is often argued that a social clause to eradicate exploitative forms of child labour is unrealistic. In fact these two claims are often unsustainable. ILO studies have shown that the notion that only children with small fingers have the necessary skills to work in a number of industries, especially the production of fine hand-knotted carpets, is simply not true. In virtually all of these industries, adults can be found working side by side with children performing unskilled tasks. In an empirical study of over 2,000 weavers it was found that children were no more likely than adults to make the finest knots. Some of the best carpets with the greatest density of small knots are in fact woven by adults. Indeed, not only can adults easily replace children, they may also be more skilful in performing delicate tasks.

    In most cases, child workers are paid less than their adult counterparts. But the lower wages and other advantages claimed for child labour are not always as clear and compelling as is suggested. Recent ILO studies in India estimate that labour cost savings resulting from the employment of children, calculated as a proportion of the final price of carpets or bangles, are surprisingly small. These calculations show that the savings are less than 5% for bangles and between 5 and 10% for carpets . The problem is that the loom owners, who are usually poor small contractors themselves, are working to very slim profit margins and have an incentive to seek these marginal cost savings. It would thus only take a very small levy on the consumer price to subsidise the cost of using exclusively adult labour.

    These findings, from an extremely competitive, labour-intensive industry, thought by some to be among the most dependent on child workers, raise serious doubts that any industry at all has to depend on child workers in order to be competitive. However, as noted by the recent ILO report on child labour, in a global market in which countries compete in producing similar products, abolishing such practices in only one country would simply transfer production to others that still employ it. The market for goods using child labour is international. Action to discourage it needs to encompass all the major producers so as to avoid ‘beggar-my-neighbour’ competition. Often, the major reason for hiring children are non-economic. Children are less aware of their rights, more willing to take orders and to do monotonous work without complaining, less likely to steal, and less likely to be absent from work.


    4.1 Economic Arguments For and Against the Social Clause

    Is there an economically convincing case for creating a social clause in international trade agreements? At present, answers to this question appear to be divided between two opposing schools of economic thought: the neo-classical and the neo-institutional. According to the neo-classical position, which is supported by many economists, there are neither theoretical nor empirical grounds for enforcing core labour standards through international trade agreements. Those from the neo-institutional school on the other hand argue that markets, both domestic and international, have to operate within a framework of regulations which establish transparent and enforceable rights, standards, obligations and dispute procedures. They assert that without such a framework, market mechanisms can have potentially destructive outcomes which conflict with economic and social objectives.

    This conceptual debate between the neo-classical and neo-institutional schools is important because critics of the social clause are only questioning how international labour standards are enforced but are casting doubt on whether labour standards serve a useful purpose. Given the prevailing trend of uncritically accepting free market solutions, it is important to demonstrate that a laissez-faire attitude to issues such as labour standards is unrealistic and untenable.

    4.1.1 Conceptual Debates: The Neo-Classical vs. the Neo-Institutional

    In reflecting on the conceptual economic debate over the social clause, it is useful to recall Feis’s classic essay ,"International labour legislation in the light of economic theory", and the more recent work by Langille, "Labour standards in the globalized economy and the free trade/fair trade debate". In his essay, Feis reflects on the apparently irreconcilable divergence between neo-classical economic theory, which teaches that:

  • the conditions of workers in any country depend upon the real income of the country and that even allowing for possible variations in sharing out of the national income, hours of work will be long, wages low and conditions burdensome, if the total income of the country is low relative to the number of inhabitants...
  • and the view that conditions of work can be improved by some form of joint international action. From the neo-classical economic perspective, which is taken by the majority of economists, free trade and an open market economy are the major preconditions which promote economic efficiency and thus enable countries to reach their highest possible level of prosperity.

    Within this paradigm, apart from the acknowledged instances of market failure, it is concluded that competition is best left unregulated as the market outcome will always be the most efficient. If this is accepted, there would appear to be little reason for either domestic or international labour standards, since regulations are perceived to interfere with market forces, impede efficiency, create sub-optimal labour allocations, and act as a deterrent to investments and a constraint on growth. Indeed, exponents of free market ideas would go as far to say that labour standards would facilitate the achievement of results that are actually antithetical to what is intended..

