Skirting the Issue: The impact of changes to AGOA on women workers in the garment sector

We went to see the King on the AGOA issue. We got down on our knees and begged him to save our jobs. We went to see him on behalf of the women in the textile sector, unionised and un-unionised, to make him aware of the problems we are facing with AGOA leaving the country - (ILO, 2014).


Aisha Bahadur is a partner at Civil Society Research and Support Collective (CSRSC ) based in Durban, South Africa. She provides strategic support to civil society organisations, and predominantly organised labour, in the areas of research, education and

April 21st, 2015

We went to see the King on the AGOA issue. We got down on our knees and begged him to save our jobs. We went to see him on behalf of the women in the textile sector, unionised and un-unionised, to make him aware of the problems we are facing with AGOA leaving the country - (ILO, 2014).

So begins the story of how three women shop stewards put their pride in their pocket and prostrated themselves to Africa’s last absolute monarch, Swaziland’s King Mswati III, in the hope that he would institute reforms that would prevent divestment in the countries garment sector. 

The country also has one of the world’s longest running states of emergency, currently in its forty-second year. As Swaziland has failed to make progress on reforms to uphold human and worker rights and show respect for the rule of law, it became inevitable that the US would withdraw Swaziland’s eligibility to preferential trade access under the US African Growth and Opportunity Act (AGOA). This was announced by US President Barak Obama in June 2014 and will come into effect from 1 January 2015 (US Office of the Press Secretary, 2014).

AGOA is a US Act passed in early 2000 that governs the preferential trade access offered by the US to the sub-Saharan countries it deems eligible. Exports from the region to the US rose from US$16.5 billion in 2000 to US$38.2 billion in 2013 under AGOA, reaching a record of close to US$80 billion in 2008 (USITC, 2014). However, more than 90 percent of these exports are petroleum crude oil and petroleum products. 

Exports of products other than crude petroleum from the region grew from US$1 billion in 2001 to a record US$5.1 billion in 2008 before falling drastically as a result of the economic recession in 2009 (USITC, 2014). These exports recovered and reached US$4.8 billion in 2013 (ibid). While initially apparel made up most of the non-petroleum exports to the US under AGOA, apparel made up only 20 percent of this in 2013 (US$907 million) (ibid). Despite these changes, jobs created in the textile and garment sector are still touted as a major success story of AGOA.

Apparel exports under AGOA from Swaziland were valued at US50 million in 2013 and the sector is the second largest foreign exchange earner after sugar and sugar-based concentrates (Woolfrey, 2014). The Swazi garment sector has suffered with exports under AGOA much lower than a decade earlier when foreign investors took advantage of access to the US market while the global quota regime for textile and garments, the Multi-Fibre Agreement (MFA), was still in force (Jones & Williams, 2012). In particular, those from Taiwan invested in Swaziland, as Taiwan has formal recognition from Swaziland in its controversy with China regarding its political status. Swaziland is one of the few remaining countries to formally recognise Taiwan and favourable bilateral relations exist between the two countries (Molapo, 2014).

When the MFA expired at the end of 2004, thousands of jobs were lost in Swaziland as well as in other African countries that had attracted investment in the garment sector under AGOA. Employment in the textile and garment sector shrunk from 30,000 jobs in 2004 to less than 15,000 in 2005 (Madonsela, 2006). Currently, employment in the sector is around 16,000 (ATUSWA, 2014). About half of these jobs rely on the US market under AGOA while the other half are employed in production for South Africa (ATUSWA, 2014).

In a small country where unemployment is estimated to be as high as 40 percent (Woolfrey, 2014), the loss of 8,000 jobs is substantial. These jobs are also important because there are fewer formal employment opportunities for women (Brixiová, Fakudze & Kangoye, 2012) and over 90 percent of those employed in the garment sector are women. A Swazi shop steward explains: 

We are the breadwinners of our families. We provide for the needs of the family including food clothing and education and school fees are very high up to R10,000 a year. About five percent are employed women that share the responsibility of providing for the family with a husband. The rest of us are women that are widows, single parents or that have unemployed husbands and so we are relied on solely to provide for the family (ILO, 2014).