    Thus from the neo-classical perspective, labour standards, either national or international, cannot be justified on efficiency grounds. Indeed, to harmonise regulatory frameworks on labour and other matters, via multilateral agreements would be to argue against the rationale of trade itself, since national differences are seen to be a necessary requirement for international trade. This is clear from conventional economic theory on international trade, which basically states that each country should specialise in the production of goods in areas in which it holds a comparative advantage. If, as a result of foreign competition, an industry is rendered uncompetitive, workers and capital within the blighted industry would be expected to shift to other unaffected industries. The disruption caused would be transitory, since it is assumed that there are no barriers to such shifts taking place. . As noted by Figueroa,

  • neo-classical theory treats the labour market as equivalent to the potato market or the fish market, which are Walrasian markets. Labour markets always clear: no workers willing to work at the prevailing wage will be unable to find a job.
  • In reality, labour markets rarely clear. Indeed, the capacity for capital and labour to shift from one industry to others, under the pressure of competition, is often slow, painful and incomplete. There are often many structural impediments that make such shifts more than just a transitory difficulty. As a result, shifts in international competitiveness becomes less desirable, despite the potential efficiency gains.

    Thus it is not surprising that policy prescriptions offered by neo-classical economic analysis are often difficult to accept and often at variance with the wider objectives of public policy. Neo-classical economic doctrine postulates a highly improbable world where the only relevant concerns are the production and exchange of goods and services, in response to market signals. Within this world, rational or economically acceptable actions are judged only to be those which maximise the production of goods and services. Political, social and other factors which either complicate or contradict this view of the world are simply assumed away. Take, for instance, the following statement from a theoretical analysis of the relationship between trade and labour standards:

  • We look here at a world of many countries, each one of which is assumed to be too small to affect world prices. Each country interest is then its own policy choice......In this context, no country on economic grounds would care about the policy choices of other individual countries.
  • This typical assumption in neo-classical economic analysis completely misses the point, that it is not just changes in the world price which concern countries, but their relative shares in trade. Even if actions by individual countries do not change the world price, there may nevertheless be significant changes in the pattern of trade, resulting in either a relative gain or loss. In addition, countries are very much concerned on economic grounds, especially with the onset of globalization, about the policy choices of other countries.

    However, such realities are often not considered by neo-classical economic analysis, and established theoretical models are usually embraced dogmatically despite their increasing variance from reality. In the case of international trade, for instance, the world has been revolutionised by increased capital mobility and increased regulatory competition. These developments, combined with the relative immobility of labour and the absolute immobility of jurisdictions, call for an urgent reshaping of the simple model upon which established international trade theory is based. Neither Adam Smith nor David Ricardo envisaged that government regulatory policy could be a relevant factor endowment. Langille makes the important argument that, given these contradictions, the demand for multilateral agreements on labour standards should be seen as a natural and inevitable component of the policy of free trade.

    In brief, Langille shows that since the level of tariffs is now tending towards zero, the question arises as to the extent to which they will be replaced by differences in governmental regulation of competition as a cause of trade distortions. In this respect, he takes the view that there is no discernible difference between the effect of a tariff on an imported good, a direct subsidy for domestic employers, an easing of a requirement of employment law and a decision not to regulate. Should this be the case, neo-classical trade theory faces the question as to what constitutes a ‘natural’, ‘neutral’ or non-subsidising level of regulation. The neo-classical approach would propose international regulatory competition as the solution. This would essentially mean that the market would decide upon the optimal level of regulation.

    Langille demonstrates that this circular reasoning is faulty and illustrates how rationally-motivated, self-interested behaviour (i.e. the standard behaviour of players in the market place) can lead to socially sub-optimal results, and that these results can be avoided through cooperation rather than competition. International regulatory competition, where law is a product for which mobile capital shops, would result in a race to the bottom. Like the well-known concept of ‘prisoner’s dilemma’, international regulatory competition involves a strategic decision where the reward to each player depends upon the reward to all and the choice of each depends on the choice of all.