It is predicted that the garment industry in Swaziland will suffer a dramatic decline in the absence of AGOA, which will result in further job losses. The shop stewards detailed their vulnerability when they spoke to the King:

We told him that the Taiwanese investors are selling their sewing machines and furniture and are preparing to leave us without jobs because of the withdrawal of AGOA from Swaziland. We told him that we are afraid that we will die of hunger, that workers on ARV and TB treatment will die when they cannot take care of themselves anymore, that our children will have to drop out of school because we can’t pay fees, that we will have to resort to prostitution to survive, increasing risk of contracting HIV. We told him that textile workers love him and he is our only hope (ILO, 2014).

It is highly probable that the exclusion of Swaziland from AGOA will leave women more vulnerable but is this a failing of the Act itself and, if so, to what extent? Section 104 of AGOA details the eligibility requirements and, among other things, requires that a country to have established or is making progress towards ‘protection of internationally recognized worker rights, including the right of association, the right to organise and bargain collectively’ (AGOA Online, 2014). It also requires a country not to ‘engage in gross violations of human rights’ (ibid). 

The US required Swaziland to address five problems by mid-2014 (Khumalo, 2014). Most of these relate to trade unions and human rights including amendments to the Industrial Relations Act which would allow for the registration of trade unions and employer federations (ibid). It also requires the dissemination and implementation of the Code of Good Practice on Protest and Industrial Action and recognition of the freedom of assembly, speech and organisation (ibid). At the centre of the withdrawal of eligibility is the Swazi government’s deregistration of the Trade Union Congress of Swaziland (TUCOSWA) formed after a merger between two labour federations in 2012 (James, 2014). The government claimed that they did not have scope in the law to register a federation, despite both merging federations having been registered in the past (Lee, 2012).  The government then failed to institute reforms that would allow a federation to register (James, 2014). 

The fact that workers are forced to bear the consequences of eligibility withdrawal brings into question the eligibility requirements and whose interest they serve. Does this further the interests of vulnerable groups by achieving improvements in environments where rights abuses occur? Or does the withdrawal serve the interests of the US which include ensuring AGOA is not associated with human rights abuses? 

Outcomes point to the latter. After the withdrawal of AGOA eligibility on 8 October 2014, the Swaziland Minister of Labour and Social Security suspended all federations with immediate effect, pending legal reforms (IndustriALL Global Union, 2014a). This affected employers’ associations as well as TUCOSWA. The Minister specifically mentioned the suspension of the Amalgamated Trade Union of Swaziland (ATUSWA) (ibid), which was formed in 2013 after a merger of nine unions, predominantly from the manufacturing sector. This effectively constitutes a ban on organised labour and is a gross violation of the International Labour Organisation’s (ILO) Convention No. 87 on the freedom of association. This is despite continued efforts by a number of interest groups to promote reform, including an ILO fact-finding mission in January that recommended the registration of worker and employer federations (IndustriALL Global Union, 2014a). 

It appears that the AGOA eligibility withdrawal has not provided any impetus for reform in Swaziland and this has also been the case in Madagascar. When Madagascar lost its AGOA eligibility on 31 December 2009, it was already experiencing a downward trend in US exports from its peak in 2004 (US$323 million) but it was still the second-highest exporter of apparel under AGOA (US$211 million) (Fukunishi, 2013). Apparel exports from Madagascar in 2010 thus dropped by 75 percent with employment dropping drastically from 57,250 in 2008 to 26,600 in 2010 (ibid). While the country did take a knock on foreign exchange earnings through AGOA exports, it was poor women garment workers who faced the immediate and devastating consequences through the loss of their livelihoods. Studies on the effect of the withdrawal of AGOA eligibility in Madagascar suggest that the loss of this status has had more of a negative effect on the economy and livelihoods than on political instability (Fukunishi, 2013). Of the 19 African companies producing exclusively for the US market in 2009, only six survived until 2011 (ibid). However, Madagascar still has preferential access to the EU and increased exports to South Africa, which has helped sustain the garment industry. 