  • Assume that we have two jurisdictions with identical labour standards policies. These jurisdictions also seek to attract capital investment in the name of job creation and other benefits. Capital will shop for the jurisdiction which has the lower regulatory price. The two jurisdictions would be better off if they agreed to co-operate and not reduce labour standards from their current level. But each jurisdiction sees that, at least potentially, it is in its interest to reduce their labour standards with no net impact on investment. The outcome of this version of the prisoner’s dilemma is particularly striking - that even in circumstances of equal starting labour standards, jurisdictions will rationally engage in a race to lower standards.
  • This ‘prisoner’s dilemma’ scenario is well known to the world trading system. Indeed, the multilateral trading system administered by the WTO was constructed to deal with the ‘beggar-my-neighbour’ policies of the Depression years. As noted by Stein:

  • All nations would be wealthier in a world that allows goods to move unfettered across national borders. But any single nation, or group of nations, could improve its position by cheating - erecting trade barriers and restricting imports. The State’s position remains improved only so long as other nations do not respond in kind. Such response is, however, the natural course for those other nations. When all nations pursue their dominant strategies and erect trade barriers, however, they can engender the collapse of international trade and depress all national incomes. This is what happened in the 1930s and what nations wanted to avoid after World War II.
  • If non-market mechanisms are required to avoid the ‘prisoner’s dilemma’ in international trade, why should this not also be the case with labour? In the idealised world of the neo-classical economist the best policy may well be to let the market decide the optimal level of regulation. Policies, however, have to operate in the real world where rationality is conditioned by strategic factors. Put simply, if rational economic behaviour, the building block of neo-classical economic theory, can lead to undesirable results in international trade why should it not also create problems for what are increasingly international labour markets? If neo-classical economic theory is to have any relevance in today’s world, it has to come to grips with the growing gap between the highly stylised world of perfect competition and the reality of the global market place.

    Economists should also not forget that markets are there to serve the needs of society and not vice versa. All too often, success is measured only in terms of how well markets are functioning and not in terms of what the consequences have been on society as a whole. Even in cases where some consideration is given to welfare effects, this is usually simplistic and often refers to just a consideration of the increase or decrease in the price and volume of goods received. Effects on labour and other distributional factors are often not considered, despite their relevance to a country’s welfare. As noted by Feis,

  • To those directly affected it appears, and with considerable truth, that their wages and conditions are governed not so much by the level of industrial effectiveness within the country as by the fluctuations in the terms of international competition to which they are subject. Their loss - in its direct and indirect effects - seems to, and sometimes may actually, outweigh the consumer gains.
  • So far, the economic case against the social clause has been examined from a conceptual perspective. It is important to add an empirical dimension to this examination by considering the impact of core labour standards on wage levels, competitiveness, investment and economic growth.

    4.1.2 Economic Grounds for Core Labour Standards

    The case for core labour standards and in turn the social clause should not be judged purely on economic grounds. Moral and human right considerations are of greater importance. Nevertheless, it is not possible to ignore the economic dimensions of core labour standards and the social clause. As noted earlier, according to neo-classical economic theory, economic efficiency is best achieved by perfectly competitive markets. But for this to hold, the following conditions must be fulfilled:

    However, it is acknowledged that perfectly competitive markets do not and cannot exist in the real world. Economic actors are often constrained in their decision-making and, in many cases, optimising rational behaviour is simply not possible. There is no such thing as perfect information. In most markets, information is a source of power and is often jealously guarded. In the labour market, inefficiencies increase when workers and some employers are poorly informed about their work environment, particularly in regard to health and safety hazards. Many product markets are characterised by imperfect competition with the existence of monopolies, oligopolies and monopsonies. Similarly, in labour markets there exists a highly unequal power relationship between the employer and the worker.

    In the absence of core labour standards, imbalances of power in the labour market are likely to be aggravated, leading to significant losses in efficiency. For each of the core labour standards, it can be shown that there are strong economic reasons for their enforcement.

    Freedom of association and collective bargaining: The bargaining power of an individual worker is usually very limited compared to that of a powerful employer or group of employers. In many respects, employers may act as monopsonists, in the sense that they are the only buyer of labour, and are thus in a position to force workers to accept whatever terms of employment they wish to impose. In such a situation, workers are also less likely to invest in firm-specific skills. This situation is clearly an infringement of the condition of equal power. Freedom of association and collective bargaining are thus necessary rights to counter-balance the market power of employers and help to enhance labour market competition.

    Collective bargaining, by allowing a more equal sharing of benefits, can help to improve worker-management cooperation and encourage the sharing of information to enhance productivity. Uneven market power becomes an even greater problem for workers belonging to groups that traditionally have had little voice in society - children, women, and ethnic and religious minorities. ×

    Freedom from forced labour: The practice of forced labour clearly contravenes the freedom of choice condition. Forced labourers have little scope for utility maximisation and do not have the freedom to undertake work which matches their skills. The allocation of labour resources is thus distorted, leading to a loss of economic efficiency and output.

    Non-discrimination in employment: Discrimination in the labour market on racial, religious, sexual, political or other grounds is economically inefficient and will also lead to a sub-optimal allocation of labour. Take, for instance, discrimination against women workers in certain occupations. This entails waste as workers will not be employed in the occupations where they will be the most productive. Similarly, when particular categories of the population are not allowed to work, be it for racial, religious or political reasons, the total output will be lower than otherwise. Discrimination leads to market outcomes that are not only inequitable but inefficient.

    Elimination of exploitative forms of child labour: Exploitative forms of child labour, such as bonded labour, are often synonymous with forced labour. Thus child labour constitutes one clear departure from the freedom of choice condition. In addition, children represent an important pool of future human capital. Ill-treatment and exploitation of children rapidly depreciates society’s future potential for development. In addition, the market may fail to provide enough incentives to ensure an adequate level of investment in education and training. This is partly because education and training are typically long-term investments, whereas the planning horizon of individual agents is often a short-term one. Thus while it may be profitable in the short term for unscrupulous employers to exploit children, it is certainly not profitable for society as a whole.

    Governments have an important role to play in ensuring that labour markets function efficiently, as they set the rules for economic transactions. The rules governing formal labour markets should define the rights of workers, unions and employers, the conditions for collective bargaining, and a system for settling disputes. Governments can also intervene directly in the labour market to achieve particular social goals.

    4.2 Some Empirical Findings on Core Labour Standards and the Social Clause

    The debate over the social clause has generated a relatively sizeable body of literature from both opposing and supporting camps. Despite the fact that much of this debate has been on the likely economic effects of core labour standards and in turn the social clause, there have been few empirical studies on this topic. The next section reviews the few studies that have sought to fill this gap and reports on some of the main findings. First, it is important to establish the impact of core labour standards on economic variables such as output, productivity and wages. Second, the question arises of the effect national differences in the observance of core labour standards have on trade performance and FDI. In particular, does non-observance of core labour standards necessarily imply lower wages and thus greater competitiveness? Third, to what extent can the unemployment problems of the developed countries be blamed upon the non-observance of core labour standards in the developing countries?

    4.2.1 Output, Productivity and Wages

    According to a recent OECD study, the relationship between the degree of observance of freedom of association and per capita GDP is weak. However, it is interesting to note that the study found that GDP growth increased significantly in a number of countries (Argentina, Panama, Peru, Philippines, Uruguay and Venezuela) after an improvement in freedom of association. On the other hand, in eight countries (Brazil, Ecuador, Fiji, South Korea, Surinam, "Taiwan, China", Thailand and Turkey), GDP growth fell significantly after the improvement, while in other countries (Dominican Republic, Guatemala and Honduras), no change was recorded.