Any changes to trade policy in the garment sector are likely to have an impact on women workers. Therefore these policies should not be gender blind and should rather advance gender equality and the empowerment of women, particularly the economic empowerment of women. Garment industry jobs are often done by women and are particularly important for gender equity, offering the opportunity to gain economic status (UNIDO, 2012). However, wages in the garment sector can be very low, so these are usually vulnerable workers who are more than likely very poor with very few other employment opportunities available to them due to high unemployment rates (Bahadur, 2004). Given the general pattern of comparative advantage in low-income countries, unplanned cancellation of duty-free access negatively affects labour-intensive industries and low-skilled workers. 

In addition to good governance requirements of countries for eligibility, there should also be requirements for companies exporting under AGOA to uphold human and worker rights. Poor working conditions have been well documented by trade unions and NGOs in some companies exporting under AGOA, particularly in the garment sector. These include poor health and safety, exploitative wages, no overtime pay, physical punishment, verbal threats, extremely strict leave policies and no maternity rights. AGOA should recognise the responsibility of companies producing under AGOA to uphold worker and human rights or face exclusion (Bahadur, 2004; Solidarity Centre, 2014). 

Since AGOA’s inception, workers’ interests and human rights have progressed in some ways. The Ruggie Framework, adopted by the United Nations Human Rights Commission (UNHRC) in 2008, is the basis of the Guiding Principles for Business and Human Rights (Business & Human Rights Resource Centre, n.d.). The framework has a three-pillar policy for dealing with the problem of corporate human rights abuses: (a) the state has a duty to protect human rights, (b) corporate bodies have a duty to respect them and (c) effective remedies must be available at public and private levels to enforce them (ibid).

Under AGOA, only the first pillar is recognised, making AGOA a very blunt instrument in terms of worker and human rights standards. The danger of developing a domestic set of demands around rights is that human rights become a tradable commodity in the race to attract investment. International norms are needed as this is the only way to remove competition through externalised costs, be they human or environmental. 

For grievance mechanisms to be effective, binding international standards should be established and judicial mechanisms to enforce these must be functional and respected. The tighter the regulatory nature and stronger the recourse to judicial justice, the stronger the nature of company level and multi-stakeholder grievance mechanisms will be. Thus, current efforts of the UNHRC to establish binding standards on human rights for multinational corporations should be championed, especially in the US where the government is opposed to such standards being developed.

Review mechanisms on AGOA eligibility need to be refined to include social dialogue on recommendations made to countries. This dialogue should be multilateral between the relevant government ministries, export producers and labour. Other stakeholders, such as the ILO, should also be consulted where appropriate. The purpose of this dialogue should be to put in place a plan within a stipulated timeframe for reform with progress indicators that must be achieved. If governments fail to make sufficient progress towards these indicators, only then should eligibility be revoked as a final resort when all other efforts have been exhausted. 

Had there been social dialogue in Madagascar, thousands of jobs may have been saved. No lessons were learnt from Madagascar on the impact of and changes needed to AGOA eligibility standards for workers in Swaziland. This suggests that the eligibility requirements are not aimed at furthering the interests of vulnerable groups but to ensure that US trade policy is not directly associated with human rights abuses.  

Recovery once eligibility is reinstated may not be easy to achieve if investment in the country has already declined as a result. President Obama reinstated Madagascar’s AGOA eligibility in the same announcement in which he withdrew Swaziland’s eligibility (US Office of the Press Secretary, 2014). It remains to be seen whether exports and employment will recover in Madagascar, given the closure of numerous companies a few years ago. 

Investment in the garment sector favours weak labour standards enforcement, low wages and poor trade union organisation – conditions that exist in Madagascar and so there is a good chance that the sector will recover. If improvements for workers are going to be achieved, however, it is important to have improved worker’s rights provisions in AGOA or any trade agreement related to countries which tolerate poor conditions for their workers. Without these, trade agreements will facilitate a rapid decrease in wages and conditions. This would be counterproductive in terms of the social benefit goal that seeks to increase trade. At the very least, core labour standards including freedom of association and the right to collective bargaining need to be ensured. 

While separate factors, trade is linked to investment. Oftentimes investment takes place to exploit market access resulting from preferential trade agreements. This means that there is often more capital outflow than inflow from developing countries as a result of such agreements. The only benefit to the developing country from export-orientated production is the wage component and its potential stimulus for the local economy. Thus, initiatives like AGOA, which are designed to promote exports, should have strong worker provisions to ensure that workers are able to organise themselves and place pressure on employers on the issue of wages. 