    On wages and productivity, the study found that there is no empirical evidence that raising core labour standards resulted in higher real wage growth. At the aggregate level, there appears to be no correlation between real wage growth and freedom of association. On the whole, the OECD study suggests that freedom of association (the indicator used for gauging the level of observance of core labour standards) is not correlated with output, wages or productivity. In part this finding can be explained by the fact that in most developing countries there is practically an "unlimited supply of labour". The formation of independent trade unions are thus by themselves unlikely to have a significant effect on wages. The reason is that with an "unlimited supply of labour", the ability of trade unions to counter the market power of employers is likely to be limited.

    It is also important to note the limitations of the OECD study. First, core labour standards are qualitative variables which are by nature difficult to quantify. Second, given the lack of information on the enforcement of core labour standards, the empirical investigation was based only on the right to freedom of association. Third, the investigation was also based on macro-economic indicators, such as GDP, which are only able to capture aggregate effects. Fourth, the study does not distinguish between the effects of other factors such as technological change, factor accumulation, oil price changes, economic policies, etc. Given the general uncertainty over the economic effects of core labour standards, it is difficult to form a conclusive view on the matter. Nevertheless, at present, the suggestion appears to be that greater observance of core labour standards is unlikely to adversely affect economic growth.

    Should this be the case, the fears expressed by certain countries about this supposed effect are misplaced. At the same time, if greater observance of labour standards has a negligible effect on wages, it should be accepted by the developed countries that competition from developing countries is a result of a natural factor endowment. There would thus be no economic grounds for alleging that this is somehow ‘unfair’. Low wages are usually the result of low levels of productivity, which are in turn due to low levels of attainment in important competitive factors such as technology.

    In this respect, research by Golub has shown that differences in productivity are more or less matched by differences in unit labour costs. For instance, in 1990, manufacturing wages in Malaysia were only 15% of those in the US, while Malaysia’s average productivity was also about 15% of the US’s. In other words, unit labour costs were roughly equal in the two countries. In addition, recent studies by the World Bank and earlier work by Amsden and Lim all point to the general conclusion that the positive economic performance of Asian NICs have not been due to either the brutal exploitation of labour or downward real wages. Of greater importance have been the productivity gains from education, infrastructure development and the restructuring of markets from low wage labour intensive to upgraded value-added and skill intensive techniques.

    4.2.2 Effects on Trade

    Does trade between a high-standards country and a low-standards country reduce real income and output in the former? The OECD study found that international differences in freedom of association and collective bargaining appear to have no detectable impact on cross-country trade performance. Moreover, in the countries studied, more liberal labour policies have been accompanied by better productivity in the long run.

    Can the pattern of comparative advantage that is justified on the basis of natural resource endowments and technology be altered by different degrees of enforcement of core labour standards? A review of reports of the ILO Committee on Freedom of Association suggests that in a number of cases, complaints have occurred in sectors where countries have a comparative advantage. However, it is difficult to be certain of the impact of core labour standards on a country’s comparative advantage. Nevertheless, some countries have deliberately attempted to strengthen the comparative advantage of a particular sector through the non-enforcement of core labour standards. Malaysia, for instance, has special legal restrictions on freedom of association in the electronics sector, a sector in which it holds a strong comparative advantage. Whether this works or not, is of course, a different matter.

    Will a country that improves its observance of core labour standards suffer a loss of competitiveness and trade performance? From a study of US import prices of textiles it was found that prices of imports from developing countries tend to be rather uniform, even though the degree of observance of the right to freedom of association varies substantially among these countries. In addition, the use of child labour in the manufacture of hand-made carpets does not seem to have reduced the export prices of these carpets compared with other countries where the practice of child labour is not widespread. A similar estimate has been carried out for the export of carpets manufactured with artificial fibres. For this type of product, Belgium and the Netherlands have lower average export prices than Turkey, where there is some evidence of child labour. Studies by the ILO have also shown that in countries where child labour is used, the labour cost savings realised through the employment of children are small.

    On the whole, the empirical evidence suggests that core labour standards do not play a significant role in shaping trade performance. It is highly unlikely that low-standards countries will enjoy export gains to the detriment of high-standards countries. Moreover, even if there is some gain from repressing core labour standards, these are likely to be short-lived and are likely to be outweighed in the long term by economic costs associated with labour repression. Therefore, the fear that better observance of core labour standards would result in a loss of competitiveness appears to be unfounded.

    4.2.3 Effects on Foreign Direct Investment

    What factors influence the location of FDI? It is generally believed that FDI is attracted by political stability, size of domestic or regional markets, availability of resources, low labour costs, etc. Of these factors, what emphasis is given to core labour standards? Empirically it is difficult to isolate the relevance of core labour standards due to the multitude of factors involved and the lack of relevant data.

    Nevertheless, the main point to note here is that while labour costs may be an important factor to TNCs, this does not mean that countries which enforce core labour standards will necessarily risk negative repercussions on FDI flows. This is because it is not proven that there is a causal link between core labour standards and labour costs. The problem is that many countries have chosen to ignore this and maintain, as if it were an absolute truth, that there is a positive causal relationship between core labour standards and labour costs. In many cases labour repression may actually have more to do with domestic politics than with the need to create or maintain international competitiveness or to attract foreign capital.

    At the same time, TNCs are not innocent bystanders and have in many cases encouraged countries to enforce repressive labour policies (see section 1.2 and 1.3). But is such behaviour economically rational? Savings from the repression of core labour standards are likely to be limited, while dynamic gains arising from respect for core labour standards, in terms of a motivated and productive workforce, are likely to be significant. If this is the case, why is it that certain governments and TNCs appear to be committed to a path of severe labour repression? TNCs are willing to adopt a ‘slash and burn’ attitude, as they can easily move on as soon as all the resources of the country - including human resources - have been exhausted. For them, a short-term strategy makes sense, as they do not have to deal with the consequences.

    Governments on the other hand have to deal with the consequences and should therefore resist such short term strategies. However, unless governments are willing to act together against such strategies, they will always be vulnerable to being played off against one another by increasingly powerful TNCs. The problem is that countries in which labour is severely repressed tend to be run by governments which are un-democratic and have no regard for human rights or the welfare of their citizens. Thus while host countries may be able to enforce core labour standards without risking negative repercussions on FDI flows, they are unlikely to do so without some form of external pressure.

    4.3 Core Labour Standards: Obstacles to or Catalysts of Development?

    Opposition by developing countries to the social clause, on the basis that observance of core labour standards would somehow undermine their development potential, stems from equally weak arguments. These often ignore the fact that core labour standards refer essentially to a set of human rights and not to labour codes on wages, working-time, health and safety, welfare and social security, etc. It is quite clear that respect for the right to freedom of association and collective bargaining, non-discrimination in employment and freedom from forced or child labour does not depend on whether a country is rich or poor. Given the nature of the standards which are being proposed for the social clause, it is difficult to see how they could possibly act as direct obstacles to economic growth.

    As has been shown in previous sections, respect for core labour standards may actually help accelerate development. For instance, it has been highlighted that a highly unequal income distribution may act as an obstacle to economic growth. One way in which it does this is by impeding the emergence of a mass market in durable consumer goods, thus reducing the possibility of developing countries emulating the Fordist growth model experienced by the developed countries. Freedom of association and collective bargaining, by helping to counter-balance the market power of employers, may help to reduce income inequality and thus assist in creating a mass market.

    In addition, core labour standards promote the development of human capital, without which no economic growth is possible. Wages close to or below the minimum subsistence level makes it impossible for workers to invest in their own education or that of their children. Appropriate remuneration, on the other hand, enables workers to maintain and enhance their qualifications and increase their incentive to adopt performance-oriented behaviour. Moreover, as noted by Scherrer, core labour standards are necessary to help make the transition from an extensive to an efficient use of labour.