Failing to ensure adequate wages for workers in export sectors has been a failing of AGOA. In Lesotho, the top exporter of apparel under AGOA, good social dialogue exists, harmonious industrial relations and there is compliance with international labour standards. This is in part due to the Better Work Programme in Lesotho, a joint ILO and International Finance Corporation programme that aims to improve compliance with labour standards and competitiveness in global supply chains in the garment sector. However, wages in the garment industry in Lesotho remain low and are the same as in Swaziland, at US$120 a month (ATUSWA, 2014). 

There are five unions which organise 40,000 workers in Lesotho and, even though workers are twice as many as those in Swaziland, the lack of unity undermines bargaining strength (IndustriALL Global Union, 2014b). Thus there is a situation where good practices on labour rights exist despite weak unions and low wages (Eppel, 2014). In fact, a study conducted by the ILO in Lesotho found that workers had suffered from a real wage decrease as increases in the garment sector had not kept up with inflation over a period of years (Koen, 2012).

To ensure social benefit is attained, labour standards should be linked to a decent wage, particularly in labour intensive sectors like the garment sector. A distinction needs to be made between trade agreements and preferential trade agreements, like AGOA. Preferential trade agreements are about preferential market access for the purposes of stimulating development. This is key to the ‘trade not aid’ agenda. For this model to achieve its developmental objectives, ensuring wage benefits to the developing country with preferential market access is vital.

Growth, in particular export-led growth, does not necessarily lead to development, especially when this growth is based on production in low value-added sectors and low wage regime. AGOA has increased Africa’s trade with the US but has not resulted in substantial development. For example, despite economic growth of 6.5 percent in 2012 and 3.4 percent in 2013 in Lesotho, mostly due to recovery in the apparel sector, the proportion of households living below the poverty line has increased from 56.6 percent in 2003 to 57.1 percent in 2013 (AFDB, 2014). With Lesotho’s Gini coefficient at 0.51, a poverty level of 57 percent and a 25 percent unemployment rate, policies need to make growth more inclusive (ibid). 

AGOA should require governments to set a minimum wage classified according to industry sectors and review these wage standards on an annual basis to ensure that wages are at the very least keeping up with inflation. Without such measures, real wages are eroded over time. Governments should also be required to uphold the right for trade unions to bargain at the industry level with employers and for this to be facilitated by a bargaining council and factory-level bargaining. 

In fact, African governments should be championing wage-led growth, which not only provides workers with a living wage but builds internal markets. It is also important to unlink trade and development. The economist, Ha-Joon Chang (2003), argues that ‘the potential value of a policy or an institution to a country should be determined more by what it will do to promote internal development than by what the international investors will think about it.’ This requires the states to have both the political will and the policy space to drive a developmental agenda forward. Trade and export-led growth has some role in this but does not lead to development. An example of development that benefited from export-led growth in China is provided by economist Joseph Stiglitz. China ‘took measures to ensure that significant portions of that increased output went to the poor, some went to provide for public education, and much was reinvested in the economy, to provide more jobs’ (Stiglitz, 2012). 

Such an approach in Swaziland seems unlikely as it is a strongly patriarchal country. While gender equality is enshrined in its constitution, Swaziland has made little progress to revise its discriminatory civil laws and traditional laws that continue to restrict women to inferior roles. Swazi women contribute substantially to the economic wellbeing of their families and communities but their contributions have been routinely undervalued. So it is not surprising the desperate pleas to save 8,000 jobs of women garment workers made by the women trade unionists that knelt at the King’s feet, despite pleasing him with declarations of gratitude and love and bearing an expensive gift, a Brahman bull they can ill afford, go unheeded. 

Aisha Bahadur s a partner at Civil Society Research and Support Collective (CSRSC ) based in Durban, South Africa. She provides strategic support to civil society organisations, and predominantly organised labour, in the areas of research, education and advocacy.


AFDB (2014) Lesotho Economic Outlook. African Development Bank. From: . Accessed: 17 November 2014.

AGOA Online (2014) AGOA Country Eligibility. From: . Accessed: 17 November 2014.

ATUSWA (2014) Report by the Amalgamated Trade Union of Swaziland at the ILO Actrav Decent Work in the Textile and Garment Sector workshop. 14 to 16 November, Durban, South Africa.

Bahadur A (2004) ‘Taking the Devils Rope’ – AGOA and Sweatshops in the Apparel Sector. OSISA. From: . Accessed: 17 November 2014.

Brixiová Z, Fakudze R & Kangoye T (2012) Labor Markets in Swaziland: The Challenge of Youth Employment. Youth Policy, UNDP. From: . Accessed: 17 November 2014.

Business & Human Rights Resource Centre (n.d.) UN ‘Protect, Respect and Remedy’ Framework and Guiding Principles. From: . Accessed: 17 November 2014.

Chang H (2003) Kicking Away the Ladder: Development Strategy in Historical Perspective. Anthem Press: London, UK. 

Eppel S (2014) Prospects for Survival and Growth of the African Garment and Textile Sector. SACTWU presentation at the ILO Actrav Decent Work in the Textile and Garment Sector workshop. 14 to 16 November, Durban, South Africa.

Fukunishi T (2013) Political crisis and suspension of duty free access in Madagascar: Assessment of Impacts on the Garment Industry. Institute of Developing Economies, Discussion Paper No. 422. From: . Accessed: 17 November 2014.

ILO (2014) Interview conducted with three Swazi shop stewards. ILO Actrav Decent Work in the Textile and Garment Sector workshop. 14 to 16 November, Durban, South Africa.

IndustriALL Global Union (2014a) Swaziland Bans Trade Unions. From: . Accessed: 17 November 2014.

IndustriALL Global Union (2014b) Report by IndustriALL Global Union at the ILO Actrav Decent Work in the Textile and Garment Sector workshop. 14 to 16 November, Durban, South Africa.

James C (2014) The Need for a Balanced Debate on Swaziland and AGOA. From: . Accessed: 17 November 2014.

Jones VC & Williams BR (2012) US Trade and Investment Relations with sub-Saharan Africa and the African Growth and Opportunity Act. From: . Accessed: 17 November 2014.

Khumalo S (2014) Swaziland Attending to AGOA Recommendations – Prime Minister. Times of Swaziland, 2 June. From: . Accessed: 17 November 2014.

Koen M (2012) Towards a Living Wage: Assessment of Lesotho Manufacturing, Workers Household Needs using a Subsistence and Basic Needs Baskets Approach with Associated Price Data and Wage Implication Calculations. ILO Report.

Lee R (2012) Swazi Authorities Panic as Protests Loom. OSISA. From:  . Accessed: 17 November 2014.

Madonsela WS (2006) The Textile and Clothing Industry of Swaziland. In Jauch H & Traub-Merz R (eds) The Future of the Textile and Clothing Industry in Sub-Saharan Africa. Friedrich-Ebert-Stiftung: Bonn, Germany. 

Molapo T (2014) A comparative analysis of the protection of women’s labour rights in the apparel industries of the Southern African Customs Union member states of South Africa and Lesotho, under the African Growth and Opportunity Act. UCT, Cape Town, South Africa. From: . Accessed: 17 November 2014.

Solidarity Centre (2014) Survey Reveals Abuse of Textile Workers in Swaziland. From: . Accessed: 17 November 2014. 

Stiglitz J (2012) The Price of Inequality. Penguin: London, UK. 

UNIDO (2012) Gender inequality and its effects on industrial development. UN Industrial Development Organization. From: . Accessed: 17 November 2014.

US Office of the Press Secretary (2014) Presidential Proclamation – AGOA. The White House. From: . Accessed: 17 November 2014. 

USITC (2014) AGOA: Trade and Investment Performance Overview. United States International Trade Commission. From: . Accessed: 17 November 2014.

Woolfrey S (2014) Drawing lessons from Swaziland’s AGOA exclusion. Trade Law Centre for Southern Africa. From: . Accessed: 17 November 2014. 


  • 1 Hood Avenue/148 Jan Smuts; Rosebank, GP 2196; South Africa
  • T. +27 (0)11 587 5000
  • F. +27 (0)11 587 5099