  • Under the prevailing system of sweatshops, employers have no particular interest in using labour intensively, first because workers are paid on a piece basis and hence no fixed labour costs arise, and secondly because their capital stock is usually small and consists of outdated machinery that cannot be used more efficiently. The resulting low labour productivity in turn precludes raising wages. In such a situation, social standards could increase interest in measures to raise productivity by changing the structure of incentives for firms and workers. For firms, they would make the extensive use of labour less attractive; for workers, they would make it more rewarding to strive for the success of the firm....... to succeed, certain preconditions must be met which ensure that workers can earn better wages, show themselves to be co-operative and acquire professional qualifications. Social standards could help create those preconditions.
  • Despite the contribution that core labour standards can make to economic development, the non-observance of standards is used in many countries as a tool to increase exports and FDI, especially in EPZs (see section 1.2). Trade unions are prohibited in the EPZs of Bangladesh and Pakistan, while they are discouraged in Guatemala and Panama. These cases cannot be explained in terms of socio-cultural differences in the approach to human rights or in terms of low levels of economic development (see section 3.3).

    Indeed, such practices appear to be part of a deliberate strategy to improve export performance. Whether they achieve this objective or not is however not clear. It has been observed in a number of countries, that the development and expansion of the export sector does not necessarily lead to an improvement in living and working conditions. Developments in the Caribbean over the last decade or more show that despite a massive increase in export production and employment, working conditions did not improve . Even in the countries that experienced the economic miracles of the 1980s, such as South Korea, a visible improvement in living standards was achieved only as a result of very hard trade union bargaining.

    4.4 The Role of Trade Unions

    Trade unions therefore play an extremely important role in helping to solve problems of unequal market power, discrimination and insufficient information. Unions provide their members with important services, negotiating on their behalf for better working conditions, protecting them from unfair treatment, and dividing the cost of obtaining information among a large number of workers. Recent analysis of the relationship between trade unions and productivity in Malaysia showed that unions can enhance productivity and efficiency. Unionised Malaysian firms tended to train their workers more and to use job rotation to enhance flexibility and efficiency. They were also more likely to adopt productivity-raising innovations relating to technological change, changing product mix and the reorganization of work.

    A study in the Republic of Korea in 1988-90 found that unions there placed great value on wage equalisation and that the degree of wage dispersion in the unionised sector was 5.2% lower than in the non-unionised sector. In Mexico, union action appears to have helped reduce discrimination. A study using wage data for 1989 concluded that in the non-union sector, men enjoyed a 17.5% wage advantage over women with identical skills and experience, but the study found no significant wage differential between men and women performing equivalent duties in the unionised sector. Similarly, the study found a significant wage disadvantage for indigenous peoples in the non-unionised sector but no discrimination in firms covered by trade unions.

    Trade unions are often criticised for improving wages and working conditions for their members at the expense of capital holders, consumers and non-union labour. However, existence of a union wage premium is not always proof of negative distributional effects. For example the actions of trade unions in South Africa helped increase the wages of unionised black workers and bring them closer to those of white workers; thus union action helped to improve the distribution of income. Moreover, there are situations where the union wage premium is very small or non-existent. A 1991 study in South Korea estimated that the wages of unionised production workers were only 2 to 4 % higher than those of non-union workers.

    4.5 From Human Rights and Political Economy Perspectives

    Some analysts speculate that the issue of core labour standards may follow the path that has already been taken by environmental issues and intellectual property rights. However, there is a major difference that would make agreement on the social clause much more difficult to achieve. The problem is that denial of the right to freedom of association and collective bargaining is in some countries part of the political system. Trade unions can be an important source of opposition to authoritarian regimes. Governments that presently do not allow these rights are not just fearful of the possible wage and cost implications arising from collective bargaining, but are also fearful that the right to freedom of association might undermine their authority and may directly threaten their very existence.

    It is for these reasons that trade unionists have historically been targeted as enemies by repressive regimes. While some countries may eventually be convinced of the merits of a multilateral social clause, it is hard to see any compromise from the more repressive governments without a wider political change in their countries. For this reason, discussions about the social clause should not be viewed as simply a labour issue. Ultimately, the social clause is about human rights. Repression, discrimination and acts of violence against workers often go beyond the workplace and generally reflect the state of human rights suffered in the country at large.

    Countries which are persistent violators of core labour standards tend to argue that this is due to a "harsh economic reality". Is this really credible? Is a country’s Gross Domestic Product the factor which determines whether women, men and children are arrested, beaten, tortured, kidnapped and murdered, just because they belong to a trade union, or wish to do so, or because they are struggling for humane conditions? Economics can only explain so much, it does not explain everything. Similarly, it is a ‘red-herring’ to continue discussing the economic growth vs. human rights dilemma. There is in fact no dilemma. There is no proof that better observance of human rights adversely affects growth and even if there is some tenuous relationship, it has already been universally accepted by countries which are signatories to the UN Universal Declaration of Human Rights (1948) that economic growth cannot be at the expense of human rights.

    International consensus has been achieved over the decades since the Second World War that human rights protection in any State can no longer be exclusively regarded as its internal affair, but that the manner in which governments treat persons on their territory is the "legitimate concern" of the international community. . The problem is that, all too often, industrialised countries have failed to apply pressure consistently on countries where major human rights violations have taken place; in fact human rights issues have at times been cynically ignored when it was perceived to be in the economic interest of industrialised countries. .

    It is thus not surprising that some developing countries are suspicious of the real intent of linking core labour standards with trade. However, such cynical behaviour does not lessen the need for international involvement in the protection of human rights. In terms of core labour standards, it highlights the need for a transparent multilateral mechanism, such as the social clause, to enforce the human rights embodied in these standards. Without the social clause, it is possible that the larger countries may make greater use of unilateral measures for protectionist purposes, under the guise of concern for human rights.

    This is a very real danger, given the willingness and ability of countries such as the US to seek remedies via the extra-territorial application of its laws. Indeed, US national legislation already provides the means for taking a stance of aggressive unilateralism. This is combined with a determination to identify which are ‘outlaw’ nations and what are ‘outlaw’ values. Despite the nearly unanimous view of international jurists that such legislation is incompatible with international law, there is no guarantee that under pressure from the groups concerned it will not be used to gain supposedly strategic ground. From a systemic perspective, this could lead to an eventual erosion of the rule of law in international trade and in turn threaten the multilateral trading system.

    In addition to unilateralism, there is also a danger that intransigence on the social clause could end up pushing regional arrangements away from their present open nature to a closed one. It is not too difficult to envisage the setting up of homogenous economic blocs as regards social protection, sheltered from what is perceived to be unfair competition. As it is, social policy - including core labour standards - is an issue that is already at the top of the agenda of the most important regional groupings in the world. As noted in an earlier section, as jobs continue to be lost in the developed countries, greater support will be given to the notion that trade with the developing countries is to blame.

    Governments will have to be seen to be doing something, if they are to maintain popular support. Allegations of ‘unfair competition’ are particularly useful, as they help to deflect attention away from the failure of domestic policy and effectively externalise the problem. Given this scenario, it is not too hard to see the attraction of a political platform of virulent economic nationalism that calls for more trade barriers. To resist the social clause is to play into the hands of those with a protectionist agenda. Thus, if only for these reasons, it is in the interest of most countries to negotiate the establishment of a multilateral social clause.

    While this section has shown that there are very strong economic grounds for the social clause, it is important not to lose sight of the fact that the fundamental reason for strengthening core labour standards is that it is the right thing to do. The arbitrary arrest, torture and murder of trade union organisers, the enslavement of children, the locking of doors in garment factories resulting in hundreds of workers burning to death, and many other human rights violations are not matters that should be thought about purely in economic terms.

    Continue to chapter 5

    Globalization and Workers' Rights


    International Labour Office
    Bureau for Workers' Activities
    CH-1211 Geneva 22
    Fax: +41 22 799 6570
    ACTRAV Homepage: