May 2021

Spinning around workers’ rights

International companies linked to forced labour in Tamil Nadu spinning mills

Pauline Overeem Martje Theuws SPINNING AROUND WORKERS’ RIGHTS | 1 Diewertje Heyl

Colophon

AUTHORS Pauline Overeem (SOMO), Martje Theuws (SOMO), Diewertje Heyl (Arisa). TEXT CORRECTIONS Leigh Chambers LAYOUT Karen Paalman. www.getlos.nl Contributions by Meike Skolnik (SOMO) and Hans Maas (Arisa).

Mondiaal FNV has actively supported and contributed to the research.

Thanks to Martha Kapazoglou and Eva Loeve, who supported work on this report during their internships at SOMO in 2020 and 2021. Also, many thanks to the Dutch anthropologist1 who joined the research team in a volunteer capacity, from mid-October 2019 to mid-January 2020.

Arisa and SOMO are part of the Dutch coalition of the Clean Clothes Campaign (CCC).

This publication has been made possible with financial assistance from Mondiaal FNV and the Dutch Ministry of Foreign Affairs (Strategic Partnership - FGG).

Disclaimer: The content of this publication is the sole responsibility of Arisa and SOMO and does not necessarily reflect the views of the Ministry of Foreign Affairs and Mondiaal FNV.

May 2021.

SOMO SOMO investigates multinationals. We conduct action-oriented research to expose the impact and unprecedented power of multinationals. Cooperating with hundreds of organisations around the world, we ensure that our information arrives where it has the most impact: from communities and courtrooms to civil society organisations, media and politicians. www.somo.nl

Arisa Arisa is an independent human rights organisation aiming to support and strengthen human rights in South Asia, together with local NGOs and trade unions. The focus of our work is on the rights of vulnerable groups in global supply chains. www.arisa.nl

Indian partners The Indian counterparts are well-respected and well-established labour rights and human rights organisations and labour researchers based in Tamil Nadu who SOMO and Arisa have been working with for years. We cannot reveal their identities in this public report, for fear of retaliation.

SPINNING AROUND WORKERS’ RIGHTS | 2

Preface

It would be easy to assume, as a consumer in a globalised economy, that there are no secrets about how prod- ucts are made. We are bombarded daily with images of new clothes, worn by models on billboards, or endorsed by influencers on social media around the world. And with brands increasingly committed to safeguarding both environmental standards and human rights, and governments slowly imposing laws to prevent violations of these rights throughout supply chains, then surely a fair apparel production process seems within reach? Unfortunately, as this report shows, such a situation is still a long way away.

Behind the scenes - and away from the scrutiny of buyers, brands, and inspectors - millions of people are working in conditions you do not wish upon anyone, to create the yarn and fabric required for clothing.

The apparel supply chain is highly fragmented, both vertically and horizontally. Yarn and fabric can be sold from one entrepreneur to another before finally ending up in factories producing branded goods. To the brands, it may seem like their t-shirts, shirts and trousers are created under fair circumstances because the working conditions and labour standards at the factory are regularly inspected, and the brands receive positive assessments of what is happening there. The same applies for factories where household textiles are produced.

Yarn and fabric suppliers further up the supply chain receive, however, much less attention. This report by SOMO and Arisa, supported by Mondiaal FNV, includes a detailed description and analysis of the human rights situation in the textile mills of Tamil Nadu where workers’ rights are being violated, workers are too afraid to object to sub- standard working conditions and excessive overtime for fear of losing their jobs, women are suffering harassment on the factory floor and in the hostels where they are obliged to live, and workers are not being paid a proper wage. This report makes painfully clear that there is a dire need for all corporate actors along the apparel supply chain to commit to heightened Human Rights Due Diligence in all phases of product development, also in the upstream supply chain. It includes clear recommendations to brands on what action they should take, and where their focus should lie. One such recommendation is engaging in rigorous supply chain mapping, along with, and this recom- mendation is close to my heart, actively committing to furthering freedom of association and collective bargaining. Facilitating the establishment and functioning of democratically elected, independent, factory-level trade unions that give workers a voice in improving their working conditions is essential to finally creating supply chains.

Karen Brouwer Managing Director Mondiaal FNV

SPINNING AROUND WORKERS’ RIGHTS | 3

Table of contents

Executive summary 6

1. Introduction 12 1.1. Why this research? 12 1.2. Reading guide 12 2. Methodology 13 2.1. Selection of mills 13 2.2. Field research 14 2.3. Supply chain research 15 2.4. Company review 15 2.5. Methodological challenges 15 3. Context 17 3.1. India’s cotton, textile, and garment industry 17 3.2. The Tamil Nadu textile and garment industry 18 3.3. Trade unions 20 3.4. Social exclusion based on caste and tribal background 20 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu 21 4. Normative framework - fundamental labour standards 22 4.1. International guidelines for responsible business conduct 22 4.2. Fundamental labour standards 22 4.3. Forced labour indicators 23 4.4. Legislative developments and existing supply chain legislation 27 4.5. Labour laws in India and Tamil Nadu 27 4.6. Caste-based discrimination legislation in India 30 5. Evidence of existence and risk of forced labour 32 5.1. Characteristics of the 29 investigated mills 32 5.2. Respondents’ profiles 33 5.3. Evidence of forced labour – using the 11 ILO indicators 37 6. Impact of COVID-19 68 6.1 Loss of income 68 6.2 Emergency relief 68 6.3 After lockdown 69 6.4 Safety measures inadequate 70 7. Responses to the research 71 7.1 Supply chain links 71 7.2 Links established based on corporate disclosure 73 7.3 Links with vertically integrated manufacturers 75 7.4 Company responses with regard to the draft research findings 77 8. Conclusions: Beyond 29 mills – a sector risk analysis 79 8.1 Incidence and risks of forced labour 79 8.2 Lack of publicly available data and supply chain transparency 82 8.3 Flawed business model 82 8.4 Restricted interpretation of supply chain responsibility 83 8.5 Reliance on social auditing 83 8.6 Frontrunners bear the brunt 83 8.7 Leverage 84 9. Recommendations 85 9.1 Corporate actors 85 9.2 Government actors 89 9.3 Need for further independent research 91

SPINNING AROUND WORKERS’ RIGHTS | 4

Abbreviations

AGT: Dutch Agreement on Sustainable Garments and Textile AITUC: All India Trade Union Congress CITU: Centre of Indian Trade Unions BC: Backward Caste BCI: Better Cotton Initiative BHRE: Business, Human Rights and the Environment Research Group CMT: Cut Make Trim CoC: Chain of Custody ESI: Employee’s State Insurance EPF: Employees’ Provident Fund ETI: Ethical Trading Initiative FC: Forward Caste FGD: Focus Group Discussion FLA: Fair Labor Association FNV: Federatie Nederlandse Vakbeweging FWF: Fair Wear Foundation GOTS: Global Organic Textile Standard HRDD: Human Rights Due Diligence ICC: Internal Complaint Committees ICN: India Committee of the Netherlands (now Arisa) IDSN: International Dalit Solidarity Network ILO: International Labour Organization ITUC: International Trade Union Confederation MBC: Most Backward Caste mHRDD: Mandatory Human Rights Due Diligence MSI: Multi-Stakeholder Initiative NGO: Non-Governmental Organisation OBC: Other Backward Class OT: Overtime PST: German Partnership for Sustainable Textiles RBC: Responsible Business Conduct RBI: Responsible Business Initiative SAI: Social Accountability International SC: Scheduled Caste SIMA: Southern India Mills’ Association ST: Scheduled Tribe TASMA: Tamilnadu Spinning Mills Association TEA: Tirupur Exporters’ Association UNGP: United Nations Guiding Principles on Business and Human Rights WC: Works Committee

SPINNING AROUND WORKERS’ RIGHTS | 5

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide Executive summary 2. Methodology 2.1. Selection of mills 2.2. Field research 2.3. Supply chain research The state of Tamil Nadu in the southern part of India, is a major hub in the global cotton-based textile and garment 2.4. Company review industry. Countless foreign factories, brands and retailers are sourcing a variety of products made in Tamil Nadu, 2.5. Methodological challenges from yarns and fabrics to apparel items and household textiles. Hundreds of spinning mills operate around the 3. Context clock to churn out huge quantities of yarns and fabrics. Over sixty per cent of all Indian spinning mills are located 3.1. India’s cotton, textile, and garment industry in this state and employ over 280.000 workers. Tamil Nadu is the second largest contributor to the national Indian 3.2. The Tamil Nadu textile and textiles industry, accounting for 19 per cent of the country’s overall textile output. garment industry 3.3. Trade unions But the Tamil Nadu textile industry is also the scene of persistent labour rights violations. In the past ten years, 3.4. Social exclusion based on 2 3 caste and tribal background SOMO and Arisa , along with other civil society actors have reported on a range of labour rights violations, includ- 3.5. Caste discrimination in garment ing child labour, discrimination on the basis of caste and tribal background, and forms of forced labour in garment factories and spinning mills in and textile factories. Tamil Nadu 4. Normative framework - fun- damental labour standards ‘We are working non-stop, without knowing whether it is day or night.’ 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards About the research 4.3. Forced labour indicators For this research, Arisa and SOMO4 again looked at labour conditions in the Tamil Nadu textile industry, this time 4.4. Legislative developments and existing supply chain legislation focusing on the risk and existence of forced labour, using the 11 indicators of forced labour as defined by the In- 4.5. Labour laws in India and ternational Labour Organization (ILO) as a guide? Tamil Nadu 4.6. Caste-based discrimination These 11 indicators are: legislation in India 5. Evidence of existence and risk of forced labour 1. Abuse of vulnerability; 5.1. Characteristics of the 29 2. Deception; investigated mills 3. Restriction of movement; 5.2. Respondents’ profiles 4. Isolation; 5.3. Risks and evidence of forced labour – using the 11 ILO 5. Physical and sexual violence; indicators 6. Intimidation and threats; 6. Impact of COVID-19 7. Retention of identity documents; 6.1 Loss of income 8. Withholding of wages; 6.2 Emergency relief 6.3 After lockdown 9. Debt bondage; 6.4 Safety measures inadequate 10. Abusive working and living conditions; 7. Responses to the research 11. Excessive overtime. 7.1 Supply chain links 7.2 Links established based on The aim of SOMO and Arisa is to help enable structural improvements to employment, working, and living con- corporate disclosure 7.3 Links with vertically integra- ditions for workers in the Indian textile and garment industry, and in particular, for the most vulnerable worker ted manufacturers groups (such as child and adolescent workers, female workers, migrant workers, and workers with a tribal, ethnic 7.4 Company responses with or low caste background, including Dalits5). regard to the draft research findings 8. Conclusions: Beyond 29 mills Methodology – a sector risk analysis 8.1 Incidence and risks of forced For this report we conducted desk research to chart the trade flows of textiles and apparel items from Tamil Nadu labour to their export destinations, and uncover the supply chain relationships between buying companies and spinning 8.2 Lack of publicly available data and mills. supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of At the centre of this report are the findings from the large-scale field research we undertook, which looked in- supply chain responsibility depth at 29 spinning mills, located in various districts in Tamil Nadu. Our Indian counterparts conducted lengthy, 8.5 Reliance on social auditing structured interviews with 725 workers: 441 men and 284 women. This sample was enough to give us a good 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 6 research

Executive summary

1. Introduction 1.1. Why this research? insight into the Tamil Nadu textile sector, which encompasses more than 2,000 spinning mills. As well as individual 1.2. Reading guide interviews, we organised focus group discussions on three specific topics, with groups of workers from four mills. 2. Methodology 2.1. Selection of mills All the interviews took place between October 2019 and January 2020 and, to increase our understanding of the 2.2. Field research impact of the corona crisis on the sector and workers, we conducted further interviews with 15 workers in October 2.3. Supply chain research 2020. 2.4. Company review 2.5. Methodological challenges Before publication, we shared parts of the draft report with the garment manufacturers, brands and retailers that 3. Context 3.1. India’s cotton, textile, and we found to have links with the 29 investigated mills. The outcomes of this review are described in detail in the garment industry report. 3.2. The Tamil Nadu textile and garment industry For security reasons, we anonymised the mills in this report; the mills have numbers (1-30). As the mills are not 3.3. Trade unions 3.4. Social exclusion based on mentioned recognisably in the report, we have not included the mills in the review. caste and tribal background 3.5. Caste discrimination in garment We do not disclose the names of the interviewed workers. Quotes of workers have been anonimised. This is for factories and spinning mills in seccurity reasons, there is a conceivable risk of retaliation by employers and local authorities. Tamil Nadu 4. Normative framework - fun- damental labour standards Forced labour in the Tamil Nadu textile industry – existence and risk 4.1. International guidelines for responsible business conduct In this report, we present evidence of the existence and/or the risk of forced labour in the 29 spinning mills that we 4.2. Fundamental labour standards investigated. The indicators of most concern to us are: abuse of vulnerability, deception, intimidation and threats, 4.3. Forced labour indicators 4.4. Legislative developments and abusive working and living conditions, and excessive overtime. These indicators were most present in our sample existing supply chain legislation of mills. 4.5. Labour laws in India and The new findings presented in this report, combined with analysis done earlier by Arisa, SOMO and other organ- Tamil Nadu isations, allow us to conclude that the problems we found are not limited to the researched mills; forced labour is 4.6. Caste-based discrimination a major risk throughout the entire Tamil Nadu textile sector. legislation in India 5. Evidence of existence and risk of forced labour ‘I am not happy, I am feeling trapped. This makes me feel angry. I feel like running 5.1. Characteristics of the 29 away from this place but I can’t leave due to my family’s situation.’ investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced This research shows that a considerable proportion of workers in this sector are labour migrants. From various labour – using the 11 ILO lingual groups, these workers face language barriers that complicate their communications with employers, HR indicators managers, and supervisors, as well as with other workers, both socially and on work-related topics. This situation 6. Impact of COVID-19 6.1 Loss of income makes workers vulnerable and isolated. 6.2 Emergency relief 6.3 After lockdown ‘Only my salary keeps my family alive. It is very difficult to work here but there is no 6.4 Safety measures inadequate other choice.’ 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on The majority of the workers we interviewed had received incorrect information during their recruitment, about corporate disclosure the working and living conditions in their prospective jobs. Only once they began working, and were living in the 7.3 Links with vertically integra- hostel, did they discover that their wages were lower than expected, their working hours were longer than they ted manufacturers 7.4 Company responses with had been told, and annual leave was unpaid. They also found that money was being deducted from their wages regard to the draft research for food and accommodation, contrary to what they had been told during recruitment. This situation amounts to findings deceptive recruitment. 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced Interviewed workers described the huge pressure they are under to do overtime to meet their employer’s schedule labour and demands. This pressure can take all forms: scolding, getting ‘a black mark on our record’, having to do an extra 8.2 Lack of publicly available data and shift, threat of dismissal, delayed payment of wages, and deductions from wages. The picture painted by those supply chain transparency interviewed was of a climate of fear within the spinning mills. The workers internalise the threats they receive, and 8.3 Flawed business model think that whatever their employers and superiors are threatening them with will actually happen. 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing Spinning mill workers work extremely long hours. Interviewed workers spoke of 12-hour shifts, and having to work 8.6 Frontrunners bear the brunt two 8-hour shifts in a row is not uncommon. Occasionally, workers are even made to work three 8-hour shifts 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 7 research

Executive summary

1. Introduction 1.1. Why this research? back-to-back. Hostel workers reported that they are regularly woken up in the middle of the night in order to work. 1.2. Reading guide This is clear evidence of excessive and involuntary overtime. 2. Methodology 2.1. Selection of mills 2.2. Field research ‘Most of my life I am spending with the machines. There is absolutely no contact 2.3. Supply chain research with the outside world.’ 2.4. Company review 2.5. Methodological challenges 3. Context Low wages are another obvious problem. Interviewed workers said their wages were not sufficient to maintain 3.1. India’s cotton, textile, and their families. Wages are often further reduced by deductions or financial penalties imposed by management for garment industry ‘misconduct’. Many workers have no formal written confirmation of their employment terms and employment rela- 3.2. The Tamil Nadu textile and tionship, so have nothing to fall back on to claim what was promised to them. This is evidence of abusive working garment industry 3.3. Trade unions conditions, deception, and the abuse of vulnerable workers. 3.4. Social exclusion based on caste and tribal background ‘We can’t take leave, if we take leave our wages will be delayed. We have to keep 3.5. Caste discrimination in garment factories and spinning mills in working for the sake of our families and most of us have debts and loans to pay off.’ Tamil Nadu 4. Normative framework - fun- damental labour standards 4.1. International guidelines for Linking Tamil Nadu spinning mills to international garment brands responsible business conduct 4.2. Fundamental labour standards and retailers 4.3. Forced labour indicators The Tamil Nadu textile and garment industry is mainly export-oriented, producing various goods for foreign gar- 4.4. Legislative developments and existing supply chain legislation ment factories, brands, and retailers, including yarns, fabrics, apparel items and household textile. We focused, in 4.5. Labour laws in India and this research, on the European and US brands and retailers that cater to Dutch and European markets. Despite Tamil Nadu painstaking desk research, it proved difficult to establish concrete links between the 29 spinning mills we investi- 4.6. Caste-based discrimination gated, and buying companies, because of the pervasive lack of transparency about trade flows and supply chain legislation in India 5. Evidence of existence and relationships. This lack of transparency is a major issue in the global textile and garment industry. risk of forced labour 5.1. Characteristics of the 29 On the basis of our supply chain research, we were able to establish direct and/or indirect links between 10 gar- investigated mills ment brands and retailers, and the mills we investigated. SOMO and Arisa use the term ‘direct link’ to mean cases 5.2. Respondents’ profiles where yarn and/or fabric used in the brand/retailer’s products comes from one of the investigated mills. This does 5.3. Risks and evidence of forced labour – using the 11 ILO not mean that the brand/retailer has a contractual relationship with the investigated spinning mill; there may be indicators one or more suppliers and/or agents in between (for example, garment factories). We speak of indirect links be- 6. Impact of COVID-19 tween brands/retailers and the mills under investigation, in cases where a brand/retailer sources from a CMT-unit 6.1 Loss of income that is part of the same vertically integrated company as the mill we investigated but yarn/ fabrics do not come 6.2 Emergency relief 6.3 After lockdown from the investigated mill but from another spinning unit within this company; or even from a spinning unit out- 6.4 Safety measures inadequate side this company. 7. Responses to the research 7.1 Supply chain links SOMO and Arisa are of the opinion that also in the latter case, there is a responsibility to conduct ‘heightened hu- 7.2 Links established based on corporate disclosure man rights due diligence’. If labour rights and human rights issues become apparent, for example by means of CSO 7.3 Links with vertically integra- reports like the current SOMO-Arisa report, in units belonging to the same company as where branded products ted manufacturers are made regardless of the direct or indirect relation to the unit where issues are reported, then these are signals 7.4 Company responses with to be taken seriously and require the buyers to investigate and take action. regard to the draft research findings 8. Conclusions: Beyond 29 mills We are aware that we have only been able to establish a limited number of supply chain links. There are many other – a sector risk analysis corporate actors involved besides the ones mentioned in this report. 8.1 Incidence and risks of forced labour In alphabetical order these ten companies are: 8.2 Lack of publicly available data and supply chain transparency 1 Carrefour 6 Sainsbury’s 8.3 Flawed business model 2 GAP 7 Tesco 8.4 Restricted interpretation of 3 IKEA 8 The Cookie Company supply chain responsibility 4 Marc O’Polo 9 WE Fashion 8.5 Reliance on social auditing 5 NEXT 8.6 Frontrunners bear the brunt 10 Zeeman 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 8 research

Executive summary

1. Introduction 1.1. Why this research? We distinguish five ways, or scenarios, in which brands and retailers could be linked to spinning units. Note: more 1.2. Reading guide than one scenario may apply in the supply chain of a buyer company. 2. Methodology 2.1. Selection of mills 2.2. Field research Scenario 1. (direct link) The spinning units are part of the supply chain of a brand/retailer because the brand/ 2.3. Supply chain research retailer works with a sourcing model which identifies the spinning mills that should supply yarn and/or fabrics 2.4. Company review (‘nominated’ or ‘preferred’ suppliers) for the production of end-products. 2.5. Methodological challenges Example: Tesco (mill 14). 3. Context 3.1. India’s cotton, textile, and garment industry Scenario 2. (direct link) A brand/retailer is linked to a spinning unit through a garment factory - that may be based 3.2. The Tamil Nadu textile and In India or another country – that sources yarn/fabrics from the spinning unit in question. The brand/retailer out- garment industry sources the responsibility to select the yarn/fabric supplier to the first-tier supplier (a garment factory or an agent). 3.3. Trade unions 3.4. Social exclusion based on Examples: Zeeman (mill 6), WE Fashion (mill 3). caste and tribal background 3.5. Caste discrimination in garment Scenario 3. (direct link) A brand/retailer sources from a CMT unit of a vertically integrated company that also factories and spinning mills in encompasses one or more spinning units. Yarn/fabrics are supplied by the spinning unit of the umbrella company. Tamil Nadu 4. Normative framework - fun- damental labour standards Scenario 4. (indirect link) A brand/retailer sources from a CMT unit of a vertically integrated company that also 4.1. International guidelines for encompasses one or more spinning units. Yarn/fabrics are supplied by a different unit or company. responsible business conduct Examples: IKEA (mill 19), GAP (mill 7), NEXT (mill 7), Sainsbury’s (mill 7), The Cookie Company (mill 27), Marc 4.2. Fundamental labour standards O’Polo (mill 14). 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation Scenario 5. Brands and retailers source from a garment factory that sources from one of the investigated spinning 4.5. Labour laws in India and mills. Yarns and/or fabrics used in the products of the brand/retailer are, however, sourced from another supplier. Tamil Nadu There are obviously many more international brands and retailers - both big and small – that are linked to Tamil 4.6. Caste-based discrimination legislation in India Nadu spinning mills. The lack of supply chain transparency makes it impossible to identify the specifics of these 5. Evidence of existence and links. In some cases, we found brands/retailers to be sourcing from (or to have had sourcing relations with) a CMT risk of forced labour unit within a vertically integrated companies that also encompasses one of the 29 spinning units that we investi- 5.1. Characteristics of the 29 gated, without us being able to establish the full details of the sourcing relation. investigated mills One example is Carrefour (mill 7) and The Cookie Company (mill 9). 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO Summing up: the brands and retailers that we found to have direct or indirect links with the mills we researched indicators are (in alphabetical order): Tesco, WE Fashion, Zeeman (direct links), and Carrefour, GAP, IKEA, Marc O’Polo, NEXT, 6. Impact of COVID-19 Sainsbury’s, The Cookie Company (indirect links). 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate The bigger picture 7. Responses to the research Hundreds of companies must be sourcing from garment factories in Tamil Nadu, given the size of the Tamil Nadu 7.1 Supply chain links industry. The following 25 apparel importers (listed in alphabetical order) are selected on the basis of information 7.2 Links established based on corporate disclosure derived from the trade database Panjiva, particularly the number of shipments, and shipment value 2019 – 2020: 7.3 Links with vertically integra- ted manufacturers 1. Ann Taylor (US) 13. NEXT (UK) 7.4 Company responses with 2. Asda (UK) 14. Orchestra (France) regard to the draft research 15. P&P Loyal (Italy) findings 3. ASOS (UK) 8. Conclusions: Beyond 29 mills 4. G-Star Raw (The Netherlands) 16. PEPCO (Poland) – a sector risk analysis 5. GAP (US) 17. Primark (Ireland) 8.1 Incidence and risks of forced 6. Guess (US) 18. Ralph Lauren (US) labour 7. H&M (Sweden) 19. River Island (UK) 8.2 Lack of publicly available data and supply chain transparency 8. J. Crew (US) 20. Takko (Germany) 8.3 Flawed business model 9. Jockey (US) 21. Tesco (UK) 8.4 Restricted interpretation of 10. Kidiliz Group (France) 22. The Children’s Place (US) supply chain responsibility 11. Marks & Spencer (UK) 23. Tom Tailor (Germany) 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 12. Mast Industries (US) 24. Sainsbury’s (UK) 8.7 Leverage 25. Walmart (US) 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 9 research

Executive summary

1. Introduction 1.1. Why this research? For GAP, Next, Sainsbury’s and Tesco we found direct or indirect links with the investigated spinning mills or with 1.2. Reading guide the larger companies of which the investigated spinning units are part of. It is likely that the other 21 apparel im- 2. Methodology 2.1. Selection of mills porters on this list are also using yarns and fabrics produced by Tamil Nadu spinning mills. So far, we have been 2.2. Field research unable to find information about direct or indirect links between these apparel importers and a specific spinning 2.3. Supply chain research mill or spinning unit. 2.4. Company review 2.5. Methodological challenges 3. Context Frontrunners catching fire 3.1. India’s cotton, textile, and garment industry Brands and retailers that are frontrunners in terms of supply chain transparency are allowing themselves to be 3.2. The Tamil Nadu textile and linked to potential human rights and labour rights violations in their supplier factories. If these links to violations garment industry and abuses are discovered, it could result in exposure in the media, public disapproval, and reputation damage. 3.3. Trade unions Competitors that lag behind in terms of supply chain transparency may very well have human rights and labour 3.4. Social exclusion based on caste and tribal background rights violations occurring in their supply chain, but do not face the same type of exposure, and so stay quietly 3.5. Caste discrimination in garment under the radar. A level playing field does, therefore, not exist; Arisa and SOMO are calling for this to change. factories and spinning mills in Tamil Nadu 4. Normative framework - fun- Impact of the coronavirus crisis damental labour standards 4.1. International guidelines for In July 2020, we conducted additional interviews with 15 workers to hear first-hand from them how the corona responsible business conduct crisis had affected their lives. These interviews gave us an impression of the problems facing workers since the 4.2. Fundamental labour standards pandemic. None of the workers had received their regular wages during the period of lockdown. The Tamil Nadu 4.3. Forced labour indicators state government made a one-off payment of INR 1,000 (€ 12), but only six workers received the additional 4.4. Legislative developments and existing supply chain legislation promised ration cards for food. Only two of the interviewed workers were able to stay in the hostel during the 4.5. Labour laws in India and lockdown, the others were sent home. All the respondents said that, following lockdown, the workforce had been Tamil Nadu almost halved, due to a dramatic decrease in demand for products, and difficulties with delivery of inputs (cotton), 4.6. Caste-based discrimination because businesses had closed. Post-lockdown, workers who were rehired were no longer paid for overtime. legislation in India 5. Evidence of existence and Workers also described inadequate measures being taken against coronavirus, in the workplace and especially in risk of forced labour the hostels. 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles Conclusions and recommendations 5.3. Risks and evidence of forced labour – using the 11 ILO This report presents clear evidence of the existence and/or risks of forced labour in the 29 spinning facilities in indicators Tamil Nadu that we investigated. This evidence has led us to conclude that the risk of forced labour practices 6. Impact of COVID-19 throughout the Tamil Nadu textile industry is sky-high. 6.1 Loss of income 6.2 Emergency relief Foreign brands and retailers, directly or indirectly sourcing yarn and/or textiles from spinning facilities in Tamil 6.3 After lockdown 6.4 Safety measures inadequate Nadu, risk becoming linked – directly or indirectly - with such inhumane practices. 7. Responses to the research 7.1 Supply chain links For many years now, Arisa and SOMO have been reporting on the appalling labour conditions in the garment and 7.2 Links established based on textile industry in Tamil Nadu. It is deeply depressing that we are now reporting again on similar issues, but this corporate disclosure 7.3 Links with vertically integra- time in a much larger sample than previously. ted manufacturers 7.4 Company responses with In the past years, all kinds of voluntary corporate improvement schemes and activities have been implemented regard to the draft research but, given the situation we found, we can only conclude that these initiatives are not enough, as there has been findings insufficient change in the respect for, and protection of, human and labour rights. More, and stronger, measures are 8. Conclusions: Beyond 29 mills – a sector risk analysis required to bring about the necessary structural changes to labour conditions in the Tamil Nadu textile industry. 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 10 research

Executive summary

1. Introduction 1.1. Why this research? This report offers a set of concrete, detailed recommendations. The main points are: 1.2. Reading guide 2. Methodology 2.1. Selection of mills • International buying companies (brands and retailers) must acquire detailed insight into their upstream sup- 2.2. Field research ply chains, and map their suppliers beyond first-tier end-product manufacturers up to the level of spinning 2.3. Supply chain research mills/units. Using blockchain-based documentation is one option; 2.4. Company review 2.5. Methodological challenges • International buying companies (brands and retailers) must publicly disclosure full supply chain details; 3. Context • At the very least, international buying companies must insist with their first-tier suppliers (end-product 3.1. India’s cotton, textile, and garment industry manufacturers) that the latter make sure that their suppliers of yarns and fabrics (spinning mills) adhere 3.2. The Tamil Nadu textile and to international labour standards. A minimum requirement that end-product manufacturers must demand garment industry from spinning mills is that all spinning mill workers receive proper employment contracts and monthly pay 3.3. Trade unions slips; 3.4. Social exclusion based on caste and tribal background • International buying companies must gain control over and consolidate their supply chain, and establish 3.5. Caste discrimination in garment direct relationships with spinning mills. A sourcing model based on ‘preferred or nominated mills’ and ‘dec- factories and spinning mills in Tamil Nadu larations of traceability’ is one option; 4. Normative framework - fun- • International buying companies must take responsibility for the communication about human rights and damental labour standards 4.1. International guidelines for labour rights with spinning mills, the setting of standards at spinning mills and the monitoring thereof; responsible business conduct • International buying companies should base HRDD policies and practices on a progressive interpretation 4.2. Fundamental labour standards 4.3. Forced labour indicators of what constitutes the production network. If non-compliance practices - such as human rights or labour 4.4. Legislative developments and rights issues -become apparent, for example by means of CSO reports, in units belonging to the same com- existing supply chain legislation pany as where your products are made regardless of the direct or indirect relation to the problematic unit, 4.5. Labour laws in India and then these are signals to be taken seriously and require the buyer to investigate and take action; Tamil Nadu 4.6. Caste-based discrimination • As part of their human rights due diligence obligations, companies must increase and improve collaborative legislation in India efforts to prevent, mitigate, and remediate human rights and labour rights violations; 5. Evidence of existence and risk of forced labour • At all stages of the human rights due diligence processes, companies must include rights-holders, their rep- 5.1. Characteristics of the 29 resentative organisations, and other relevant stakeholders in a meaningful way; investigated mills 5.2. Respondents’ profiles • Alongside the voluntary actions companies are urgently required to take, Arisa and SOMO also advocate for 5.3. Risks and evidence of forced brands and retailers to join enforceable brand agreements; labour – using the 11 ILO indicators • SOMO and Arisa argue for legislative and regulatory measures, at both national and international levels, to 6. Impact of COVID-19 force companies to take responsibility for human rights violations happening in their international supply 6.1 Loss of income chains, and hold them accountable if they fail to do so. 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 11 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 1. Introduction 2. Methodology 2.1. Selection of mills 2.2. Field research 1.1. Why this research? 2.3. Supply chain research 2.4. Company review Tamil Nadu in south India, is a major hub in the global cotton-based textile and garment industry. Yarn and fabric 2.5. Methodological challenges producers based in Tamil Nadu supply garment factories in India and elsewhere; these factories then go on to 3. Context supply countless brands and retailers around the world. Tamil Nadu is also the location of persistent labour rights 3.1. India’s cotton, textile, and garment industry violations. 3.2. The Tamil Nadu textile and garment industry In the past years, SOMO and Arisa (formerly the India Committee of the Netherlands (ICN)) have conducted 3.3. Trade unions repeated research into labour issues within the export-oriented garment and textile industry in Tamil Nadu.6 In 3.4. Social exclusion based on caste and tribal background previous reports, SOMO and Arisa mapped out the risks and presented evidence of the ongoing violations of 3.5. Caste discrimination in garment fundamental labour rights in this region, such as child labour, discrimination on the basis of caste and tribal back- factories and spinning mills in ground, and forced labour. In other publications, we have also flagged up the problems caused by lack of supply Tamil Nadu chain transparency and ineffective grievance mechanisms7. 4. Normative framework - fun- damental labour standards 4.1. International guidelines for The main purpose of this research was to investigate the scale and risks of the forced labour - as defined by the responsible business conduct International Labour Organization (ILO) – occurring in the textile industry in Tamil Nadu. Furthermore, we aimed 4.2. Fundamental labour standards to establish and define the links that exist between international garment brands and retailers, and a number of 4.3. Forced labour indicators spinning mills in Tamil Nadu. 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and The main objective of this, and previous research projects, is to help bring about structural improvements of em- Tamil Nadu ployment, working and living conditions in the Indian textile and garment industry. We are particularly concerned 4.6. Caste-based discrimination for the groups among the labour force prone to abuse and susceptible to becoming victom of forced labour, such legislation in India 5. Evidence of existence and as: child and adolescent workers; female workers; migrant workers; workers with a distinct tribal, ethnic or caste risk of forced labour background;workers that have more than one of the aforementioned characteristics. 5.1. Characteristics of the 29 investigated mills In this research, we focus on European and US buying companies. Firstly, as organisations based in Europe, we 5.2. Respondents’ profiles work to hold companies headquartered in Europe to account. Secondly, the purchasing practices that prevail in 5.3. Risks and evidence of forced labour – using the 11 ILO the sector contribute to the continued exploitation of labour in global garment supply chains. Brands and retailers indicators have a responsibility to cease, prevent, mitigate and remediate adverse impacts on human rights occurring in their 6. Impact of COVID-19 supply chains. In practical terms, SOMO, Arisa, and our Indian counterparts will use these research findings to 6.1 Loss of income communicate and engage with industry and government actors. 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 1.2. Reading guide 7.1 Supply chain links Chapter 2 gives details of the research methodology. In chapter 3 we describe the textile and garment industry in 7.2 Links established based on India and Tamil Nadu. Chapter 4 identifies the existing normative frameworks on forced labour. Chapter 5 presents corporate disclosure 7.3 Links with vertically integra- the findings of the field research and describes the labour conditions of the workers in the spinning mills (using ted manufacturers the 11 indicators of forced labour as categories). Chapter 6 studies the interviews with workers which focused 7.4 Company responses with on the impact of the coronavirus crisis. Chapter 7 summarises the responses from brands, retailers, certification regard to the draft research initiatives and Responsible Business Conduct (RBC) initiatives. In chapters 8 and 9 we present our conclusions and findings 8. Conclusions: Beyond 29 mills recommendations. – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 12 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 2. Methodology 2. Methodology 2.1. Selection of mills 2.2. Field research The current report is the result of findings from large-scale field research and desk research into the recruitment, 2.3. Supply chain research and the working and living conditions, of intrastate and interstate migrant workers in Tamil Nadu spinning mills, 2.4. Company review focusing on the risk and existence of forced labour. 2.5. Methodological challenges 3. Context 3.1. India’s cotton, textile, and 2.1. Selection of mills garment industry 3.2. The Tamil Nadu textile and There are more than 2,000 spinning mills in the state of Tamil Nadu8 but up-to-date information about where these garment industry mills are in the various districts is hard to find. 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background For this research, we looked at 29 spinning mills. We recognise that, while this cannot be a statistically represent- 3.5. Caste discrimination in garment ative sample), we believe it is enough to draw conclusions about what is happening in the sector. factories and spinning mills in Tamil Nadu In this chapter, we explain how we came to select these 29 mills. Chapter 3 includes more information on the 4. Normative framework - fun- spinning mill industry in Tamil Nadu. damental labour standards 4.1. International guidelines for responsible business conduct We first contacted various stakeholders and corporate actors, and asked them to nominate mills to investigate. We 4.2. Fundamental labour standards approached member companies of AGT, contacts at FWF, FLA, and at various Indian NGOs. These organisations 4.3. Forced labour indicators did give us some suggestions but, unfortunately, not as many concrete leads as we had hoped. 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and We formulated a number of criteria to help identify which mills to investigate, including: geographical distribution Tamil Nadu across the districts of Tamil Nadu; the size of the workforce at the mill; the presence of intrastate and/or interstate 4.6. Caste-based discrimination migrants, and the presence of factory hostels. We also included mills that were not part of the improvement pro- legislation in India 5. Evidence of existence and jects currently being run by AGT and PST. risk of forced labour 5.1. Characteristics of the 29 We ensured that stand-alone and vertically integrated mills9 were included in our research, and we also looked at investigated mills membership of relevant mill associations and employers’ associations in Tamil Nadu. 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO Interviews were conducted at 30 facilities indicators but, during the review process, we became 6. Impact of COVID-19 aware that one of the 30 facilities (number 6.1 Loss of income 10) was not a spinning unit but a garmenting 6.2 Emergency relief Salem 6.3 After lockdown unit. Because this report focuses on spinning 6.4 Safety measures inadequate mills we have since removed all references Namakkal 7. Responses to the research to facility number 10. The other 29 spinning Erode 7.1 Supply chain links mills under investigation have retained their 7.2 Links established based on corporate disclosure original numbering (which explains why there 7.3 Links with vertically integra- is a spinning mill number 30 in the report). Coimbatore ted manufacturers 7.4 Company responses with Tirupur regard to the draft research findings 8. Conclusions: Beyond 29 mills Dindigul – a sector risk analysis THE MILLS ARE LOCATED IN 9 8.1 Incidence and risks of forced Theni DIFFERENT DISTRICS IN labour Virudhunagar 8.2 Lack of publicly available data and TAMIL NADU supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of Thoothukudi supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 13 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 2.2. Field research 2. Methodology 2.1. Selection of mills Individual interviews 2.2. Field research 2.3. Supply chain research Key to this research are the structured interviews conducted with spinning mill workers. Between October 2019 2.4. Company review and January 2020, 725 workers (women and men) were interviewed. 25 workers were interviewed at each of the 2.5. Methodological challenges 29 mills being investigated. 3. Context 3.1. India’s cotton, textile, and The selection of respondents was done by our Indian counterparts. garment industry 3.2. The Tamil Nadu textile and garment industry All the interviews were conducted outside the workplace, usually at, or near, the markets where workers shopped 3.3. Trade unions after leaving the factory premises, or near churches and temples, in public parks, etc. 3.4. Social exclusion based on caste and tribal background 3.5. Caste discrimination in garment In about half of the cases, the interviewed worker was accompanied by either co-workers, friends or relatives. factories and spinning mills in Tamil Nadu Most interviews were not done all at once, but took place over a number of short conversations. 4. Normative framework - fun- One interview was 15 minutes long but the rest lasted between 30 minutes and 1.5 hours. damental labour standards 4.1. International guidelines for responsible business conduct Focus group discussions (FGD) 4.2. Fundamental labour standards 4.3. Forced labour indicators As well as the 725 individual interviews, there were 12 focus group discussions (FGDs) on three topics: recruit- 4.4. Legislative developments and ment; disciplinary measures; and occupational health and safety. 95 workers - from four mills (number 4, 11, 16 existing supply chain legislation 4.5. Labour laws in India and and 22) - also took part in the FGDs. Tamil Nadu The FGDs were held in December 2019 and January 2020. 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and Additional interviews to gain insights into the impact of COVID-19 outbreak risk of forced labour 5.1. Characteristics of the 29 15 female workers were interviewed for a second time, between 10 and 16 July 2020, about how they fared dur- investigated mills ing the COVID-19 outbreak. One interview was conducted by telephone; the other interviews were held in-per- 5.2. Respondents’ profiles son. The in-person interviews took place at a range of locations including the worker’s home, a workers’ hostel, 5.3. Risks and evidence of forced and in public places such as teashops. These interviews took an average of 48 minutes, and all the interviewers labour – using the 11 ILO were female. indicators 6. Impact of COVID-19 All the female workers were migrant workers from districts within Tamil Nadu. Ten of the workers self-identified as 6.1 Loss of income Scheduled Caste, the other five as Most Backward Caste. 6.2 Emergency relief These 15 workers worked for eight different mills (number 11, 12, 13, 15, 16, 17, 19 and 22) . 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links Research tools 7.2 Links established based on corporate disclosure Specific sets of questions were developed for the individual interviews, the FGDs, and the COVID-19 interviews. 7.3 Links with vertically integra- SOMO, Arisa, and our Indian counterparts, worked together to formulate the questions which were originally ted manufacturers 7.4 Company responses with in English and later translated into Tamil. The interview questions took the following order: about the respond- regard to the draft research ent’s background (gender, age, marital status, level of education, caste or tribal background, mother tongue, etc.); findings work-related questions about a range of topics including recruitment, function at the mill, working hours, wage, 8. Conclusions: Beyond 29 mills deductions, leave, grievance mechanism, etc; living conditions. The interviews included open and closed questions. – a sector risk analysis 8.1 Incidence and risks of forced labour An open interview format was used for the FGDs, which were facilitated by a discussion leader, with a notetaker 8.2 Lack of publicly available data and also present. FGDs lasted between 2 and 2.5 hours. supply chain transparency 8.3 Flawed business model The COVID-19 interviews focused both on the lockdown period upto July 2020 and the re-opening of the mills 8.4 Restricted interpretation of supply chain responsibility post-lockdown. The questions examined loss of wages and bonuses, basic needs, relief provisions, layoffs, and 8.5 Reliance on social auditing the worker’s current working conditions (working hours, workload, COVID-19 measures, and the workers’ needs 8.6 Frontrunners bear the brunt and concerns). All the interviews concluded with open questions, inviting workers to comment upon their current 8.7 Leverage working life and suggest possible changes. 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 14 research

Executive summary

1. Introduction 1.1. Why this research? Interviewers 1.2. Reading guide 2. Methodology The interviews were conducted by a team of 20 interviewers, hailing from the Indian labour rights and human 2.1. Selection of mills rights organisations working with SOMO and Arisa on this research project. The team comprised 15 male inter- 2.2. Field research 2.3. Supply chain research viewers, and five female interviewers. Most interviewers spoke Tamil; four interviewers also spoke Hindi. Five 2.4. Company review interviewers used local interpreters when interviewing non-Tamil speaking respondents. All interviewers were 2.5. Methodological challenges extensively prepared and briefed. 3. Context 3.1. India’s cotton, textile, and From mid-October 2019 to mid-January 2020, the Indian research team was joined by a Dutch anthropologist10, in garment industry 3.2. The Tamil Nadu textile and a volunteer capacity, who conducted interviews with civil society experts, labour recruiters, and workers. For some garment industry interviews, this volunteer used local interpreters. 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background 2.3. Supply chain research 3.5. Caste discrimination in garment 11 factories and spinning mills in Links between the spinning mills and buying companies were established using the Open Apparel Registry (OAR) Tamil Nadu and the Panjiva trade database12(a subscription-based database which includes detailed shipment information for 4. Normative framework - fun- a number of countries, including India). damental labour standards 4.1. International guidelines for responsible business conduct 2.4. Company review 4.2. Fundamental labour standards 4.3. Forced labour indicators In December 2020, a drafts of chapters 5 (findings of the field research) was sent to a number of companies – 4.4. Legislative developments and garment manufacturers, brands and retailers – identified as having links to the investigated mills. In some cases, existing supply chain legislation because of the lack of supply chain transparency throughout the sector, we were unable to establish any links 4.5. Labour laws in India and Tamil Nadu between the investigated spinning mills and their customers. The companies – brands, retailers or garment manu- 4.6. Caste-based discrimination facturers –included in the review, were only given the names of mills that we were able to link to them. We asked legislation in India these companies if they had been, or still were, sourcing from the named spinning mill(s). We also asked for details 5. Evidence of existence and of the company’s sourcing relationship with the mill(s) (direct sourcing, indirect sourcing, nominated mills, etc.), risk of forced labour 5.1. Characteristics of the 29 when the sourcing relationship started, when the sourcing relationship ended (if applicable), and what specific investigated mills products and volumes/values per product, they sourced. 5.2. Respondents’ profiles 5.3. Risks and evidence of forced We also asked the companies to inform us of any factual mistakes in the draft chapters they were sent. labour – using the 11 ILO indicators 6. Impact of COVID-19 2.5. Methodological challenges 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown Restricted access to workers 6.4 Safety measures inadequate Because, in most spinning mills, young female workers live in hostels located inside the factory premises, these 7. Responses to the research 7.1 Supply chain links workers can only be interviewed when they are on leave or allowed to leave the premises for a (weekly or monthly) 7.2 Links established based on trip to the market. In most cases, interviewers were not able to enter the factory grounds or hostels. The oppor- corporate disclosure tunity to interview female hostel workers was, therefore, limited, and explains why female workers are underrep- 7.3 Links with vertically integra- resented in this research. ted manufacturers 7.4 Company responses with regard to the draft research Male migrant workers often stay in hostels outside the factory grounds where, in principle, they are able to receive findings visitors. While interviews could take place at the hostel, or in nearby teashops, the presence of interviewers did 8. Conclusions: Beyond 29 mills not always go unnoticed. Spinning mills are often located in small, out-of-the-way places where the coming and – a sector risk analysis goings of strangers are easily spotted. Hostels are often near the factories. During the field research, hostel -war 8.1 Incidence and risks of forced labour dens or factory supervisors sometimes noticed that the workers had ‘unknown’ visitors and came to check what 8.2 Lack of publicly available data and was going on. This form of social control made the workers and interviewers very uneasy, with interviewers forced supply chain transparency to give evasive answers to questions, or even pause the interview and continue at a later moment. 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility Time 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt As this research shows, workers in the spinning mills industry in Tamil Nadu have little free time. The working days 8.7 Leverage are long – with overtime a near standard element of a working week – and workers have very few days holiday in a 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 15 research

Executive summary

1. Introduction 1.1. Why this research? year. Many workers, especially women, also have other obligations outside of the mill. If they are married and have 1.2. Reading guide children, for example, women workers have double responsibilities. Convincing workers to spend 1 – 1.5 hours 2. Methodology 2.1. Selection of mills being interviewed is not easy. 2.2. Field research 2.3. Supply chain research 2.4. Company review Language issues 2.5. Methodological challenges Migrants from other states do not speak the local language (Tamil). The interview team consisted of Tamil and 3. Context 3.1. India’s cotton, textile, and Hindi speakers. Some interviewers made use of local interpreters. garment industry 3.2. The Tamil Nadu textile and garment industry Lack of awareness of labour rights 3.3. Trade unions Workers often have limited awareness about their own legal entitlements, and the legal obligations of their em- 3.4. Social exclusion based on caste and tribal background ployers. This lack of awareness and information, creates problems for workers, and makes field research on these 3.5. Caste discrimination in garment issues more difficult. For example, if a worker has no clue about the purpose or potential role of labour unions or factories and spinning mills in Internal Complaints Committees, it is difficult for her or him to fully answer questions on these topics. And if a Tamil Nadu worker has never been informed about a mechanism such as the Provident Fund, she or he cannot be expected to 4. Normative framework - fun- provide full and relevant answers. damental labour standards 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards Sensitivity around certain questions 4.3. Forced labour indicators Certain topics are more sensitive than others. Workers are, for example, often reluctant to answer questions about 4.4. Legislative developments and existing supply chain legislation their caste and tribal background, their level of education, and/or their age. 4.5. Labour laws in India and Tamil Nadu Sexual harassment, or issues relating to reproductive health (menstruation, pregnancy), are also not so easily dis- 4.6. Caste-based discrimination cussed. We ensured we had female interviewers in the team to cover such topics. legislation in India The privileged position of the interviewers and researchers may have had an intimidating effect on the workers. 5. Evidence of existence and risk of forced labour The researchers included educated NGO employees and students. We were aware that there would be a huge gap 5.1. Characteristics of the 29 between the interviewers and the respondents in terms of education and societal status, and had discussed this investigated mills with the interview team during the preparatory phase. 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO Job insecurity indicators 6. Impact of COVID-19 Interviewed workers were reluctant to speak freely, and critically, of their experiences in the spinning mills in case 6.1 Loss of income they lost their jobs. This reluctance was particularly marked in migrant workers. Migrants do not know who they 6.2 Emergency relief can trust, and many migrant workers said they were afraid of losing their job; Even though working conditions 6.3 After lockdown 6.4 Safety measures inadequate were bad for them, losing their job would be worse. Migrant workers were, therefore, reluctant to share informa- 7. Responses to the research tion about exploitative working arrangements and employment conditions. In these situations, it was important 7.1 Supply chain links that the interviewers took time to build trust, and clearly explained the objectives of the research and ensured 7.2 Links established based on the workers’ identities remained confidential. Such circumstances also emphasised the importance of conducting corporate disclosure interviews outside of the factory grounds/hostels, wherever possible, to make sure no supervisors, wardens or 7.3 Links with vertically integra- ted manufacturers managers were in close vicinity. 7.4 Company responses with regard to the draft research findings Lack of supply chain transparency 8. Conclusions: Beyond 29 mills The lack of supply chain transparency is a methodological obstacle of a very different kind, but still very real. This – a sector risk analysis 8.1 Incidence and risks of forced research focuses on yarn and fabric manufacturers. While some brands and retailers have taken steps in recent labour years to map, monitor and, in a few cases, disclose which spinning mills are part of their supply chain, the over- 8.2 Lack of publicly available data and whelming majority of brands and retailers have not. As a result, there is little information on where the yarn and supply chain transparency fabrics produced by the investigated spinning mills end up. While, we were able to use information provided by a 8.3 Flawed business model 13 8.4 Restricted interpretation of handful of brands, retailers, and improvement initiatives, as well as from the Panjiva database , to establish some supply chain responsibility links, it was not possible, in many cases, to determine the end users of textiles produced in Tamil Nadu. 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 16 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 3. Context 2. Methodology 2.1. Selection of mills 2.2. Field research 3.1. India’s cotton, textile, and garment industry 2.3. Supply chain research 2.4. Company review 2.5. Methodological challenges The textile and garment industries are of major importance to the Indian economy. The textile and clothing indus- 3. Context try is one of the largest sources of employment in India. From 2018 to 2019, according to the Ministry of Textiles, 3.1. India’s cotton, textile, and 14 garment industry the textile and clothing industry directly employed over 45 million people , and accounted for 12 per cent of the 3.2. The Tamil Nadu textile and country’s total export revenues15. garment industry 3.3. Trade unions India is a major producer of cotton, yarn, textiles, and garments. It is the world’s third largest exporter of cotton 3.4. Social exclusion based on 16 caste and tribal background and cotton yarn (see Graph X). Though the growth rate of ready-made garments for export has decreased , in 3.5. Caste discrimination in garment 2019, India was still the world’s seventh largest exporter of such goods, and its domestic market also continues factories and spinning mills in to expand. Tamil Nadu 4. Normative framework - fun- The major export destinations of Indian produced garments include the US, the UK, the UAE, Germany, Spain, damental labour standards 4.1. International guidelines for and France (see Graphs 2 and 3). Indian produced yarns and fabrics are mainly exported to Bangladesh and China responsible business conduct where they are processed further (see Graph 1). The EU imports most of its clothing from China and Bangladesh 4.2. Fundamental labour standards so, indirectly, yarn and fabrics produced in India end up in garments sold in European stores. Despite knowing this, 4.3. Forced labour indicators however, it is very difficult to trace back the origin of the yarn and fabrics used to produce textile garments. 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and While limited supply chain transparency makes it difficult to identify where yarn and fabrics produced in Tamil Tamil Nadu Nadu eventually end up, an increasing number of garment brands and retailers publicly disclose supplier lists. 4.6. Caste-based discrimination Unfortunately, the vast majority of brands and retailers disclose only first-tier suppliers (leaving producers of yarn legislation in India 5. Evidence of existence and and fabrics outside the scope of their disclosure), while other key industry players refrain from disclosing any kind risk of forced labour of supply chain information at all. 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators Top 5 export desitations 2019 6. Impact of COVID-19 6.1 Loss of income Cotton, cotton yarn & cotton textiles 6.2 Emergency relief Bangladesh $1.548.714.621 6.3 After lockdown China $1.043.236.397 6.4 Safety measures inadequate Sri Lanka $242.833.352 7. Responses to the research Vietnam $231.526.822 7.1 Supply chain links Egypt $216.380.057 7.2 Links established based on corporate disclosure Apparel, not knitted or crocheted 7.3 Links with vertically integra- ted manufacturers USA $2.052.794.052 7.4 Company responses with UAE $1.126.080.902 regard to the draft research UK $759.027.662 findings Germany $559.792.593 8. Conclusions: Beyond 29 mills France $354.633.387 – a sector risk analysis 8.1 Incidence and risks of forced Apparel, knitted or crocheted labour USA $2.298.885.743 8.2 Lack of publicly available data and UK $834.094.478 supply chain transparency UAE $733.359.666 8.3 Flawed business model 8.4 Restricted interpretation of Spain $481.850.736 supply chain responsibility Germany $452.881.076 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 17 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 3.2. The Tamil Nadu textile and garment industry 2. Methodology 2.1. Selection of mills There are three main garment production hubs in India: National Capital Region (NCR) comprising New Delhi and 2.2. Field research 2.3. Supply chain research the surrounding areas, Bangalore district (Karnataka), and Tirupur district (Tamil Nadu). 2.4. Company review 2.5. Methodological challenges The Tirupur garment production hub benefits from the presence of spinning mills in nearby districts, where cotton 3. Context yarns and cotton fabrics are made. 3.1. India’s cotton, textile, and garment industry 3.2. The Tamil Nadu textile and Out of approximately 3,400 spinning mills in India, more than 2,000 - 60 per cent - are in Tamil Nadu. Despite this, garment industry the state accounts for only 35 per cent of India’s total yarn production. According to the Handlooms, Handicrafts, 3.3. Trade unions Textiles & Khadi Department of the Tamil Nadu state government, productivity is low because of the obsolete 3.4. Social exclusion based on 17 caste and tribal background machinery being used by many spinning mills . Nevertheless, Tamil Nadu is still the second largest contributor to 18 3.5. Caste discrimination in garment the Indian textiles industry, accounting for 19 per cent of the country’s textile output . factories and spinning mills in Tamil Nadu Tamil Nadu has the second highest number of power looms installed, for use in the weaving process19. The state 4. Normative framework - fun- accounts for 70 per cent of the country’s cotton knitting capacity, and is home to Tirupur, the biggest knitting damental labour standards 4.1. International guidelines for cluster in India, which produces 45 per cent of total knitwear exports. There are more than 6,000 knitting and responsible business conduct garmenting units in, and around, Tirupur20. 4.2. Fundamental labour standards 4.3. Forced labour indicators According to a 2019 report by the Tamil Nadu Handlooms, Handicrafts, Textiles & Khadi Department, the Tamil 4.4. Legislative developments and 21 existing supply chain legislation Nadu spinning industry provides employment to 280,000 workers . The majority of these workers are intrastate 4.5. Labour laws in India and migrants, coming from different districts within Tamil Nadu. A growing percentage of workers are interstate mi- Tamil Nadu grants, coming from other states in India, including Kerala, Karnataka, and Andhra Pradesh. Over the past few 4.6. Caste-based discrimination years, labour migration from Odisha, Jharkhand, Bihar, Chhattisgarh, and West Bengal has also been on the rise.22 legislation in India This report focuses on the human rights and labour rights situation at 29 spinning mills based in Tamil Nadu. Spin- 5. Evidence of existence and risk of forced labour ning is the process of converting cotton or man-made fibre into yarn, one of the steps necessary to making clothes 5.1. Characteristics of the 29 and textile products out of cotton. investigated mills 5.2. Respondents’ profiles The process begins with cottonseed production. The cotton plants grow and mature and, after being harvested, the 5.3. Risks and evidence of forced labour – using the 11 ILO fibre is separated from the seed (known as ‘ginning’). After ginning, the cotton is prepared for spinning. indicators Once the cotton has been spun, the yarn is either woven or knitted into fabrics, which are then ‘finished’. Finishing 6. Impact of COVID-19 includes bleaching, dyeing, and printing. The last stage of the process is the production of garments or other textile 6.1 Loss of income items, which includes activities such as cutting, stitching and embroidery, buttoning, labelling, and packing. 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate The state of Tamil Nadu is home to a wide variety of manufacturers that dedicate themselves to one or more of the 7. Responses to the research processes that form part of the production of garments or other textile items. 7.1 Supply chain links There are small, medium, and large spinning mills. There are also stand-alone spinning mills (non-integrated), and 7.2 Links established based on horizontally and vertically integrated manufacturers. Vertically integrated firms may provide a combination of vari- corporate disclosure 7.3 Links with vertically integra- ous processes in their plants, such as spinning, weaving and/or knitting, fabric finishing, and garmenting. Horizon- ted manufacturers tally integrated firms have multiple production facilities which all focus on one single process, such as spinning. 7.4 Company responses with Some companies are both vertically and horizontally integrated. One firm may have several units all operating regard to the draft research under different names and, despite common ownership, production facilities may be registered as separate units. findings 8. Conclusions: Beyond 29 mills – a sector risk analysis Some companies have dozens of different manufacturing units, and it is not always clear how many units belong 8.1 Incidence and risks of forced to the same company, what the units are called, and where they are located. To make matters worse, facilities with labour similar names may be part of the same group or company, or may belong to separate companies altogether. 8.2 Lack of publicly available data and These complexities, and their underlying causes, are addressed in a report by Verité (2010)23. The organisation supply chain transparency 8.3 Flawed business model describes the Tamil Nadu industry as follows: 8.4 Restricted interpretation of supply chain responsibility ‘There are vast networks of factories and subcontractors, with units ranging in size from individual home-based units to 8.5 Reliance on social auditing factories with over 1,000 workers. This multitude of manufacturing facilities of all shapes and sizes has developed, in part, 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 18 research

Executive summary

1. Introduction 1.1. Why this research? as larger factories cleave themselves to avoid falling within the confines of labour legislation and, in part, as a response to 1.2. Reading guide global market pressures for flexible and infinitely expandable and contractible production capacity. A multitude of small, 2. Methodology 2.1. Selection of mills individual units can be highly flexible and responsive to market demand, since each small factory unit relates with the 2.2. Field research tiers above and below on a strictly contract basis.’ 2.3. Supply chain research 2.4. Company review Hundreds of companies must be sourcing from garment factories in Tamil Nadu, given the size of the Tamil Nadu 2.5. Methodological challenges industry. Table 5 shows, in alphabetical order, 25 large importers of apparel items produced in Tamil Nadu. This 3. Context 3.1. India’s cotton, textile, and selection of buying companies is based on shipment information found on the trade database, Panjiva, regarding garment industry the export value and number of shipments of (knitted and crocheted) apparel (HS codes 61 and 62) produced in 3.2. The Tamil Nadu textile and Tamil Nadu in 2019 and 2020. garment industry 3.3. Trade unions 3.4. Social exclusion based on Table 1: Importers of apparel items produced in Tamil Nadu caste and tribal background 3.5. Caste discrimination in garment Company Country where company is headquartered factories and spinning mills in 1 Ann Taylor US Tamil Nadu 4. Normative framework - fun- 2 Asda UK damental labour standards 3 ASOS UK 4.1. International guidelines for responsible business conduct 4 G-Star Raw Netherlands 4.2. Fundamental labour standards 5 GAP US 4.3. Forced labour indicators 6 Guess US 4.4. Legislative developments and existing supply chain legislation 7 H&M Sweden 4.5. Labour laws in India and 8 J. Crew US Tamil Nadu 4.6. Caste-based discrimination 9 Jockey US legislation in India 10 Kidiliz Group France 5. Evidence of existence and risk of forced labour 11 Marks & Spencer UK 5.1. Characteristics of the 29 12 Mast Industries US investigated mills 13 NEXT UK 5.2. Respondents’ profiles 5.3. Risks and evidence of forced 14 Orchestra France labour – using the 11 ILO 15 P&P Loyal Italy indicators 6. Impact of COVID-19 16 PEPCO Poland 6.1 Loss of income 17 Primark Ireland 6.2 Emergency relief 6.3 After lockdown 18 Ralph Lauren US 6.4 Safety measures inadequate 19 River Island UK 7. Responses to the research 20 Sainsbury’s UK 7.1 Supply chain links 7.2 Links established based on 21 Takko Germany corporate disclosure 22 Tesco UK 7.3 Links with vertically integra- ted manufacturers 23 The Children’s Place US 7.4 Company responses with 24 Tom Tailor Germany regard to the draft research findings 25 Walmart US 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 19 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 3.3. Trade unions 2. Methodology In India, trade unions face a number of restrictions and obstacles both at the stages of formation and registration, 2.1. Selection of mills as well as on their day-to-day activities. Anti-union prejudice is prevalent, and employers enjoy a dominant posi- 2.2. Field research 2.3. Supply chain research tion with little, or no, opposition. There is widespread violence against union members, and labour activists and/or 2.4. Company review striking workers are threated with criminalisation. NGOs also experience the same issues. 2.5. Methodological challenges 3. Context The International Trade Union Confederation’s (ITUC) 2020 Global Rights Index describes the disproportionate 3.1. India’s cotton, textile, and garment industry police violence used against workers calling for the payment of due wages and better working conditions. Accord- 3.2. The Tamil Nadu textile and ing to ITUC, thousands of workers have been detained for exercising their right to strike, and the report concludes garment industry that the situation in India has deterioriated since its previous annual index. Worker rights are under great pressure, 3.3. Trade unions and ITUC now classifies India as one of the ten worst countries in the world for workers24. 3.4. Social exclusion based on caste and tribal background 3.5. Caste discrimination in garment Furthermore, ITUC foresees that this situation may worsen as India is adopting a range of measures intended to factories and spinning mills in make the labour market more flexible. These measures reduce protection for workers, and undermine unions by Tamil Nadu replacing long-term employment contracts with fixed-term, flexible, temporary contract labour and other catego- 4. Normative framework - fun- ries of precarious work25. damental labour standards 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards 3.4. Social exclusion based on caste and tribal background 4.3. Forced labour indicators 4.4. Legislative developments and India is a caste-based country and, because of the caste system, people are divided into social groups. The caste existing supply chain legislation 4.5. Labour laws in India and system originated in ancient India and has, over time, developed into a system of social stratification – including Tamil Nadu assignment of rights - based on birth. This assignment of rights is unequal and hierarchical among the various 4.6. Caste-based discrimination caste groups; the highest caste group is assigned the most rights, and the lowest caste group is assigned the least, legislation in India or no, rights.26 A large part of India’s population, Dalits, are considered out-caste and, therefore, placed outside 5. Evidence of existence and risk of forced labour the caste hierarchy. They are the lowest social group and face severe discrimination, including untouchability and 5.1. Characteristics of the 29 segregation. investigated mills 5.2. Respondents’ profiles Despite a legal ban on negative caste-based discrimination, the caste system remains a major cause of the seg- 5.3. Risks and evidence of forced regation evident in Indian society. A person’s caste determines not only their living arrangements, but who they labour – using the 11 ILO 27 indicators may or may not marry, and their working opportunities (based on the jobs assigned to them). Because Dalits, 6. Impact of COVID-19 for example, are considered ‘lesser human beings’ and impure, they are assigned the dirtiest jobs, such as manual 6.1 Loss of income scavenging and disposing of dead animals.28 Being forced to take on such dirty and hazardous jobs is a form of 6.2 Emergency relief modern slavery.29 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research According to India’s constitution and legislation, the caste groups are divided into: Forward Caste, Other Backward 7.1 Supply chain links Classes, Scheduled Castes, and Scheduled Tribes. Forward Caste, also referred to as General Class or General 7.2 Links established based on Category comprise the non-backward classes who are in a relatively better social and economic position. Other corporate disclosure 7.3 Links with vertically integra- Backward Classes are described as ‘socially and educationally backward classes’. Scheduled Castes include as ted manufacturers Dalits. The term ‘Dalit’ literally means ‘broken people’, and is a self-designated term for so-called ‘untouchables’ 7.4 Company responses with who traditionally occupy the lowest place in Indian society. Scheduled Tribes are tribes or tribal communities (also regard to the draft research referred to as Adivasi). According to the 2011 census, there are over 200 million people categorised as Scheduled findings Castes in India, and over 100 million as Scheduled Tribes. 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced Additional caste groups are specified at state level. In Tamil Nadu, the category of Other Backward Classes is labour further broken down into in Backward Classes, Most Backward Classes, and Denotified Communities. Most Back- 8.2 Lack of publicly available data and ward Classes are communities which were found to be as disadvantaged as the Scheduled Castes but could not be supply chain transparency 8.3 Flawed business model classified as such, because the concept of untouchability does not exist in these groups. Denotified Communities 8.4 Restricted interpretation of were classified as criminal tribes under British colonial rule and faced severe repression. After independence, the supply chain responsibility Government of India repealed the Criminal Tribes Act, and these communities were then classified as Denotified 8.5 Reliance on social auditing Communities.30 The specification of caste groups under the set categories affects access to social schemes under 8.6 Frontrunners bear the brunt the reservation system. 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 20 research

Executive summary

1. Introduction 1.1. Why this research? The government-run reservation system facilitates access to education, employment, and political seats for disad- 1.2. Reading guide vantages castes and classes. The system is grounded in Articles 15 and 16 of the Indian Constitution, which allows 2. Methodology 2.1. Selection of mills the Government of India to set quotas to ensure any ‘socially and educationally backward classes of citizens’ are 2.2. Field research properly represented. Reservations were first arranged in the mid twentieth century, for Scheduled Castes and 2.3. Supply chain research Scheduled Tribes. In 1992 they were also arranged for Other Backward Classes.31 Both Scheduled Castes and 2.4. Company review Scheduled Tribes are eligible for quotas in education and government jobs. Despite this system of affirmative 2.5. Methodological challenges action, however, large-scale violations of human rights continue, especially affecting Scheduled Castes and Sched- 3. Context 32 3.1. India’s cotton, textile, and uled Tribes. garment industry 3.2. The Tamil Nadu textile and Scheduled Caste and Scheduled Tribes face widespread discrimination, such as: social and economic exclusion; garment industry segregation in housing; limited, or no, access to public and private services and employment; being denied access 3.3. Trade unions 33 3.4. Social exclusion based on to water, schools, jobs, and land. Because of this, many people from lower castes or tribal communities live in se- caste and tribal background vere poverty, making them vulnerable to exploitation, such as child labour, slavery, and trafficking. In caste-based 3.5. Caste discrimination in garment countries, those subjected to modern slavery are mainly from the lowest castes or tribal communities.34 factories and spinning mills in Tamil Nadu Scheduled Castes and Scheduled Tribes also regularly experience violent crime such as rape, lynching, and murder. 4. Normative framework - fun- damental labour standards The Centre of South Asian Studies, and Tokyo University of Foreign Studies state that, according to the National 4.1. International guidelines for Crime Records Bureau in India, between 2006 and 2016, there were reports of 422,799 crimes against Scheduled responsible business conduct Castes, and 81,332 crimes against Scheduled Tribes.35 Although laws are in place to counter such caste-based 4.2. Fundamental labour standards discrimination and criminal activity, these laws are poorly implemented. In September 2020, a 30-year review of 4.3. Forced labour indicators 4.4. Legislative developments and the Scheduled Castes and Scheduled Tribes Prevention of Atrocities Act (1989) found that 90 per cent of the cases existing supply chain legislation filed under the Act are acquitted in Tamil Nadu.36 4.5. Labour laws in India and Tamil Nadu The International Dalit Solidarity Network (IDSN) - a network of international human rights groups, development 4.6. Caste-based discrimination legislation in India agencies, national Dalit solidarity networks from Europe, and national platforms in caste-affected countries - ad 5. Evidence of existence and vocates for Dalit human rights. IDSN reports that Dalit women are particularly vulnerable to violence and discrim- risk of forced labour ination, yet lack access to justice precisely because of their weak socio-economic status. Women and girls from 5.1. Characteristics of the 29 caste-affected communities often face multiple forms of violence, including sexual violence, trafficking, abduction investigated mills and abuse, as well as forced and bonded labour and slavery. Dalit women are specifically used as debt slaves in 5.2. Respondents’ profiles 5.3. Risks and evidence of forced brick kilns, and the agriculture and textile industries, and are subjected to caste-based forms of slavery such as labour – using the 11 ILO manual scavenging and forced prostitution.37 indicators 6. Impact of COVID-19 6.1 Loss of income 3.5. Caste discrimination in garment factories and spinning mills in 6.2 Emergency relief 6.3 After lockdown Tamil Nadu 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links The FWF India Country Report 2019 estimates that, based on interviews with local NGOs, between 70 and 80 per 7.2 Links established based on cent of workers in the garment and textile industry in Tamil Nadu are from Scheduled Castes, Scheduled Tribes, corporate disclosure and Other Backward Classes.38 Many of these workers are thought to be interstate migrants hoping for better job 7.3 Links with vertically integra- opportunities in Tamil Nadu, coming from impoverished north and north-eastern states such as Bihar, Odisha, and ted manufacturers 39 7.4 Company responses with Uttar Pradesh, where livelihood options are scarce. regard to the draft research findings The FWF country study also highlights the role caste plays regarding the living arrangements for migrant workers 8. Conclusions: Beyond 29 mills (based on interviews with local civil society stakeholders and experts in Tamil Nadu). Upper caste workers reserve – a sector risk analysis 8.1 Incidence and risks of forced bathrooms for their use only, for example, and lower caste workers must use the remaining bathrooms. Since most 40 labour migrant workers belong to lower castes, they spend more time waiting in queues for the restroom. 8.2 Lack of publicly available data and supply chain transparency The FWF states that, with regard to gender and caste: ‘Dalit female workers are abused for their caste status by 8.3 Flawed business model being given more overtime, heavier work, cleaning tasks, being given extra work during night shifts, being denied 8.4 Restricted interpretation of supply chain responsibility leave, and being subjected to verbal and physical harassment […] at the hostels, female Dalit workers are not al- 8.5 Reliance on social auditing lowed to choose rooms or complain, and several deaths were reported after cases of sexual abuse in textile mills.’41 8.6 Frontrunners bear the brunt According to a caste discrimination expert of a Tamil Nadu-based civil society organisation, segregation based on 8.7 Leverage caste and class is prevalent in canteens and hostel accommodation. Workers from Scheduled Castes and Sched- 9. Recommendations 42 9.1 Corporate actors uled Tribes are sometimes not allowed to drink from the same water tap as workers from higher castes. 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 21 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 4. Normative framework - fundamental 2. Methodology 2.1. Selection of mills 2.2. Field research labour standards 2.3. Supply chain research 2.4. Company review 2.5. Methodological challenges 4.1. International guidelines for responsible business conduct 3. Context 3.1. India’s cotton, textile, and garment industry Over the past decade, voluntary guidelines for responsible business conduct have been developed, offering guid- 3.2. The Tamil Nadu textile and ance to companies with international supply chains, MSIs, certification schemes and RBC initiatives. The 2011 garment industry United Nations Guiding Principles on Business and Human Rights (UNGP) is a set of guidelines covering the 3.3. Trade unions duty of the state to protect human rights, the responsibility of enterprises to respect human rights, and access to 3.4. Social exclusion based on remedy for victims of business-related abuses. In accordance with UNGP 17, companies should assess how their caste and tribal background 43 3.5. Caste discrimination in garment activities and business relationships may cause, or contribute to, (risks of) human rights violations. factories and spinning mills in Tamil Nadu The OECD Guidelines for Multinational Enterprises(2011) are recommendations to multinational companies that 4. Normative framework - fun- operate in a global context. These guidelines specify that it is the responsibility of businesses to perform due dili- damental labour standards 4.1. International guidelines for gence, a process which allows enterprises to identify, prevent, and mitigate potential, and actual, adverse human 44 responsible business conduct rights and environmental impacts. The following steps are part of the due diligence process as described in the 4.2. Fundamental labour standards OECD Due Diligence Guidance for Responsible Business Conduct (RBC) : 4.3. Forced labour indicators 4.4. Legislative developments and 1. embed responsible business conduct in policies and management systems; existing supply chain legislation 4.5. Labour laws in India and 2. identify and assess adverse impacts in operations, supply chains, and business relationships; Tamil Nadu 3. cease, prevent, or mitigate adverse impacts; 4.6. Caste-based discrimination 4. track implementation and results; legislation in India 5. communicate how any identified adverse impacts are being addressed; 5. Evidence of existence and 45 risk of forced labour 6. provide for, or cooperate with, remediation when appropriate. 5.1. Characteristics of the 29 investigated mills The OECD Due Diligence Guidance for Responsible Supply Chains in the Garment and Footwear Sector (2017) 5.2. Respondents’ profiles focuses on risks to human rights in the garment and footwear sector, and details the due diligence processes to be 5.3. Risks and evidence of forced labour – using the 11 ILO undertaken by garment and footwear companies. indicators 6. Impact of COVID-19 The OECD guidance on due diligence does not include anything specifically on caste-based discrimination. Caste 6.1 Loss of income is identified as a risk factor only in Module 2. Sexual harassment and sexual and gender-based violence (SGBV) in 6.2 Emergency relief the workplace: ‘Minorities, including ethnic, religious and caste minorities, are often much more vulnerable to harass- 6.3 After lockdown 6.4 Safety measures inadequate ment and violence due to their ‘lower status’. Minorities may also have more difficulty accessing grievance mechanisms. 7. Responses to the research P. 119’.46 7.1 Supply chain links 7.2 Links established based on corporate disclosure 4.2. Fundamental labour standards 7.3 Links with vertically integra- ted manufacturers International labour standards are legal instruments drawn up by the the ILO’s constituents (governments, employ- 7.4 Company responses with ers, and workers) that set out the basic principles of workers’ rights. The governing body of the ILO has identified regard to the draft research eight ‘fundamental’ conventions47: findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 1. Freedom of Association and Protection of the Right to Organise Convention, 1948 (No. 87) 8.1 Incidence and risks of forced 2. Right to Organise and Collective Bargaining Convention, 1949 (No. 98) labour 3. Forced Labour Convention, 1930 (No. 29) (and its 2014 Protocol) 8.2 Lack of publicly available data and 4. Abolition of Forced Labour Convention, 1957 (No. 105) supply chain transparency 5. Minimum Age Convention, 1973 (No. 138) 8.3 Flawed business model 8.4 Restricted interpretation of 6. Worst Forms of Child Labour Convention, 1999 (No. 182) supply chain responsibility 7. Equal Remuneration Convention, 1951 (No. 100) 8.5 Reliance on social auditing 8. Discrimination (Employment and Occupation) Convention, 1958 (No. 111). 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 22 research

Executive summary

1. Introduction 1.1. Why this research? While India has ratified six of these eight Conventions, the two very important conventions on Freedom of Asso- 1.2. Reading guide ciation (No. 87) and Collective Bargaining (No. 98) are yet to be ratified48. However, because each of the above 2. Methodology 2.1. Selection of mills conventions is considered ‘fundamental’, ILO members are expected to respect and promote the rights promoted 2.2. Field research in all the conventions, regardless of whether or not they have ratified them. The principles enshrined in the funda- 2.3. Supply chain research mental labour conventions are also covered by the ILO Declaration on Fundamental Principles and Rights at Work 2.4. Company review (1998)49. 2.5. Methodological challenges 3. Context 3.1. India’s cotton, textile, and It is clear from previous cases, and ILO communications, that the ILO regards caste-based discrimination as be- garment industry longing to the category of ‘social origin’. The definition of discrimination, which is defined in article 1.1 of the 3.2. The Tamil Nadu textile and Discrimination (Employment and Occupation) Convention, 1958 (No. 111) includes: ‘any distinction, exclusion or garment industry preference made on the basis of race, colour, sex, religion, political opinion, national extraction or social origin, which has 3.3. Trade unions 50 3.4. Social exclusion based on the effect of nullifying or impairing equality of opportunity or treatment in employment or occupation.’ caste and tribal background 3.5. Caste discrimination in garment factories and spinning mills in 4.3. Forced labour indicators Tamil Nadu

4. Normative framework - fun- damental labour standards This research focuses on the potential risk and actual existence of forced labour in the Tamil Nadu textile indus- 4.1. International guidelines for try. To determine whether workers are trapped in forced labour situations, we use the ILO’s definition of forced responsible business conduct labour and forced labour indicators, developed to identify such situations. These indicators are based upon the 4.2. Fundamental labour standards internationally accepted definition of forced or compulsory labour specified in the ILO fundamental Forced Labour 4.3. Forced labour indicators 4.4. Legislative developments and Convention (ILO Convention 29) as:‘ work or service exacted from any person under the menace of any penalty and for existing supply chain legislation which the said person has not offered himself voluntarily’.The 2014 Protocol also states that: ‘forced or compulsory la- 4.5. Labour laws in India and bour violates the human rights and dignity of millions of women and men, girls and boys, contributes to the perpetuation Tamil Nadu of poverty and stands in the way of the achievement of decent work for all’.51. The 11 ILO indicators of forced labour 4.6. Caste-based discrimination are listed in Table 2. legislation in India 5. Evidence of existence and risk of forced labour Table 2: ILO forced labour indicators with additional descriptions by Traidcraft 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles Indicator Description by ILO Additional description by 52 5.3. Risks and evidence of forced Traidcraft labour – using the 11 ILO indicators 1. ‘Anyone can be a victim of forced labour. However, ‘When an employer takes advantage 6. Impact of COVID-19 ABUSE OF people who lack knowledge of the local language or of a worker’s vulnerable position (for 6.1 Loss of income VULNERABILITY laws, have few livelihood options, belong to a minority example, imposing excessive working 6.2 Emergency relief religious or ethnic group, have a disability or have hours or withholding wages).’ 6.3 After lockdown other characteristics that set them apart from the 6.4 Safety measures inadequate majority population are especially vulnerable to abuse 7. Responses to the research 7.1 Supply chain links and more often found in forced labour. 7.2 Links established based on The mere fact of being in a vulnerable position, for ex- corporate disclosure ample, lacking alternative livelihood options, does not 7.3 Links with vertically integra- necessarily lead a person into forced labour. It is when ted manufacturers an employer takes advantage of a worker’s vulnerable 7.4 Company responses with regard to the draft research position, for example, to impose excessive working findings hours or to withhold wages, that a forced labour 8. Conclusions: Beyond 29 mills situation may arise. Forced labour is also more likely in – a sector risk analysis cases of multiple dependency on the employer, such 8.1 Incidence and risks of forced as when the worker depends on the employer not labour 8.2 Lack of publicly available data and only for his or her job but also for housing, food and supply chain transparency for work for his or her relatives.’ 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 23 research

Executive summary

1. Introduction 1.1. Why this research? 2. ‘Deception relates to the failure to deliver what has ‘When the work is not as promised to 1.2. Reading guide 2. Methodology DECEPTION been promised to the worker, either verbally or in the worker.’ 2.1. Selection of mills writing. Victims of forced labour are often recruited 2.2. Field research with promises of decent, well-paid jobs. But once 2.3. Supply chain research they begin working, the promised conditions of work 2.4. Company review do not materialize, and workers find themselves 2.5. Methodological challenges 3. Context trapped in abusive conditions without the ability to 3.1. India’s cotton, textile, and escape. In these cases, workers have not given free garment industry and informed consent. Had they known the reali- 3.2. The Tamil Nadu textile and ty, they would never have accepted the job offer. garment industry Deceptive recruitment practices can include false 3.3. Trade unions promises regarding working conditions and wages, but 3.4. Social exclusion based on caste and tribal background also regarding the type of work, housing and living 3.5. Caste discrimination in garment conditions, acquisition of regular migration status, factories and spinning mills in job location or the identity of the employer. Children Tamil Nadu may also be recruited through false promises, made to 4. Normative framework - fun- them or their parents, concerning school attendance damental labour standards 4.1. International guidelines for or the frequency of visits by or to their parents.’ responsible business conduct 3. ‘If workers are not free to enter and exit the work Where the worker is not able to freely 4.2. Fundamental labour standards 4.3. Forced labour indicators RESTRICTION premises, subject to certain restrictions which are exit and enter the work premises, 4.4. Legislative developments and OF MOVEMENT considered reasonable, this represents a strong indica- subject to reasonable restrictions. existing supply chain legislation tor of forced labour. 4.5. Labour laws in India and Forced labourers may have their movements Tamil Nadu controlled inside the workplace, through the use of 4.6. Caste-based discrimination legislation in India surveillance cameras or guards, and outside the work- 5. Evidence of existence and place by agents of their employer who accompany risk of forced labour them when they leave the site.’ 5.1. Characteristics of the 29 investigated mills 4. ‘Victims of forced labour are often isolated in remote ‘Where the worker is denied contact 5.2. Respondents’ profiles 5.3. Risks and evidence of forced ISOLATION locations, denied contact with the outside world. with the outside world.’ labour – using the 11 ILO indicators Workers may not know where they are, the worksite 6. Impact of COVID-19 may be far from habitation and there may be no 6.1 Loss of income means of transportation available. But equally, work- 6.2 Emergency relief 6.3 After lockdown ers may be isolated even within populated areas, by 6.4 Safety measures inadequate being kept behind closed doors or having their mobile 7. Responses to the research phones or other means of communication confiscat- 7.1 Supply chain links ed, to prevent them from having contact with their 7.2 Links established based on families and seeking help.’ corporate disclosure 7.3 Links with vertically integra- ted manufacturers 5. ‘Forced labourers, their family members and close as- ‘When the worker is subjected to 7.4 Company responses with PHYSICAL sociates may be subjected to actual physical or sexual actual physical or sexual violence.’ regard to the draft research AND SEXUAL violence. Violence can include forcing workers to take findings VIOLENCE drugs or alcohol so as to have greater control over 8. Conclusions: Beyond 29 mills them. Violence can also be used to force a worker – a sector risk analysis 8.1 Incidence and risks of forced to undertake tasks that were not part of the initial labour agreement, such as to have sex with the employer 8.2 Lack of publicly available data and or a family member or, less extreme, to undertake supply chain transparency obligatory domestic work in addition to their ‘normal’ 8.3 Flawed business model tasks. Physical abduction or kidnapping is an extreme 8.4 Restricted interpretation of supply chain responsibility form of violence which can be used to take a person 8.5 Reliance on social auditing captive and then force them to work. As violence is 8.6 Frontrunners bear the brunt not acceptable as a disciplinary measure under any 8.7 Leverage circumstances, it is a very strong indicator of forced 9. Recommendations labour.’ 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 24 research

Executive summary

1. Introduction 1.1. Why this research? 6. ‘Victims of forced labour may suffer intimidation and ‘When the worker experiences intim- 1.2. Reading guide 2. Methodology INTIMIDATION threats when they complain about their conditions idation and threats (for example, if 2.1. Selection of mills AND THREATS or wish to quit their jobs. In addition to threats of they complain about their conditions).’ 2.2. Field research physical violence, other common threats used against 2.3. Supply chain research workers include denunciation to the immigration 2.4. Company review authorities, loss of wages or access to housing or 2.5. Methodological challenges 3. Context land, sacking of family members, further worsening 3.1. India’s cotton, textile, and of working conditions or withdrawal of ‘privileges’ garment industry such as the right to leave the workplace. Constantly 3.2. The Tamil Nadu textile and insulting and undermining workers also constitutes a garment industry form of psychological coercion, designed to increase 3.3. Trade unions their sense of vulnerability. The credibility and impact 3.4. Social exclusion based on caste and tribal background of the threats must be evaluated from the worker’s 3.5. Caste discrimination in garment perspective, taking into account his or her individ- factories and spinning mills in ual beliefs, age, cultural background and social and Tamil Nadu economic status.’ 4. Normative framework - fun- damental labour standards 7. ‘The retention by the employer of identity documents ‘When the worker is not able to ac- 4.1. International guidelines for RETENTION or other valuable personal possessions is an element cess their identity documents, or oth- responsible business conduct 4.2. Fundamental labour standards OF IDENTITY of forced labour if workers are unable to access these er valuable personal items, and feels 4.3. Forced labour indicators DOCUMENTS items on demand and if they feel that they cannot that they are not able to leave the job 4.4. Legislative developments and leave the job without risking their loss. In many cases, without risking the loss of them.’ existing supply chain legislation without identity documents, the worker will not be 4.5. Labour laws in India and able to obtain other jobs or access essential services, Tamil Nadu and may be afraid to ask for help from authorities or 4.6. Caste-based discrimination legislation in India NGOs.’ 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced 8. ‘Workers may be obliged to remain with an abusive ‘When wages are purposefully with- labour – using the 11 ILO WITHHOLDING OF WAGES employer while waiting for the wages that are owed held in order to force the worker to indicators 6. Impact of COVID-19 to them. The fact of irregular or delayed payment of remain, and prevent them from chang- 6.1 Loss of income wages does not automatically imply a forced labour ing employer.’ 6.2 Emergency relief situation. But when wages are systematically and de- 6.3 After lockdown liberately withheld as a means to compel the worker 6.4 Safety measures inadequate to remain, and deny him or her of the opportunity to 7. Responses to the research change employer, this points to forced labour.’ 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 25 research

Executive summary

1. Introduction 1.1. Why this research? 9. ‘Forced labourers are often working in an attempt to ‘When a worker is working in order to 1.2. Reading guide 2. Methodology DEBT BONDAGE pay off an incurred or sometimes even inherited debt. pay off a debt.’ 2.1. Selection of mills The debt can arise from wage advances or loans to 2.2. Field research cover recruitment or transport costs or from daily 2.3. Supply chain research living or emergency expenses, such as medical costs. 2.4. Company review Debts can be compounded as a result of manipulation 2.5. Methodological challenges 3. Context of accounts, especially when workers are illiterate. 3.1. India’s cotton, textile, and Debt bondage may also arise when children are re- garment industry cruited in exchange for a loan given to their parents or 3.2. The Tamil Nadu textile and relatives. Employers or recruiters make it difficult for garment industry workers to escape from the debt, by undervaluing the 3.3. Trade unions work performed or inflating interest rates or charges 3.4. Social exclusion based on caste and tribal background for food and housing. 3.5. Caste discrimination in garment factories and spinning mills in Debt bondage – or bonded labour – reflects an Tamil Nadu imbalance in power between the worker-debtor and 4. Normative framework - fun- the employer-creditor. It has the effect of binding the damental labour standards 4.1. International guidelines for worker to the employer for an unspecified period of responsible business conduct time, anything from a single season, to years, or even 4.2. Fundamental labour standards successive generations. It bears no resemblance to 4.3. Forced labour indicators taking a ‘normal’ loan from a bank or other independ- 4.4. Legislative developments and ent lender, for repayment on mutually agreed and existing supply chain legislation acceptable terms.’ 4.5. Labour laws in India and Tamil Nadu 10. ‘Forced labour victims are likely to endure living and ‘Work performed under conditions 4.6. Caste-based discrimination legislation in India ABUSIVE WORKING working conditions that workers would never freely that are degrading or hazardous and in 5. Evidence of existence and AND LIVING accept. Work may be performed under conditions that serious breach of labour law.’ risk of forced labour CONDITIONS are degrading (humiliating or dirty) or hazardous (diffi- 5.1. Characteristics of the 29 cult or dangerous without adequate protective gear), investigated mills and in severe breach of labour law. Forced labourers 5.2. Respondents’ profiles 5.3. Risks and evidence of forced may also be subjected to substandard living condi- labour – using the 11 ILO tions, made to live in overcrowded and unhealthy indicators conditions without any privacy. 6. Impact of COVID-19 6.1 Loss of income Extremely bad working and living conditions alone do 6.2 Emergency relief not prove the existence of forced labour; unfortu- 6.3 After lockdown 6.4 Safety measures inadequate nately, people may sometimes ‘voluntarily’ accept 7. Responses to the research bad conditions because of the lack of any alternative 7.1 Supply chain links jobs. However, abusive conditions should represent 7.2 Links established based on an ‘alert’ to the possible existence of coercion that is corporate disclosure preventing the exploited workers from leaving the job.’ 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with 11. ‘Forced labourers may be obliged to work excessive ‘When an employee has to work regard to the draft research EXCESSIVE OVERTIME hours or days beyond the limits prescribed by national more overtime than is allowed under findings law or collective agreement. They can be denied national law, because of some form 8. Conclusions: Beyond 29 mills breaks and days off, having to take over the shifts and of threat (for example, dismissal, or in – a sector risk analysis 8.1 Incidence and risks of forced working hours of colleagues who are absent, or by order to earn the minimum wage).’ labour being on call 24 hours a day, 7 days a week. 8.2 Lack of publicly available data and supply chain transparency The determination of whether or not overtime con- 8.3 Flawed business model stitutes a forced labour offence can be quite complex. 8.4 Restricted interpretation of supply chain responsibility As a rule of thumb, if employees have to work more 8.5 Reliance on social auditing overtime than is allowed under national law, under 8.6 Frontrunners bear the brunt some form of threat (e.g. of dismissal) or in order to 8.7 Leverage earn at least the minimum wage, this amounts to 9. Recommendations forced labour.’ 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent Source: ILO indicators of forced labour SPINNING AROUND WORKERS’ RIGHTS | 26 research

Executive summary

1. Introduction 1.1. Why this research? In some cases, according to the ILO, the presence of a single indicator may signify a situation of forced labour but, 1.2. Reading guide most of the time, it is the combination of several indicators that points to a situation of forced labour. 2. Methodology 2.1. Selection of mills 2.2. Field research 2.3. Supply chain research 4.4. Legislative developments and existing supply chain legislation 2.4. Company review In various countries, predominantly in Europe, steps have been taken towards mandatory human rights due dili- 2.5. Methodological challenges gence (mHRDD). At the European level, an important moment took place in March 2021, when the European Par- 3. Context 53 3.1. India’s cotton, textile, and liament voted in favour of a proposal for an EU Directive on due diligence. Significant developments at a national garment industry level include the German Government’s announcement that it will introduce a due diligence law. Germany is set 3.2. The Tamil Nadu textile and to ‘introduce fines for companies procuring parts or materials abroad from suppliers who fail to meet minimum garment industry human rights and environmental standards’54. In the Netherlands, in March 2021, four political parties submitted a 3.3. Trade unions bill on human rights and environmental due diligence to the Dutch Parliament55. 3.4. Social exclusion based on caste and tribal background 3.5. Caste discrimination in garment Some countries already impose legal obligations on businesses to ensure respect for human rights throughout factories and spinning mills in their operations and supply chains. Tamil Nadu

4. Normative framework - fun- 56 damental labour standards The French introduced a ‘Duty of Vigilance’ law in 2017 which established: 4.1. International guidelines for ‘a legally binding obligation for parent companies to identify and prevent adverse human rights and environmental im- responsible business conduct pacts resulting from their own activities, from activities of companies they control, and from activities of their subcontrac- 4.2. Fundamental labour standards tors and suppliers, with whom they have an established commercial relationship’57. 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation This law applies to French companies with more than 5,000 employees based in France, and companies operating 4.5. Labour laws in India and worldwide and based in France with more than 10,000 employees. Tamil Nadu Other countries have legislation that addresses particular human rights risks, and often places reporting require- 4.6. Caste-based discrimination ments on companies: legislation in India 5. Evidence of existence and risk of forced labour • The Dutch Child Labour Due Diligence Law58 requires all companies selling products on the Dutch mar- 5.1. Characteristics of the 29 ket to determine if child labour occurs in their supply chains, and develop a plan to combat it. Companies investigated mills covered by this law must submit a statement to regulatory authorities declaring they have carried out due 5.2. Respondents’ profiles 59 5.3. Risks and evidence of forced diligence related to child labour, throughout all their supply chains . labour – using the 11 ILO • The UK’s Modern Slavery Act (2015) 60 requires enterprises either domiciled, or conducting business, in indicators 6. Impact of COVID-19 the UK, and with an annual turnover of at least £36 m, to publish an annual statement, disclosing whether 6.1 Loss of income they have undertaken due diligence measures to prevent slavery or human trafficking in their supply chains. 6.2 Emergency relief 6.3 After lockdown • The Australian Modern Slavery Act, modelled after its namesake UK law, applies to companies headquar- 6.4 Safety measures inadequate tered, or operating in, Australia, and with a turnover of over AU$ 100m. Companies covered by this law are 7. Responses to the research required to report on any due diligence undertaken in relation to modern slavery61. 7.1 Supply chain links 62 7.2 Links established based on • The California Transparency in Supply Chains Act requires large manufacturers and retailers who do busi- corporate disclosure ness in California, to disclose their efforts to eradicate slavery and human trafficking in their supply chains. 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research 4.5. Labour laws in India and Tamil Nadu findings 8. Conclusions: Beyond 29 mills – a sector risk analysis Child labour 8.1 Incidence and risks of forced labour India has ratified ILO Convention no. 138 on Minimum Age, and ILO Convention no. 182 on the Worst Forms of 8.2 Lack of publicly available data and Child Labour. In 2017 it amended its own laws; the Child Labour (Prohibition and Regulation) Act, prohibits em- supply chain transparency ployment or work by children younger than 14 years in any occupation or process, before and after school hours, 8.3 Flawed business model except in family enterprises. The employment of adolescents (between 14 and 18 years of age) in hazardous occu- 8.4 Restricted interpretation of supply chain responsibility pations (section 3) and processes, during night shifts (7 pm. to 8 am.) is also prohibited under this Act. Additionally, 8.5 Reliance on social auditing adolescents can work no more than a maximum of six hours a day, and cannot be made to work overtime. The Act 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 27 research

Executive summary

1. Introduction 1.1. Why this research? has been criticised for allowing children to work in agriculture and home work, both of which can be categorised 1.2. Reading guide as ‘family enterprises’.63 2. Methodology 2.1. Selection of mills 2.2. Field research Forced and bonded labour 2.3. Supply chain research 2.4. Company review Article 23 of the Indian Constitution prohibits human trafficking and all forms of forcedlabour64. 2.5. Methodological challenges The issue of bonded labour is addressed in the Bonded Labour System (Abolition) Act (1976) where it is defined as 3. Context 3.1. India’s cotton, textile, and a system of ‘forced or partly forced labour… under which a debtor accepts an advance’ of some kind, for a pledge 65 garment industry of labour. This arrangement compels an individual to work until the debt has been paid . 3.2. The Tamil Nadu textile and garment industry In a 2019 report, Traidcraft highlighted the verdict from a 1982 Public Interest Litigation case between the Peo- 3.3. Trade unions ple’s Union for Democratic Rights and the Union of India and Others. In this case, the Supreme Court of India 3.4. Social exclusion based on caste and tribal background identified forced labour as labour paid less than the government’s stipulated minimum wage. The judgement- ar 3.5. Caste discrimination in garment gued that nobody, if they were aware of their legal entitlement, would voluntarily offer their labour for less than factories and spinning mills in the minimum wage66. Tamil Nadu 4. Normative framework - fun- damental labour standards Minimum wage and working hours 4.1. International guidelines for responsible business conduct Minimum wages in India are set at national, state, sector, and skill/occupational level. In Tamil Nadu, however, 4.2. Fundamental labour standards there is no minimum wage set for the textile sector, only a set minimum wage for apprentices in textile mills. Ap- 4.3. Forced labour indicators prentices earn a basic daily minimum wage of INR 110.00 (€ 1.27), with a dearness allowance of INR 212.70 (€ 4.4. Legislative developments and 67 existing supply chain legislation 2.45), bringing their total daily minimum wage to INR 322.70 (€ 3.72) . 4.5. Labour laws in India and Tamil Nadu The Factories Act (1948) stipulates a working week should be no more than 48 working hours, and the total num- 4.6. Caste-based discrimination ber of working hours per week - including overtime - should not exceed 60 hours. Overtime hours should be paid legislation in India 68 5. Evidence of existence and at double rate . risk of forced labour 5.1. Characteristics of the 29 investigated mills Interstate migrant workers 5.2. Respondents’ profiles The Inter-State Migrant Workmen (Regulation and Conditions of Service) Act, 1979, regulates the employment of 5.3. Risks and evidence of forced labour – using the 11 ILO interstate migrant workers recruited through contractors or middlemen, and applies to companies employing five indicators or more such workers. The Act protects interstate migrant workers, and their conditions of employment, including 6. Impact of COVID-19 working hours and fixed wages.69 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown Health and safety 6.4 Safety measures inadequate 7. Responses to the research In accordance with The Factories Act, 1948 (Act No. 63 of 1948) and the Factories (Amendment) Act, 1987 (Act 7.1 Supply chain links No. 20 of 1987), employers must ensure the health, safety, and welfare of their workers. This Act states that work- 7.2 Links established based on ers are entitled to work in safe buildings and with safe machinery, have access to toilets and drinking water, and be corporate disclosure allowed a weekly day off and paid leave. Section 41G of the Act prescribes the formation of a Safety Committee in 7.3 Links with vertically integra- ted manufacturers industrial establishments where hazardous processes take place, or where hazardous substances are handled, and 7.4 Company responses with a Managing Committee for factory canteens that employ more than 250 workers.70 regard to the draft research findings 8. Conclusions: Beyond 29 mills Employee State Insurance – a sector risk analysis 71 8.1 Incidence and risks of forced The Employee’s State Insurance (ESI) Act (1948) provides all employees (irrespective of number of hours worked, labour or the temporary or permanent nature of their contract) with benefits in case of sickness, injury, pregnancy, etc. 8.2 Lack of publicly available data and Monthly salary deductions for the ESI scheme are made at 1.75 per cent of the gross monthly wage, with the em- supply chain transparency ployer contributing 4.75 per cent of the gross monthly wage. The ESI Act also covers contract workers72. 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 28 research

Executive summary

1. Introduction 1.1. Why this research? Employees’ Provident Fund 1.2. Reading guide 2. Methodology Employees’ Provident Funds and Miscellaneous Provisions Act and Employee Provident Fund Scheme (1952) pro- 2.1. Selection of mills vide for a state retirement benefit, or pension fund, for all salaried employees who work at a workplace employing 2.2. Field research 2.3. Supply chain research more than 20 people. Employers’ and employees’ monthly contributions stand at 12 per cent of the basic monthly 73 2.4. Company review wage. 2.5. Methodological challenges 3. Context 3.1. India’s cotton, textile, and Sexual harassment at the workplace garment industry 3.2. The Tamil Nadu textile and The Sexual Harassment of Women at Workplace (Prevention, Prohibition and Redressal) Act, 2013, seeks to pro- 74 garment industry tect women from sexual harassment at their place of work. The Act provides for the establishment of Internal 3.3. Trade unions Complaints Committees (ICC) in workplaces with more than 10 employees, where victims of sexual harassment 3.4. Social exclusion based on can file complaints75. caste and tribal background 3.5. Caste discrimination in garment factories and spinning mills in Hostel accommodation in Tamil Nadu Tamil Nadu 4. Normative framework - fun- The Tamil Nadu Hostels and Homes for Women and Children (Regulation) Act was adopted in 2014. damental labour standards This law regulates hostel accommodation, including the hygiene and safety of the accommodation. It mandates all 4.1. International guidelines for hostels or lodging houses that accommodate women or children, to register and obtain a licence from the District responsible business conduct 76 4.2. Fundamental labour standards Collector (at the district in Tamil Nadu where the hostel is located). 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation Freedom of association and collective bargaining 4.5. Labour laws in India and Tamil Nadu In India, ‘freedom of association’ is regarded as a fundamental right of all citizens under Article 19 (1) of the Consti- 77 4.6. Caste-based discrimination tution of India: ‘All Citizens shall have the right … to form associations, unions or cooperative societies’. Despite legislation in India this, legal and practical restrictions are placed on freedom of association. Employers in most states are not legally 5. Evidence of existence and obliged to recognise trade unions, which limits collective bargaining.78 As a result, there is a very low level of un- risk of forced labour ionisation in India. According to FWF, less than 5 per cent of India garment workers are unionised.79 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles Another ambiguity relating to collective bargaining is the prescription of grievance redressal committees by The 5.3. Risks and evidence of forced Industrial Disputes Act (1947), which creates a legal duplication regarding worker representation. Every industrial labour – using the 11 ILO establishment with 100 or more workers is required to set up a Works Committee. This Committee must comprise indicators 6. Impact of COVID-19 representatives from both the employer and the workers’ side. The workers’ representatives must be chosen from 6.1 Loss of income among the workforce, and in consultation with their trade union (if one exists). The objective of the Works Com- 6.2 Emergency relief mittees is: ‘to promote measures for securing and preserving good relations between the employer and workers; 6.3 After lockdown and to comment upon matters in the common interest of the employer and employees, and to endeavour to reach 6.4 Safety measures inadequate a compromise in any material difference of opinion on such matters’80. 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on 81 corporate disclosure Grievance redressal committees for industrial establishments: 7.3 Links with vertically integra- Table 3 gives an overview of different grievance redressal committees prescribed for industrial establishments, ted manufacturers under different Indian labour laws. 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 29 research

Executive summary

1. Introduction 1.1. Why this research? Table 3: Factory committees required by law in India 1.2. Reading guide 2. Methodology 2.1. Selection of mills Committee Applicable Number of Representa- Meeting Tenure Legal reference 2.2. Field research factory size/ committee tives frequency of 2.3. Supply chain research no. of workers members committee 2.4. Company review Works Com- 100 or more Max. 20 Every 3 months 2 years Industrial Disputes (central) 2.5. Methodological challenges Employee mittee workers Rules 1957, sections 3. Context Employer 3.1. India’s cotton, textile, and 38-57 Grievance 20-99 workers Max. 6 Not specified Not spec- Industrial Disputes Act, garment industry Employee 3.2. The Tamil Nadu textile and Redressal Com- (not required Employer ified 2010 Amendment, section garment industry mittee if Works 9C 3.3. Trade unions Committee is 3.4. Social exclusion based on required) caste and tribal background 3.5. Caste discrimination in garment Internal 10 or more Min. 4 Employee** Not specified 3 years Sexual Harassment of factories and spinning mills in Complaints workers Women at Workplace (Pre- Tamil Nadu Committee vention, Prohibition and 4. Normative framework - fun- Redressal) Act 2013 damental labour standards 4.1. International guidelines for Safety & Health 250 workers Not specified Employee Every 3 months 2 years (3 Section 67B of the Punjab responsible business conduct Committee (general) years in Ta- Factory Rules 1952 4.2. Fundamental labour standards Employer 4.3. Forced labour indicators 50 workers (if mil Nadu) (applicable in Haryana as 4.4. Legislative developments and covered under well), section 62B of Uttar existing supply chain legislation Section 87 or Pradesh Factory Rules 4.5. Labour laws in India and Section 2(cb) 1950, Section 88C of Kar- Tamil Nadu of Factories nataka Factory Rules 1969, 4.6. Caste-based discrimination Act) 61M of the Tamil Nadu legislation in India 5. Evidence of existence and Factories Rules 1950. risk of forced labour Canteen Com- More than 250 4-10 Not specified 2 years Factories Act, 1948, 5.1. Characteristics of the 29 Employee investigated mills mittee workers Employer section 46 (2D) of the and 5.2. Respondents’ profiles Model Rules under the 5.3. Risks and evidence of forced Factories Act, 1948, sec- labour – using the 11 ILO tion 93. indicators 6. Impact of COVID-19 ** Only in the Internal Complaints Committee is the employer/factory management not represented. As gender-based violence and 6.1 Loss of income (sexual) harassment is often related to power relations in factories, only workers are represented in this committee. 6.2 Emergency relief Source: Worker-management dialogue in Indian legislation – a guidance document, FWF 2018.82 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on 4.6. Caste-based discrimination legislation in India corporate disclosure 7.3 Links with vertically integra- In India, it is illegal to negatively discriminate based on caste. The Indian Constitution (1948) prohibits discrimina- ted manufacturers tion based on grounds of religion, race, caste, sex, or place of birth (article 15); abolishes ‘untouchability’ (article 7.4 Company responses with 17); promotes educational and economic interests of Scheduled Castes, Scheduled Tribes and other weaker sec- regard to the draft research tions of society (article 46); and includes provisions related to reservation and protection of seats for members of findings 8. Conclusions: Beyond 29 mills Scheduled Castes, Scheduled Tribess and Other Backward Castes in local administrative structures and in public 83 – a sector risk analysis employment (article 16, 335, 338, 340, 341 & 342). 8.1 Incidence and risks of forced labour Because existing laws (such as the Protection of Civil Rights Act (1955) and the Indian Penal Code) proved inade- 8.2 Lack of publicly available data and supply chain transparency quate in preventing and eliminating atrocities (extreme and violent crimes) against members of Scheduled Castes 8.3 Flawed business model and Scheduled Tribes by people from non-Scheduled Castes and non-Scheduled Tribes, the Scheduled Castes/ 8.4 Restricted interpretation of Scheduled Tribes (Prevention of Atrocities – PoA) Act (1989) & Rules (1995) were enacted. The provisions under supply chain responsibility the PoA Act & Rules include: a criminal liability for specifically defined atrocities and penalisations, relief and com- 8.5 Reliance on social auditing pensations for victims of atrocities, and the establishment of special authorities to implement and monitor the 8.6 Frontrunners bear the brunt 84 8.7 Leverage Act. 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 30 research

Executive summary

1. Introduction 1.1. Why this research? Other Acts, as well as the Indian Labour Laws mentioned in Section 4.7, relevant for Scheduled Castes and Sched- 1.2. Reading guide uled Tribes (the caste groups which suffer must from human rights violations) include: the Prohibition of Employ- 2. Methodology 2.1. Selection of mills ment as Manual Scavengers and Their Rehabilitation Act (2013) (prohibits manual scavenging and manual cleaning 2.2. Field research of sewers and septic tanks, and the construction of insanitary latrines, without protective equipment); The Immor- 2.3. Supply chain research al Traffic (Prevention) Act, 1956 (ITPA) (the main legislation preventing trafficking for commercial exploitation). 2.4. Company review 2.5. Methodological challenges Different tools and guidance documents have been developed to support businesses in assessing the risk of caste- 3. Context 3.1. India’s cotton, textile, and based discrimination taking place in supply chains, and providing access to remedy when harm has been done; garment industry these include the IDSN Dalit Discrimination Check85, the ETI Base Code guidance: caste in global supply chains86, 3.2. The Tamil Nadu textile and and the Ambedkar Principles – Principles and Guidelines to address Caste Discrimination in the Private Sector by garment industry IDSN87. 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu 4. Normative framework - fun- damental labour standards 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 31 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 5. Evidence of existence and risk of 2. Methodology 2.1. Selection of mills 2.2. Field research forced labour in Tamil Nadu spinning mills 2.3. Supply chain research 2.4. Company review 2.5. Methodological challenges 3. Context ’We are working non-stop, without knowing whether it is day or night.’ 3.1. India’s cotton, textile, and garment industry 3.2. The Tamil Nadu textile and garment industry 3.3. Trade unions 5.1. Characteristics of the 29 investigated mills 3.4. Social exclusion based on caste and tribal background 3.5. Caste discrimination in garment To begin with, it should be noted that it is not easy to find information about some businesses. Sometimes, only factories and spinning mills in an address could be found for the mill being investigated, and much of the time there was no public information Tamil Nadu available about the corporate structure and/or workforce of a business. 4. Normative framework - fun- damental labour standards 4.1. International guidelines for Two of the investigated mills (14 and 17) are listed on the Bombay Stock Exchange, the others are privately incor- responsible business conduct porated. 4.2. Fundamental labour standards 4.3. Forced labour indicators Some of the investigated mills are part of vertically integrated companies. Many vertically integrated companies 4.4. Legislative developments and are involved in all stages of the garment process, from spinning to knitting and weaving, to dyeing and garmenting. existing supply chain legislation 4.5. Labour laws in India and One of the investigated companies (mill 16) is involved in every stage, from growing cotton to garmenting, while Tamil Nadu other mills dedicate themselves to spinning and weaving. Mills 7, 9, 16, 19, and 28 are vertically integrated. 4.6. Caste-based discrimination legislation in India While some of the investigated facilities only engage in spinning, they encompass multiple units. These mills are 5. Evidence of existence and risk of forced labour known as horizontally integrated facilities and are mill 8 (four spinning units), mill 14 (five spinning units), mill 17 5.1. Characteristics of the 29 (two spinning units) and mill 23 (four spinning units). investigated mills 5.2. Respondents’ profiles Mill 27 engages in ginning, spinning, weaving, knitting, dyeing, and garmenting. The company that owns the mill 5.3. Risks and evidence of forced also operates four spinning units and is, therefore, both vertically and horizontally integrated. labour – using the 11 ILO indicators Other investigated facilities appear to be stand-alone spinning units. 6. Impact of COVID-19 Mills 3, 7, 8, 14, 21, 24, and 27 are members of the BCI. 6.1 Loss of income Mills 2, 3, 6, and 14 are GOTS-certified. 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research Sustainability schemes: BCI and GOTS 7.1 Supply chain links Seven of the investigated mills are member of the Better Cotton Initiative (BCI). Four of the investigat- 7.2 Links established based on ed units hold a Global Organic Textile Standard (GOTS) certification. corporate disclosure 7.3 Links with vertically integra- ted manufacturers Better Cotton Initiative (BCI) 7.4 Company responses with The BCI describes itself as the ‘largest cotton sustainability programme in the world’. Its aim is to make regard to the draft research cotton production ‘better for the people who produce it, better for the environment it grows in and findings better for the sector’s future’88. 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced While the standards for BCI-certified cotton apply only to the beginning of the supply chain (cotton labour farming), BCI has members involved in other parts of the supply chain. At the time of writing, BCI 8.2 Lack of publicly available data and has 2,100 members worldwide, comprised of 221 brands and retailer members, 1,811 suppliers and supply chain transparency 8.3 Flawed business model manufacturers, 19 producer organisations, 32 civil society organisations and 17 associate members. 8.4 Restricted interpretation of Members are not certified. Instead, BCI Members must ‘agree’ to the Membership Code of Practice: supply chain responsibility ‘Members should act with integrity and should not partake in behaviour that poses a credibility risk 8.5 Reliance on social auditing to BCI. Examples of credibility risks would include cases of child labour, tax fraud, poor treatment of 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 32 research

Executive summary

1. Introduction 1.1. Why this research? employees or other controversial activities’. BCI does not audit its members but the BCI Secretariat 1.2. Reading guide regularly monitors and gathers information based on Google alerts and notifications from stakeholders 2. Methodology 2.1. Selection of mills and other members. If a member is found to be in breach of the Code of Practice, the matter is brought 2.2. Field research to the attention of the BCI Secretariat, and the member is given a particular period of time to explain, 2.3. Supply chain research and another period of time to resolve the matter. If the matter is not resolved within three months of 2.4. Company review receiving the formal warning, the member may be suspended. 2.5. Methodological challenges

3. Context 3.1. India’s cotton, textile, and Global Organic Textile Standard garment industry Global Organic Textile Standard (GOTS) is a textile processing standard for organic fibres, and includes 3.2. The Tamil Nadu textile and ecological and social criteria. GOTS is a ‘standard for the entire post-harvest processing (including spinning, garment industry knitting, weaving, dyeing and manufacturing) of apparel and home textiles made with certified organic fibre’. 3.3. Trade unions 89 3.4. Social exclusion based on In 2020, there were 10,388 GOTS-certified facilities around the world . caste and tribal background 3.5. Caste discrimination in garment On-site inspection and certification of processors, manufacturers, and traders, is undertaken by inde- factories and spinning mills in pendent third-party GOTS-accredited certification bodies and forms the basis of the GOTS monitoring Tamil Nadu system. Certified facilities undergo an annual on-site inspection90. GOTS requires social auditors have 4. Normative framework - fun- damental labour standards a SA8000 certificate, or undergo specific GOTS-SAI training. On-site inspection includes interviews 4.1. International guidelines for with workers and verifying records related to their wages, age, overtime, etc. Currently, off-site worker responsible business conduct interviews are not within the scope of GOTS’s work.91 GOTS has a complaints procedure in place to 4.2. Fundamental labour standards handle and resolve complaints regarding: unauthorised, false or misleading use of the GOTS logo, or 4.3. Forced labour indicators 4.4. Legislative developments and other claims related to GOTS (certification); failures or omissions in the course of the GOTS certifica- existing supply chain legislation tion procedure; any other abuses of the GOTS quality assurance or the licensing and labelling system: 4.5. Labour laws in India and and violations against the ownership and other rights of/in the GOTS. GOTS explains that complaints Tamil Nadu based on any of the criteria listed, can be made by any company, organisation or individual92. In an 4.6. Caste-based discrimination legislation in India October 2020 press release, GOTS stated that it had uncovered ‘systematic fraud abusing the Indian 93 5. Evidence of existence and government certification system of organic cotton production’ . risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.2. The respondents 5.3. Risks and evidence of forced labour – using the 11 ILO indicators For this research, we interviewed 725 randomly selected workers (25 workers per investigated mill). The ability to 6. Impact of COVID-19 contact workers played a key role in the selection of respondents, and therefore the selection of spinning mills. It 6.1 Loss of income proved difficult to contact spinning mill workers living in hostels on the factory premises, and there are limited op- 6.2 Emergency relief 6.3 After lockdown portunities for workers, especially female workers, to leave the factory premises. Female workers can only leave as 6.4 Safety measures inadequate part of organised outings to the market, which usually take place once a month. The research team could only engage 7. Responses to the research in conversations with workers who enjoyed a certain level of freedom, and were able to move around during their 7.1 Supply chain links outings. In other cases, workers were strictly monitored during their time outside the factory premises. The respond- 7.2 Links established based on corporate disclosure ents interviewed for this research enjoyed more freedom of movement than many workers in other spinning mills. 7.3 Links with vertically integra- While we do not claim these 725 respondents are a statistically representative sample, we do consider the sample ted manufacturers enough to provide an insight into the sector. 7.4 Company responses with regard to the draft research findings Male/female 8. Conclusions: Beyond 29 mills – a sector risk analysis We interviewed 725 workers: 441 men and 284 women. While women make up the majority of the labour force in 8.1 Incidence and risks of forced the Indian textile and garment industry, more male workers were interviewed. There are two reasons for this dis- labour proportionality: the research focuses on migrant workers, both from within the state of Tamil Nadu and from other 8.2 Lack of publicly available data and supply chain transparency states, and in this category women are less clearly overrepresented; the proportion of male workers is most strik- 8.3 Flawed business model ing in the group of respondents hailing from outside Tamil Nadu, where 247 out of the 253 respondents were male. 8.4 Restricted interpretation of Another reason for the overrepresentation of men in the research sample is that most female textile workers live supply chain responsibility in hostels on factory premises, and have very little freedom of movement. It is therefore more difficult to access 8.5 Reliance on social auditing these workers. 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 33 research

Executive summary

1. Introduction 1.1. Why this research? Poverty 1.2. Reading guide 2. Methodology Without exception, the interviewed workers gave their poor background as the reason for taking a job in a spin- 2.1. Selection of mills ning mill. This is as true for workers from Tamil Nadu as it is for workers from other states. The migrant workers 2.2. Field research 2.3. Supply chain research who were interviewed – even the young people - often mentioned debt as their main motivation for wanting a 2.4. Company review mill job in Tamil Nadu. In their home areas, there are very few job opportunities, with some workers from farming 2.5. Methodological challenges families saying that the dry climate has now made farming impossible. Workers at mill 28 said that agents came to 3. Context their home village, just after a cyclone had ruined crops, to recruit labourers to work in Tamil Nadu. Many workers 3.1. India’s cotton, textile, and are from landless low-caste families. garment industry 3.2. The Tamil Nadu textile and garment industry 3.3. Trade unions Labour migration 3.4. Social exclusion based on The 725 workers interviewed, came from 11 different states. While most of the workers (472) came from Tamil caste and tribal background Nadu, the figures also illustrate how yarn producers in Tamil Nadu are turning to migrant workers for their labour 3.5. Caste discrimination in garment factories and spinning mills in force. The states of Bihar, Odisha, and Uttar Pradesh are becoming important sources of migrant workers. Tamil Nadu 4. Normative framework - fun- damental labour standards 4.1. International guidelines for THE WORKERS COME FROM responsible business conduct 4.2. Fundamental labour standards 10 DIFFERENT 4.3. Forced labour indicators STATES IN INDIA 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu Uttar Pradesh 4.6. Caste-based discrimination Bihar legislation in India Assam 5. Evidence of existence and Chhattisgarh risk of forced labour 5.1. Characteristics of the 29 investigated mills Jharkhand 5.2. Respondents’ profiles 5.3. Risks and evidence of forced Gujarat labour – using the 11 ILO Odisha indicators 6. Impact of COVID-19 Maharashtra 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate Karnataka 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure Kerala6 7.3 Links with vertically integra- TAMIL NADU ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis The 472 respondents from Tamil Nadu came from 28 different districts, another sign that the spinning mill industry 8.1 Incidence and risks of forced in Tamil Nadu depends on migrant labourers. labour 8.2 Lack of publicly available data and The distances between the home districts of the workers and the location of the mills they work in varied greatly supply chain transparency but, in most cases, the distances (even within Tamil Nadu) were considerable, especially given the lack of affordable 8.3 Flawed business model 8.4 Restricted interpretation of public transport for low-paid workers. As a result, most of the workers from within Tamil Nadu, resided in mill hostels. supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 34 research

Executive summary

1. Introduction 1.1. Why this research? Tiruvallur 1.2. Reading guide Tiruvannamalai 2. Methodology Krishnagiri Vellore 2.1. Selection of mills 2.2. Field research Dharmapuri Kanchipuram 2.3. Supply chain research Salem 2.4. Company review 2.5. Methodological challenges Namakkal Viluppuram 3. Context Erode 3.1. India’s cotton, textile, and Cuddalore garment industry Perambalur 3.2. The Tamil Nadu textile and Ariyalur garment industry Nagapattinam 3.3. Trade unions Coimbatore 3.4. Social exclusion based on Tirupur Thanjavur caste and tribal background 3.5. Caste discrimination in garment Karur factories and spinning mills in Dindigul Pudukkottai Tamil Nadu THE RESPONDENTS COME FROM Theni 4. Normative framework - fun- 28 DIFFERENT DISTRICTS Madurai Sivaganga damental labour standards 4.1. International guidelines for IN TAMIL NADU Virudhunagar responsible business conduct Ramanathapuram Tirunelveli 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination legislation in India Recruitment 5. Evidence of existence and risk of forced labour Almost 60 per cent of the men and women interviewed were recruited by agents. Nearly 40 per cent heard about 5.1. Characteristics of the 29 the job and employment terms from family, friends, or acquaintances. investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced 13 workers at mill 26, and five workers at mill 27 (all women), found their jobs through Regional Skills Training labour – using the 11 ILO Centres. In some cases, this was years ago, in the early 2000s. indicators 6. Impact of COVID-19 6.1 Loss of income Language 6.2 Emergency relief 6.3 After lockdown A considerable number of the workers - 297 out of the 725 - did not speak Tamil, the language spoken in the state 6.4 Safety measures inadequate of Tamil Nadu where the investigated spinning mills are located. The mother tongues of these workers reflected 7. Responses to the research their migrant background: besides Hindi (spoken in the northern part of India) (262 workers), workers spoke As- 7.1 Supply chain links samese (an Indo-Aryan language spoken in the northeast state of Assam), Bhojpuri and Bihari (Indo-Aryan lan- 7.2 Links established based on corporate disclosure guages both spoken in Bihar), Kanada (Dravidian language spoken in the south western region of India), Malayalam 7.3 Links with vertically integra- (Dravidian language spoken in the state of Kerala), Mythili (Indo-Aryan language spoken in Bihar and Jharkand), ted manufacturers Odishi (Indo-Aryan language spoken in the state of Odisha), and Telugu (Dravidian language spoken in Andhra 7.4 Company responses with Pradesh). regard to the draft research findings 8. Conclusions: Beyond 29 mills The interview team conducted interviews in Tamil and Hindi. Some of the interviewers used local interpreters. – a sector risk analysis 8.1 Incidence and risks of forced labour Age 8.2 Lack of publicly available data and supply chain transparency Respondents were selected regardless of their age, so it is significant that 10 of the workers in this random sample 8.3 Flawed business model were only 16 or 17 years old at the time of the interviews, a sign that adolescent labour persist in this sector. 8.4 Restricted interpretation of Besides this specific group of underage workers, almost two-thirds of the respondents were between 18 and 30 supply chain responsibility 8.5 Reliance on social auditing years old. 204 workers were between 30 and 50 years, and just six workers were 50 years or older. 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 35 research

Executive summary

1. Introduction 1.1. Why this research? When the respondents were asked how old they were when they started working at their current workplace, more 1.2. Reading guide than-one tenth (76 workers - 55 female and 21 male) said they were under 18 years old. Of this group, 64 workers 2. Methodology 2.1. Selection of mills said they were between 15 and 17 years old when they started working, and 12 (in four different mills, all female) 2.2. Field research said they were younger than 15. In the other 25 investigated mills, we found no other workers who had begun 2.3. Supply chain research working so young (at least not within our sample). 2.4. Company review 2.5. Methodological challenges 3. Context 3.1. India’s cotton, textile, and Caste/tribal background garment industry 3.2. The Tamil Nadu textile and All 725 workers were asked ‘What is your caste and/or tribal background?’ but less than two-thirds (460) answered garment industry the question, while a considerable group (265) did not. In 43 cases, the non-responses were not accompanied by 3.3. Trade unions 3.4. Social exclusion based on any comment, while in the remaining 222 cases, the respondents said they did not see the importance of answer- caste and tribal background ing the question. The high number of those who did not answer may indicate sensitivities about caste. 3.5. Caste discrimination in garment The Government of India distinguishes a number of overarching caste and tribal categories: Forward Caste (FC) (or factories and spinning mills in General Class), Other Backward Class (OBC), Scheduled Castes (SC) and Scheduled Tribes (ST). Tamil Nadu Workers also self-identified as BC (Backward Caste) and MBC (Most Backward Caste). Respondents also referred 4. Normative framework - fun- 94 damental labour standards to a number of sub-categories, such as Baavri, Seethri, Johi. With the help of lists drawn up by the Government 4.1. International guidelines for of India that specify, state by state, which caste communities belong to which of the four overarching categories, responsible business conduct and following advice from two Indian experts on caste, we analysed the respondents’ answers and categorised 4.2. Fundamental labour standards them under the following main categories. A small number of answers (11) could not be categorised. None of the 4.3. Forced labour indicators 4.4. Legislative developments and interviewed workers fit in the Forward Caste/General Class category; about one third of the respondents fit in the existing supply chain legislation category of Scheduled Castes (235); another 184 fit in the category Other Backward Class (which also includes 4.5. Labour laws in India and Backward Castes (BC) and Most Backward Caste (MBC)). Tamil Nadu 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and Table 4: Caste and tribal background of interviewed workers risk of forced labour 5.1. Characteristics of the 29 Caste and Tribal category Number of workers investigated mills 5.2. Respondents’ profiles FC (Forward Caste/General Class) 0 5.3. Risks and evidence of forced OBC (Other Backward Class) 184 labour – using the 11 ILO SC (Scheduled Caste) 235 indicators 6. Impact of COVID-19 ST (Scheduled Tribe) 30 6.1 Loss of income No answer 43 6.2 Emergency relief 6.3 After lockdown Did not see the importance of answering 222 6.4 Safety measures inadequate Caste category unknown for categorisation 11 7. Responses to the research 7.1 Supply chain links TOTAL 725 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers Level of education 7.4 Company responses with regard to the draft research 25 workers described themselves as uneducated or illiterate. Four-fifths of the workers said they had gone through th th findings primary school (80), middle school (322), secondary school or 10 grade (199). 12 standard is required to pursue 8. Conclusions: Beyond 29 mills a higher education; only 52 of the interviewed workers had finished 12th standard (higher secondary school). – a sector risk analysis Seven workers said they had taken higher education, but did not specify further. Not all workers gave information 8.1 Incidence and risks of forced about their level of schooling; 40 workers did not answer the question. This may indicate sensitivity on the issue labour 8.2 Lack of publicly available data and of education. supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 36 research

Executive summary

1. Introduction 1.1. Why this research? Hostel workers 1.2. Reading guide 2. Methodology The vast majority of the interviewed workers (625) lived in hostels. Residing in a hostel is often a requirement 2.1. Selection of mills imposed by management, and not necessarily the worker’s choice. We distinguished that there were hostels lo- 2.2. Field research 2.3. Supply chain research cated at the factory premises (481 respondents), and hostels outside the factory premises (144 respondents). 100 2.4. Company review workers had arranged other accommodation and did not live in hostels. 2.5. Methodological challenges 3. Context 3.1. India’s cotton, textile, and Employment duration garment industry 3.2. The Tamil Nadu textile and All respondents were asked how long they had been working at their current workplace. 16 workers did not an- garment industry swer the question; 66 workers said they had been working there for up to a year. 247 workers said they had been 3.3. Trade unions at their workplace for between one and three years, and 165 said they had been working there between three and 3.4. Social exclusion based on six years. About one-third (231) said they had been working more than six years at their current workplace. caste and tribal background 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu 4. Normative framework - fun- 5.3 Risks and evidence of forced labour – using the 11 ILO indicators damental labour standards 4.1. International guidelines for We have summarised and grouped our research findings using the 11 forced labour indicators as definied by the responsible business conduct 95 4.2. Fundamental labour standards ILO as headings: 4.3. Forced labour indicators 4.4. Legislative developments and 1. Abuse of vulnerability existing supply chain legislation 2. Deception 4.5. Labour laws in India and Tamil Nadu 3. Restriction of movement 4.6. Caste-based discrimination 4. Isolation legislation in India 5. Physical and sexual violence 5. Evidence of existence and 6. Intimidation and threats risk of forced labour 5.1. Characteristics of the 29 7. Retention of identity documents investigated mills 8. Withholding of wages 5.2. Respondents’ profiles 9. Debt bondage 5.3. Risks and evidence of forced 10. Abusive working and living conditions labour – using the 11 ILO 11. Excessive overtime indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 1. Abuse of vulnerability 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links Abuse of vulnerability 7.2 Links established based on ‘Anyone can be a victim of forced labour. However, people who lack knowledge of the local corporate disclosure language or laws, have few livelihood options, belong to a minority religious or ethnic group, 7.3 Links with vertically integra- have a disability or have other characteristics that set them apart from the majority popula- ted manufacturers 7.4 Company responses with tion are especially vulnerable to abuse and more often found in forced labour. regard to the draft research findings The mere fact of being in a vulnerable position, for example, lacking alternative livelihood 8. Conclusions: Beyond 29 mills options, does not necessarily lead a person into forced labour. It is when an employer takes – a sector risk analysis 8.1 Incidence and risks of forced advantage of a worker’s vulnerable position, for example, to impose excessive working hours or labour to withhold wages, that a forced labour situation may arise. Forced labour is also more likely in 8.2 Lack of publicly available data and cases of multiple dependency on the employer, such as when the worker depends on the em- supply chain transparency ployer not only for his or her job but also for housing, food and for work for his or her relatives.’ 8.3 Flawed business model ILO forced labour indicators96 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 37 research

Executive summary

1. Introduction 1.1. Why this research? Interviewing the 725 respondents for this research gave us plenty of insight into the background of textile workers 1.2. Reading guide in Tamil Nadu, and provided ample evidence that these are vulnerable workers, whose vulnerability is purposely 2. Methodology 2.1. Selection of mills abused during, both during recruitment and employment. Here, we detail the factors/sub-indicators (including 2.2. Field research some defined by the ILO) used to identify the vulnerable among the current labour force in the Tamil Nadu textile 2.3. Supply chain research industry. As well as lack of knowledge of the local language or laws, and limited livelihood options, we found these 2.4. Company review workers to have other characteristics that set them apart (from the majority population). Some of the characteris- 2.5. Methodological challenges tics identified (such as age, belonging to a tribal and/or caste group) overlap and reinforce one another. 3. Context 3.1. India’s cotton, textile, and garment industry 3.2. The Tamil Nadu textile and Workers have limited livelihood options garment industry Tamil Nadu textile and garment producers look for workers from beyond the immediate vicinity of the spinning 3.3. Trade unions mill or garment factory. Recruitment takes place in various districts across Tamil Nadu and, increasingly, in oth- 3.4. Social exclusion based on caste and tribal background er states. This research identifies the states of Bihar, Odisha, and Uttar Pradesh as important bases for migrant 3.5. Caste discrimination in garment workers. Other sources also point to the north and northeast of India as sources of migrant labour97. Assam and factories and spinning mills in Jharkhand were also mentioned by some interviewed workers. Tamil Nadu 4. Normative framework - fun- damental labour standards Many spinning mills have outsourced worker recruitment to agents. These agents, or recruiters, trav- 4.1. International guidelines for el to areas characterised by poverty and a lack of job opportunities, often as a result of natural disasters responsible business conduct (such as drought and floods) or conflict (including armed conflicts). Without exception, the workers- inter 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research

findings r 8. Conclusions: Beyond 29 mills e l – a sector risk analysis i D g if 8.1 Incidence and risks of forced i f e o e g labour re a n nt ngu 8.2 Lack of publicly available data and & la supply chain transparency ca 8.3 Flawed business model ste 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 38 research

Executive summary

1. Introduction 1.1. Why this research? viewed cited their poor backgrounds – the lack of job opportunities at home - as the main reason for- work 1.2. Reading guide ing in the spinning mill. Some workers from farming families said that the dry climate had now made farming 2. Methodology 98 2.1. Selection of mills impossible, and others said their villages were visited by recruiting agents after a cyclone ruined their crops. 2.2. Field research 2.3. Supply chain research 2.4. Company review Low educational background 2.5. Methodological challenges Poor education was also cited as a reason for working in the mill, and compounded the limited options avail- 3. Context 3.1. India’s cotton, textile, and able to those workers from areas where there were few job opportunities. 25 of the 725 workers de- garment industry scribed themselves as ‘uneducated’ or ‘illiterate’. Four-fifths of workers said they had gone through pri- 3.2. The Tamil Nadu textile and mary school, middle school or secondary school (10th grade). 12th standard is required to pursue any garment industry form of higher education. 52 of the interviewed workers said they had finished 12th standard (higher sec- 3.3. Trade unions ondary school), but only seven said they had achieved ‘higher education’ (without specifying further). 3.4. Social exclusion based on caste and tribal background 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu LEVEL OF EDUCATION 4. Normative framework - fun- OF RESPONDENTS damental labour standards 4.1. International guidelines for 25 80 52 7 responsible business conduct 322 199 40 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and ? Tamil Nadu Illiterate Primary Middle Secondary Higher Higher No answer 4.6. Caste-based discrimination secondary legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 Indebtedness investigated mills 5.2. Respondents’ profiles Many workers in the Tamil Nadu textile industry come from very poor and indebted families. The wages they earn 5.3. Risks and evidence of forced from working in the spinning mills are used to pay off loans and debts. In some cases, the worker’s family used labour – using the 11 ILO advances on future wages – given by, or via, the recruiting agents - to settle urgent debts. Once employed, workers indicators had to compensate for these advances (for more information see paragraph on debt bondage). 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief ‘Only my salary keeps my family alive. I know this is very difficult, to work here but 6.3 After lockdown there is no other choice.’. 6.4 Safety measures inadequate 7. Responses to the research Female worker (age 22), mill 17. 7.1 Supply chain links 7.2 Links established based on ‘Due to drought in my home place, there are no job opportunities there. I am here to corporate disclosure 7.3 Links with vertically integra- support my family. With the money I earn, my family can buy food. I am also trying ted manufacturers to pay off my debt.’ 7.4 Company responses with Male worker (age 32), mill 19. regard to the draft research findings 8. Conclusions: Beyond 29 mills ‘I come from a poor family. Because of my family situation, I am here. I have two – a sector risk analysis sisters who are not yet married. I have to work for their marriage. After my sisters’ 8.1 Incidence and risks of forced labour marriage, I will think about my own life.’ 8.2 Lack of publicly available data and Male worker (age 32), mill 19. supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of ‘We have a loan to pay off so me and my husband are working here. My daughter supply chain responsibility is living with her grandparents. Once in three months, just one or two days, I will go 8.5 Reliance on social auditing and see her.’ 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 39 research

Executive summary

1. Introduction 1.1. Why this research? Female worker (age 30), mill 20. 1.2. Reading guide 2. Methodology 2.1. Selection of mills ‘I have three children, it is painful for me to live away from my children. My husband 2.2. Field research is a drunkard and everyday he used to beat me. To protect me from this situation I 2.3. Supply chain research 2.4. Company review stay in the hostel. There are so many difficulties in the mill. Sometimes I do not know 2.5. Methodological challenges how to carry on. I am bearing everything because of my children.’ 3. Context Female worker (age 26), mill 21. 3.1. India’s cotton, textile, and garment industry 3.2. The Tamil Nadu textile and garment industry 3.3. Trade unions Caste and tribal background 3.4. Social exclusion based on 460 of the 725 workers answered the question ‘What is your caste and/or tribal background?’ with specific in- caste and tribal background formation, usually referring to the main caste and tribal categories defined by the Government of India. A number 3.5. Caste discrimination in garment factories and spinning mills in of workers used other category names such as Most Backward Caste and Backward Caste, or mentioned specific Tamil Nadu sub-categories, such as Baavri, Seethri, and Johi. 4. Normative framework - fun- damental labour standards None of the interviewed workers belong to the so-called Forward Castes (FC). The majority of the respondents 4.1. International guidelines for responsible business conduct who answered the specific question on caste fell into the categories of Scheduled Castes and Other Backward 4.2. Fundamental labour standards Classes. 4.3. Forced labour indicators 4.4. Legislative developments and 265 respondents did not answer this question, possibly because of sensitivities around the issue of caste and tribal existing supply chain legislation 4.5. Labour laws in India and background. Tamil Nadu 4.6. Caste-based discrimination The caste and tribal background of a worker is unfortunately reflected in their working conditions and position in legislation in India the mill. Based on our sample, we found that workers with a Scheduled Tribes background worked longer hours 5. Evidence of existence and than workers from other backgrounds. We were also told that Schedule Caste female workers are made to do the risk of forced labour 5.1. Characteristics of the 29 dirtier chores in the hostels, and it was hinted that the job of cleaning the shared toilets in the hostels ‘automati- investigated mills cally’ falls to the women from lower backgrounds. 5.2. Respondents’ profiles 5.3. Risks and evidence of forced Female workers from Scheduled Tribes and Scheduled Castes, and/or with a migration background are also par- labour – using the 11 ILO indicators ticularly at risk of becoming the target of (sexual) harassment. This topic arose in a focus group discussion. 6. Impact of COVID-19 6.1 Loss of income ‘We are not at ease in the factory premises. When the machine is on, the male co- 6.2 Emergency relief 6.3 After lockdown workers and the supervisor will make all sorts of comments about us. Even at the 6.4 Safety measures inadequate hostel this happens. We are at fear all the time. We always try to be in the crowd 7. Responses to the research to keep ourselves safe. Whether the management, the supervisors, or our co-male 7.1 Supply chain links 7.2 Links established based on workers, they do not treat us fairly. Particularly the Scheduled Castes, Scheduled corporate disclosure Tribes and migrants [workers] are the ones harassed the highest.’ 7.3 Links with vertically integra- Focus Group Discussion (FGD), spinning mill 16. ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills Age – a sector risk analysis Previous studies by SOMO and Arisa have presented evidence of the use of child labour in the spinning mills in 8.1 Incidence and risks of forced 99 labour Tamil Nadu . This current research shows that the recruitment and employment of children remains a pervasive 8.2 Lack of publicly available data and issue in the Tamil Nadu textile industry. Child labour is forbidden by Indian law which applies to children upto 14 supply chain transparency years. Adolescents (between 14 and 17 years) are allowed to work, but under very strict conditions; they cannot 8.3 Flawed business model work more than six hours a day and are not allowed to work night shifts. The minimum age at which adolescents 8.4 Restricted interpretation of supply chain responsibility are allowed to undertake hazardous work is 18 years. 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 40 research

Executive summary

1. Introduction 1.1. Why this research? The ILO fundamental convention concerning the minimum age for the admission to employment (Convention 138) 1.2. Reading guide stipulates that the minimum age shall not be less than 15 years. According to the convention on the worst forms 2. Methodology 2.1. Selection of mills of child labour (Convention 182), hazardous work is forbidden for those under the age of 18. In line with these 2.2. Field research conventions, SOMO and Arisa have used the age categories: 15-18 years and younger than 15 years to categorise 2.3. Supply chain research the findings related to young workers in the investigated spinning mills. 2.4. Company review 2.5. Methodological challenges The working environment in spinning mills is very often strenuous, hazardous, unhealthy, and dangerous, and the 3. Context 3.1. India’s cotton, textile, and conditions prescribed by Indian law are not met. All the adolescents interviewed said they worked more than six garment industry hours a day and all, except one, said they worked night shifts100. In other words, the adolescent workers are treated 3.2. The Tamil Nadu textile and as adult workers. Children under the age of 18 are particularly vulnerable to other forms of exploitation – such as garment industry false promises, cheating, harassment, confinement, strict supervision - that come with working in sub-standard 3.3. Trade unions 3.4. Social exclusion based on spinning mills and living in mill-run hostels. caste and tribal background 3.5. Caste discrimination in garment 15-17 years factories and spinning mills in 31-49 <15 y years Tamil Nadu 12 W 64 WORKERS 4. Normative framework - fun- damental labour standards 204 4.1. International guidelines for 18-30 WORKERS years responsible business conduct >50 y 6 W 4.2. Fundamental labour standards AGE 4.3. Forced labour indicators 4.4. Legislative developments and AT THE BEGINNING OF existing supply chain legislation EMPLOYMENT 4.5. Labour laws in India and 439 Tamil Nadu WORKERS 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 Younger than 18 at the time of the interview investigated mills We identified 10 respondents younger than 18 at the time of the interviews: one girl of 16 years, and nine girls of 5.2. Respondents’ profiles 5.3. Risks and evidence of forced 17 years. These 10 adolescents worked at five different mills (mills: 11, 12, 13, 15, and 16). labour – using the 11 ILO indicators Younger than 18 at the start of employment at the mill 6. Impact of COVID-19 All the workers were asked how old they were when they had begun working at their current workplace. 76 work- 6.1 Loss of income 6.2 Emergency relief ers (55 female and 21 male) said they had started working before they were 18 years old. 6.3 After lockdown 15 of the investigated spinning mills employed workers who were below the age of 18 when they began working 6.4 Safety measures inadequate (mills: 2, 3, 9, 11, 12, 13, 14, 15, 16, 17, 19, 20, 22, 23 and 27). Three of the workers employed at mills 23 and 27, 7. Responses to the research are now in their thirties and forties, so did not recently start working at these mills (the mill’s current customers 7.1 Supply chain links 7.2 Links established based on cannot be held responsible for this past child labour). corporate disclosure 7.3 Links with vertically integra- Younger than 15 at the start of employment at the mill ted manufacturers Among the 76 workers who began working in the spinning mill before they were 18 years old, 12 were younger 7.4 Company responses with than 15 at the time of joining the workforce (11 girls started working aged 14, one girl started working aged 13). regard to the draft research findings The mills concerned are mills 11, 12, 13, and 16. These very young workers are all girls, and said they began work 8. Conclusions: Beyond 29 mills because of the need to pay off debt. – a sector risk analysis 8.1 Incidence and risks of forced Younger than 18 in the mill labour 8.2 Lack of publicly available data and The 725 respondents were asked the age of the youngest member of the current workforce at their mill. More than supply chain transparency one-third (254) said there were workers below the age of 18 in their mill, incriminating 21 mills (mills: 1, 2, 3, 6, 8, 8.3 Flawed business model 9, 11, 12, 13, 15, 16, 17, 19, 20, 21, 22, 23, 24, 25, 28, and 29). 8.4 Restricted interpretation of supply chain responsibility Children of 14 in the mill 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt Of the 21 spinning mills reported to have members of the workforce younger than 18, five (mills: 11, 12, 13, 16 8.7 Leverage and 17) had workers who also reported101 that the youngest workers there were 14 years old. (Note, this is at the 9. Recommendations time of the interviews.) 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 41 research

Executive summary

1. Introduction 1.1. Why this research? Some factors make it hard to know the exact details and scale of the problem. The research team found, for ex- 1.2. Reading guide ample, that young workers were reluctant to share their true age, and the researchers believed that not all the 2. Methodology 2.1. Selection of mills interviewed workers gave their correct age. This is something that Arisa and SOMO have seen before, in other 102 2.2. Field research field research projects . 2.3. Supply chain research 2.4. Company review Furthermore, it proved particularly difficult to speak to young workers, as some did not want to participate in the 2.5. Methodological challenges interviews. The interviewers gained the impression that these workers felt that sharing information about their 3. Context 3.1. India’s cotton, textile, and age, along with details of their employment, might get them into trouble. garment industry 3.2. The Tamil Nadu textile and ‘My father is a drinker, he will not do any work and always fights with my mother. In garment industry 3.3. Trade unions this situation I have to work for my family and take care of my brother’s education. 3.4. Social exclusion based on When I see my mother, I see the injuries on her body.’. caste and tribal background Girl worker, mill 11 (age 14 at the start of her employment, now 17). 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu ‘I dropped out of school in class 8 due to my father’s loan problem. To support my 4. Normative framework - fun- parents, me and my sister are working in this mill.’ damental labour standards 4.1. International guidelines for Female worker, mill 11, (age 15 at the start of her employment, now 20). responsible business conduct 4.2. Fundamental labour standards ‘My family is very poor. I have five younger sisters. One sister is working with me 4.3. Forced labour indicators 4.4. Legislative developments and here. The rest are staying with my parents. Only with our income the whole family is existing supply chain legislation having food.’ 4.5. Labour laws in India and Girl worker, mill 12 (age 15 at the start of her employment, now 17). Her Tamil Nadu 4.6. Caste-based discrimination younger sister is also working at this spinning mill. legislation in India 5. Evidence of existence and risk of forced labour Workers lack of knowledge of local language 5.1. Characteristics of the 29 investigated mills Workers from far-off places and other states are particularly vulnerable to abuse in the Tamil Nadu spinning mills 5.2. Respondents’ profiles because they are a long way from home and don’t speak the local language. 5.3. Risks and evidence of forced labour – using the 11 ILO A considerable number of workers in our sample - 297 out of 725 - did not speak Tamil. A large group of workers indicators spoke Hindi, not necessarily the language spoken by their employers. The owners of the spinning mills are mainly 6. Impact of COVID-19 6.1 Loss of income from Tamil Nadu and speak Tamil. Some workers explicitly described the problems they have communicating with 6.2 Emergency relief their employers, including the supervisors who give the instructions on the work floor. 6.3 After lockdown 6.4 Safety measures inadequate Table 5: Languages spoken by interviewed workers 7. Responses to the research 7.1 Supply chain links Language spoken Number of respondents 7.2 Links established based on corporate disclosure Tamil 428 7.3 Links with vertically integra- ted manufacturers Hindi 262 7.4 Company responses with Bhojpuri 11 regard to the draft research Malayalam 6 findings 8. Conclusions: Beyond 29 mills Bihari 6 – a sector risk analysis Assamese 16 8.1 Incidence and risks of forced labour Telugu 2 8.2 Lack of publicly available data and Kanada 1 supply chain transparency Odishi 1 8.3 Flawed business model 8.4 Restricted interpretation of Mythili 1 supply chain responsibility More than one language spoken 9 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 42 research

Executive summary

1. Introduction 1.1. Why this research? No employment contracts 1.2. Reading guide 2. Methodology The vast majority of workers - 568 out of 725 – had not signed a contract or appointment letter, and only seven of 2.1. Selection of mills the workers received payslips103. The workers who did receive payslips worked at mill 23 and mill 14. 2.2. Field research 2.3. Supply chain research 2.4. Company review This problem regarding lack of documents is not limited to one or even a few mills, it is a problem across the board. 2.5. Methodological challenges The worst mills are 1, 2, 3, 4, 6, 7, 8, 9, 12, 14, 18, 19, 20, 21, 22, 28, 29 and 30. In these 18 mills, none of the 3. Context interviewed workers had signed a contract or appointment letter. Mills 13, 24, 25, 26 and 27 were better; in these 3.1. India’s cotton, textile, and mills, the majority of the interviewed workers had signed a contract. garment industry 3.2. The Tamil Nadu textile and garment industry Without contracts or appointment letters, workers have no information about their terms of employment. Work- 3.3. Trade unions ers lack any written confirmation about what was promised to them during recruitment. Moreover, they have no 3.4. Social exclusion based on written proof of employment and, without payslips, have no knowledge of any deductions from their wages, their caste and tribal background 3.5. Caste discrimination in garment social security eligibility, salary structure, etc. factories and spinning mills in Tamil Nadu 4. Normative framework - fun- Lack of knowledge on social security damental labour standards Because so many workers lack written information about their employment terms, the majority of those inter- 4.1. International guidelines for responsible business conduct viewed did not have a clue whether or not they were enrolled in social security schemes. 4.2. Fundamental labour standards We questioned workers about the two most important social security schemes for workers in the organised sector 4.3. Forced labour indicators in India: Employees’ Provident Fund (EPF) and Employees’ State Insurance scheme (ESI). Employers are obliged by 4.4. Legislative developments and law to contribute to these funds. existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 600 workers did not know whether their employer contributed to EPF, and 559 did not have, or were not aware of 4.6. Caste-based discrimination having, an EPF number. Only 63 workers were able to confirm that their employer contributed to the EPF. legislation in India 5. Evidence of existence and 484 workers did not know whether their employer contributed to the ESI scheme, and 610 workers did not have, risk of forced labour 5.1. Characteristics of the 29 or were not aware of having, an ESI number. investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced Positive example: At mill 24, all the interviewed workers said contributions were made to EPF labour – using the 11 ILO indicators and ESI, and all had EPF and ESI cards/numbers. 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate WHAT RESPONDENTS SAID ABOUT 7. Responses to the research SOCIAL SECURITY 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers ? ? 7.4 Company responses with regard to the draft research No findings Yes 8. Conclusions: Beyond 29 mills 600 484 Yes – a sector risk analysis No 8.1 Incidence and risks of forced 132 labour 63 8.2 Lack of publicly available data and 109 supply chain transparency Does your employer 62 Does your employer 8.3 Flawed business model contribute to the contribute to the Employees’ 8.4 Restricted interpretation of State Insurance scheme (ESI) supply chain responsibility Employees’ Provident 8.5 Reliance on social auditing Fund (EPF)? 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 43 research

Executive summary

1. Introduction 1.1. Why this research? 2. Deception 1.2. Reading guide 2. Methodology 2.1. Selection of mills Deception 2.2. Field research 2.3. Supply chain research ‘Deception relates to the failure to deliver what has been promised to the worker, either ver- 2.4. Company review bally or in writing. Victims of forced labour are often recruited with promises of decent, well- 2.5. Methodological challenges paid jobs. But once they begin working, the promised conditions of work do not materialize, 3. Context and workers find themselves trapped in abusive conditions without the ability to escape. In 3.1. India’s cotton, textile, and these cases, workers have not given free and informed consent. Had they known the reality, garment industry 3.2. The Tamil Nadu textile and they would never have accepted the job offer. garment industry 3.3. Trade unions Deceptive recruitment practices can include false promises regarding working conditions and 3.4. Social exclusion based on wages, but also regarding the type of work, housing and living conditions, acquisition of regular caste and tribal background 3.5. Caste discrimination in garment migration status, job location or the identity of the employer. Children may also be recruited factories and spinning mills in through false promises, made to them or their parents, concerning school attendance or the Tamil Nadu frequency of visits by or to their parents.’ 4. Normative framework - fun- ILO forced labour indicators104 damental labour standards 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards 4.3. Forced labour indicators Recruitment 4.4. Legislative developments and existing supply chain legislation Spinning mill owners use a network of agents to recruit workers. These agents are paid a certain amount for every 4.5. Labour laws in India and worker they bring in. Tamil Nadu 4.6. Caste-based discrimination legislation in India As well as using agents, there are informal ways for spinning mill owners to find labourers. They may, for example, 5. Evidence of existence and ask their current workers to find new recruits – perhaps friends, relatives or acquaintances in their home village - in risk of forced labour return for some form of preferential treatment. Tempted by this promise, workers may agree to taking on the role 5.1. Characteristics of the 29 of broker. Members of a worker’s family might also be used by employers. Fathers whose daughters are working investigated mills at the mill, for example, might be asked to help recruit other villagers in return for a small amount of money. It is 5.2. Respondents’ profiles 105 5.3. Risks and evidence of forced difficult to say for sure how widespread these practices are . labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings d W a 8. Conclusions: Beyond 29 mills y e – a sector risk analysis a h 8.1 Incidence and risks of forced n a d v labour e

8.2 Lack of publicly available data and n t i o supply chain transparency g h w 8.3 Flawed business model t or , . k sly 8.4 Restricted interpretation of continuou supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 44 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide WHO GAVE YOU INFORMATION 2. Methodology ABOUT THIS JOB? 2.1. Selection of mills 2.2. Field research 2.3. Supply chain research 2.4. Company review 2.5. Methodological challenges 426 271 3. Context 3.1. India’s cotton, textile, and garment industry Regional Skill Training Centre 18 3.2. The Tamil Nadu textile and garment industry Advertisement 5 3.3. Trade unions Company 7 3.4. Social exclusion based on No answer 1 caste and tribal background

3.5. Caste discrimination in garment Family, friends, factories and spinning mills in Some workers indicated to have received information Tamil Nadu Agent / broker acquaintances from multiple sources 4. Normative framework - fun- damental labour standards 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards Conversation with a recruiter 4.3. Forced labour indicators The research team spoke with a male recruiter. Over the past 15 years, he has recruited more than 4.4. Legislative developments and 100 workers for one particular spinning mill. When he started, he received INR 500 (€ 5.76) for every existing supply chain legislation worker he recruited. Now, he receives INR 3,000 (€ 34.54) per recruited worker. He receives this pay- 4.5. Labour laws in India and Tamil Nadu ment in two instalments: the first part is paid at the start of the worker’s employment and the remain- 4.6. Caste-based discrimination der is paid when the recruited worker has completed three months of employment. For the recruiter, legislation in India these payments are on top of the income he earns from his regular job. The income from recruiting 5. Evidence of existence and workers is not sufficient to cover his needs. risk of forced labour 5.1. Characteristics of the 29 investigated mills This recruiter only recruits girls. He never speaks directly with the girls, but instead discusses the job 5.2. Respondents’ profiles offer with their parents. At first, he recruited workers from his own home village but then he began 5.3. Risks and evidence of forced looking for workers in other villages. The recruiter explained that he targets families facing economic labour – using the 11 ILO indicators hardship and in need of extra income, as he has more chance of successfully recruiting new workers 6. Impact of COVID-19 from such families. 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research Nearly two-thirds of the interviewed workers (426) found their job in the spinning mill through an agent or recruit- 7.1 Supply chain links er. More than one-third (271) found the job through family, friends, or acquaintances. A small group of 18 female 7.2 Links established based on workers found the work through a Regional Skills Training Centre. Five workers had responded to an advertise- corporate disclosure ment, and seven workers had got information about the job opportunity from the spinning mill itself. A combina- 7.3 Links with vertically integra- ted manufacturers tion of factors is also possible. 7.4 Company responses with regard to the draft research Around half of the interviewed workers said the information they had been given about the working and living findings conditions had turned out not to be true. Some spoke in strong terms about the lies they had been told. 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour ‘Now that I am here, we have to work more but the wage we get is less. We have to 8.2 Lack of publicly available data and work continuously, day and night. Leaving home has led me to feel very frustrated.’ supply chain transparency 8.3 Flawed business model Female worker (age 34), mill 21. 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 45 research

Executive summary

1. Introduction 1.1. Why this research? ‘My parents passed away. I have three sisters. While joining here they promised me 1.2. Reading guide 2. Methodology to give INR 60,000 ( € 690.75 ) per year and that food and accommodation would 2.1. Selection of mills be free. But they did not give like that. They give only INR 40,000 ( € 460.50 ) after 2.2. Field research three years. Now my daily wage is INR 300 ( € 3.45 ). But they deduct for food and 2.3. Supply chain research 2.4. Company review accommodation from that too.’ 2.5. Methodological challenges Female worker (13 at the start of her employment, now 18), mill 12. 3. Context 3.1. India’s cotton, textile, and garment industry ‘We are from a poor family, so we all are working here in this mill. My children are 3.2. The Tamil Nadu textile and not going to school because I am getting very low salary. They promised me INR garment industry 9,000 ( € 103.61 )per month but after all the deductions only INR 3,000( € 34.54 ) 3.3. Trade unions 3.4. Social exclusion based on is left.’ caste and tribal background Female worker (age 40), from Bihar, mill 12. 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu ‘The management lied to us through the agent and made us work here. We have to 4. Normative framework - fun- do more work and we get less salary.’ damental labour standards Male worker (age 24), mill 3. 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards ‘Since we do not know any other job, for us this is the best job. We were promised 4.3. Forced labour indicators free food and accommodation. The agent showed us pictures of the place which 4.4. Legislative developments and existing supply chain legislation looked good. We were told that we would live and work in a safe place and that 4.5. Labour laws in India and there would be no heat. But the situation is very different. Most information the Tamil Nadu agent gave us was incorrect. We signed up four years ago but still we have not 4.6. Caste-based discrimination legislation in India received an appointment letter. We were assured by the agent that we would earn 5. Evidence of existence and INR 9,000 ( € 103.61 ) per month but most of us are receiving less. Also we have to risk of forced labour 5.1. Characteristics of the 29 pay some fee to the agent.’ investigated mills FGD, mill 16. 5.2. Respondents’ profiles 5.3. Risks and evidence of forced When the recruitment promises are compared to the reality of the worker’s lives, the discrepancies are glaring, labour – using the 11 ILO indicators especially with regard to working hours and wages. In some cases, the respondents were very aware of the differ- 6. Impact of COVID-19 ences between the promises and the reality but, in other cases, respondents did not seem to realise that they were 6.1 Loss of income earning less than they were promised. 6.2 Emergency relief 6.3 After lockdown Below is a list of the main discrepancies between what was promised and what was delivered. 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on FALSE PROMISES corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis annual leave 8.1 Incidence and risks of forced working hours wage labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 46 research

Executive summary

1. Introduction 1.1. Why this research? Working hours 1.2. Reading guide 2. Methodology Almost all the workers said that when they were recruited, they were told they would be working eight hours a 2.1. Selection of mills day but that in reality, their shifts are between 10 and 12 hours long. Some workers even occasionally work dou- 2.2. Field research 2.3. Supply chain research ble shifts (16 hours), and sometimes (though rarely) they have to work round the clock (read more about working 2.4. Company review hours in paragraph 5.3.11). This discrepancy between promised and real working hours was found in all but five 2.5. Methodological challenges spinning mills. 3. Context 3.1. India’s cotton, textile, and In two mills (mills 29 and 30), most of the workers had known beforehand that they would be working 12 hours a garment industry 3.2. The Tamil Nadu textile and day. In two mills (mills 26 and 27), workers actually worked eight hours a day. garment industry 3.3. Trade unions 3.4. Social exclusion based on Wages caste and tribal background In a considerable number of spinning mills, monthly wages turned out to be lower than what was promised. In 3.5. Caste discrimination in garment factories and spinning mills in one mill (mill 20), none of the interviewed workers had received what they had been told they would be earning. Tamil Nadu Instead of INR 10,000 (€ 115.13), they earned between INR 8,000 - 9,000 (€ 92.10 – 103.61). 4. Normative framework - fun- damental labour standards At mill 5, workers were promised wages between INR 9,000 - 10,000 (€ 103.61 - 115.13) but only nine actually 4.1. International guidelines for responsible business conduct received that amount. The remaining 16 workers who were interviewed, received considerably less, with 14 earn- 4.2. Fundamental labour standards ing INR 7,500 (€ 113.44) or less. 4.3. Forced labour indicators 4.4. Legislative developments and At mill 11, only three workers received the promised wage; the remaining workers earned considerably less than existing supply chain legislation 4.5. Labour laws in India and what they had expected. Promised wages had ranged between INR 9,000 - 10,000 (€ 103.61 - 115.13), but actual Tamil Nadu wages were INR 5,500 - 9,000 (€ 63.32 - 103.61), INR 1,000 - 3,500 (€ 11.51 – 40.29) less than promised. 4.6. Caste-based discrimination legislation in India Likewise, at mill 12, only three workers received the salary they had been promised. Six workers received INR 5. Evidence of existence and 8,700 (€ 100.16) instead of the INR 9,000 (€ 103.61) they were made to believe they would earn. The other 16 risk of forced labour 5.1. Characteristics of the 29 workers interviewed at this mill received wages much lower than promised (ranging between INR 5,000 – 8,000 investigated mills (€ 57.56 - 92.10) as opposed to the INR 9,000 (€ 103.61) promised). 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO Wage deductions indicators 6. Impact of COVID-19 One of the reasons why the actual wages are lower than those promised, is because part of the wage is withheld 6.1 Loss of income by the employer. Of the 725 workers interviewed, 493 said that money had been deducted from their salary. 81 6.2 Emergency relief workers did not know whether or not deductions had been made. It was difficult to get a full picture of why deduc- 6.3 After lockdown tions are made and how much is deducted. 116 workers said they knew money was deducted from their salary but 6.4 Safety measures inadequate 7. Responses to the research did not know why; all they knew was the amount of their net wage. Because they have not signed contracts and/ 7.1 Supply chain links or do not receive payslips, workers lack basic information about benefits, deductions, bonuses, etc. Workers said 7.2 Links established based on that money had been deducted for food (‘canteen’) or accommodation. corporate disclosure 7.3 Links with vertically integra- Sometimes during recruitment, workers had been told that food and accommodation would be free, only to find ted manufacturers 7.4 Company responses with that, in reality, they often had to pay for it. This came up In all four focus group discussions. regard to the draft research findings ‘We were promised three free meals a day and free accommodation. We were also 8. Conclusions: Beyond 29 mills – a sector risk analysis told that we could continue our study if we wanted to; that the work is under a 8.1 Incidence and risks of forced safe roof; that there would be no heat, and more. But the reality is different after labour reaching here. Most of the information was incorrect.’ 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model ‘They are giving us tea and a bun, but we have to pay for it. They deduct money for 8.4 Restricted interpretation of food and accommodation from our salary. Only the balance is paid to us.’ supply chain responsibility 8.5 Reliance on social auditing FGD, mill 4. 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 47 research

Executive summary

1. Introduction 1.1. Why this research? Participants in the focus group discussions spoke of money being deducted from wages to pay for food, accommo- 1.2. Reading guide dation, cleaning charges, electricity, leave, etc. but they had no idea how much was being deducted for each cost. 2. Methodology 2.1. Selection of mills Other deductions are taken as an ‘end-of-term’ or ‘end-of-year’ lump sum. 105 workers, working for six different 2.2. Field research spinning mills, said that money was deducted from their salary for a so-called ‘end-of-term lump sum’. These work- 2.3. Supply chain research ers did not receive any written confirmation of this arrangement, and had no contract or monthly payslips. The 2.4. Company review workers had the impression that the sum taken was going towards a handsome amount which would be paid out 2.5. Methodological challenges to them at the end of their employment but, with no documentation, they were not in a strong position to claim 3. Context 3.1. India’s cotton, textile, and any such sum, were it to exist. These ‘end-of-term’ deductions were reported by workers of mills 15, 16, 17, 20, garment industry 24 and 27. 3.2. The Tamil Nadu textile and garment industry 3.3. Trade unions Annual leave 3.4. Social exclusion based on caste and tribal background During recruitment, workers are made promises about annual leave. Leave is important for workers, many of who 3.5. Caste discrimination in garment come from far-off places and want to know that they will be able to visit their families. Many workers are forced factories and spinning mills in to leave their children behind in their home villages and being separated from them, and their families, is a cause Tamil Nadu of sadness. Consequently, they desperately want to make use of the rare occasions available to them, to see their 4. Normative framework - fun- damental labour standards families. 4.1. International guidelines for In some spinning mills, however, workers are denied permission to take the annual leave they are entitled to or, if responsible business conduct they do take leave, it comes with a penalty (such as the loss of the job entirely, delayed payment of wages, being 4.2. Fundamental labour standards sent away temporarily, loss of working hours). In these situations, many workers are forced to take unpaid leave. 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation ‘For the sake of my family I am forced to do this work. I do not know any other work. 4.5. Labour laws in India and Here we are working as slaves. I have been losing my life here. I have to bear all the Tamil Nadu 4.6. Caste-based discrimination torture. They will not allow us to go home.’ legislation in India Female worker (age 23), mill 16. 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 ‘To go home or to take rest we have to take the permission from the supervisor, often investigated mills he does not give permission easily. If we ask continuously, they will ask us to leave 5.2. Respondents’ profiles the job, or will ask us to stand outside the work premises.’ 5.3. Risks and evidence of forced labour – using the 11 ILO FGD, mill 16. indicators 6. Impact of COVID-19 In all but two spinning mills (mills 2 and 26), workers were told that they would be entitled to a specific number 6.1 Loss of income of days of annual leave. In reality, however, in many cases they were not allowed to take all those days. At mill 6.2 Emergency relief 6.3 After lockdown 12, for example, workers were promised 35 days of holiday a year, but none of the workers ever got to take that 6.4 Safety measures inadequate much leave: five workers got 12 days leave, seven workers got 15 days leave, six workers got 16 days leave, three 7. Responses to the research workers got 18 days, and three workers got 20 days106. 7.1 Supply chain links Likewise, at mill 13, there is a big gap between the promised amount of annual leave and what workers take in 7.2 Links established based on practice. Workers are entitled to 30 leave days a year. 20 workers, however, reported that they only have 18 days corporate disclosure 7.3 Links with vertically integra- leave, and another five workers had to make do with 12 days a year. ted manufacturers Moreover, annual leave is often unpaid. 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills Positive examples. All the workers at mill 26 said that their employer kept his promises regarding – a sector risk analysis working hours, wages, and annual leave. This was confirmed by a closer analysis of the working con- 8.1 Incidence and risks of forced ditions at the mill. labour 8.2 Lack of publicly available data and supply chain transparency Worrying practices. 8.3 Flawed business model At spinning mills 12, 13 and 20, workers reported that wages, working hours, and the amount of an- 8.4 Restricted interpretation of nual leave they receive, are very different to what was promised during recruitment. supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 48 research

Executive summary

1. Introduction 1.1. Why this research? 3. Restriction of movement 1.2. Reading guide 2. Methodology 2.1. Selection of mills 2.2. Field research 2.3. Supply chain research 2.4. Company review 2.5. Methodological challenges 3. Context 3.1. India’s cotton, textile, and garment industry 3.2. The Tamil Nadu textile and garment industry 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu 4. Normative framework - fun- damental labour standards 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO Restriction of movement indicators ‘If workers are not free to enter and exit the work premises, subject to certain restrictions 6. Impact of COVID-19 which are considered reasonable, this represents a strong indicator of forced labour. 6.1 Loss of income Forced labourers may have their movements controlled inside the workplace, through the use 6.2 Emergency relief of surveillance cameras or guards, and outside the workplace by agents of their employer who 6.3 After lockdown 6.4 Safety measures inadequate accompany them when they leave the site.’ 107 7. Responses to the research ILO forced labour indicators 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- The vast majority of workers – 625 out of the 725 – lived in hostels. In most cases (481), these hostels were lo- ted manufacturers cated on the mills’ premises. Workers who reside in such hostels on factory grounds are often severely limited in 7.4 Company responses with their freedom of movement. regard to the draft research findings Table 6: Current domicile 8. Conclusions: Beyond 29 mills – a sector risk analysis Where do you currently live? Respondents 8.1 Incidence and risks of forced Hostel at mill premises 481 labour 8.2 Lack of publicly available data and Hostel outside mill premises 144 supply chain transparency Other 100 8.3 Flawed business model Total 725 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing Hostels, or dormitories for female workers, are usually located within the mills’ premises. There are often separate 8.6 Frontrunners bear the brunt dormitories for Tamil workers, and for migrant workers from other states. Hostels for male workers are sometimes 8.7 Leverage located outside of the mills’ premises, meaning male hostel workers enjoy more freedom of movement than their 9. Recommendations female co-workers. 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 49 research

Executive summary

1. Introduction 1.1. Why this research? A worker does not usually have a free choice as to whether he or she stays in a mill-run hostel. Some workers said 1.2. Reading guide they would rather stay elsewhere but cannot afford another place. Respondents also said that the only housing 2. Methodology 2.1. Selection of mills available in the vicinity of the mill is mill-run hostels. Spinning mills are often located in remote places, far from 2.2. Field research urban centres or residential areas. Public transport between these places is often far from optimal, so commuting 2.3. Supply chain research is not an option. Other workers said they stayed in mill hostels because management had ordered them to; workers 2.4. Company review from 15 of the 29 investigated spinning mills said they stayed in a hostel because management explicitly required 2.5. Methodological challenges it of them (mills: 1, 3, 4 5, 11, 12, 13, 14, 16 21, 22 27, 28, 29, and 30). 3. Context 3.1. India’s cotton, textile, and garment industry During focus group discussions with workers from four mills (mills 4, 11, 16 and 22), workers were asked whether, 3.2. The Tamil Nadu textile and when, and how often, they leave the mills’ premises, and about their contacts with the outside world. Workers garment industry said they were not free to leave the mill’s premises on their own. Their only opportunity to pass through the com- 3.3. Trade unions 3.4. Social exclusion based on pounds’ gates was for (monthly) trips to the market and even then, such trips could only be made in the presence caste and tribal background of a hostel warden. 3.5. Caste discrimination in garment factories and spinning mills in Respondents also said they had very few opportunities to speak with their families. Only some of the workers had Tamil Nadu mobile phones. There is a payable landline phone in the hostel which workers can use, but only in the presence of 4. Normative framework - fun- damental labour standards the hostel warden, or supervisor. 4.1. International guidelines for responsible business conduct Workers commented that hostel wardens monitor their every movement, and that they feel very uncomfortable 4.2. Fundamental labour standards being watched all the time. The wardens are described as very strict. When they are not working, the workers must 4.3. Forced labour indicators 4.4. Legislative developments and stay in their dormitory. In some mills, there are restrictions in place that prevent workers from freely mixing with existing supply chain legislation one another. 4.5. Labour laws in India and Tamil Nadu ‘Frankly saying I do not like to work here, there is no freedom here to go for shopping 4.6. Caste-based discrimination legislation in India or to go home. There is no proper food. We cannot share our feelings to anyone. They 5. Evidence of existence and are not treating us as human beings.’ risk of forced labour 5.1. Characteristics of the 29 Female worker (age 21), mill 17. investigated mills 5.2. Respondents’ profiles ‘I come from a poor family, but here we are also suffering. We have no freedom.’ 5.3. Risks and evidence of forced labour – using the 11 ILO Male worker (age 21), from Odisha, mill 2. indicators 6. Impact of COVID-19 ‘When I joined here, there were so many rules and regulations. Even for using the 6.1 Loss of income bathroom, they will only give five minutes. Now, there is a little liberation but still we 6.2 Emergency relief 6.3 After lockdown are not allowed to go outside of the mill premises.’ 6.4 Safety measures inadequate Female worker (age 47), mill 13. 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on ‘The workers are often not allowed to gather together. Work always - we have to do. corporate disclosure Nobody can go out, and no outsiders can enter the mill.’ 7.3 Links with vertically integra- ted manufacturers FGD, mill 4. 7.4 Company responses with regard to the draft research ‘Since I do not know other work I joined here. Physically and psychologically I am findings suffering a lot. The work is heavy and they are making us to do more work and they 8. Conclusions: Beyond 29 mills – a sector risk analysis are giving us very low salary. There is not enough food. I feel like living in the jail. I do 8.1 Incidence and risks of forced not know when I am getting relief from this.’ labour 8.2 Lack of publicly available data and Female worker (age 20), mill 16. supply chain transparency 8.3 Flawed business model Those workers who do not reside on mill grounds, experience more freedom. One female worker, working at 8.4 Restricted interpretation of mill 24, explained that, because she was staying in a private hostel outside the mill’s compound, she could vis- supply chain responsibility it her family on her weekly day off, but that this would not be possible if she were living in a mill-run hostel. 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 50 research

Executive summary

1. Introduction 1.1. Why this research? 4. Isolation 1.2. Reading guide 2. Methodology 2.1. Selection of mills 2.2. Field research 2.3. Supply chain research Isolation 2.4. Company review ‘Victims of forced labour are often isolated in remote locations, denied contact with the -out 2.5. Methodological challenges side world. 3. Context 3.1. India’s cotton, textile, and Workers may not know where they are, the worksite may be far from habitation and there may be garment industry 3.2. The Tamil Nadu textile and no means of transportation available. But equally, workers may be isolated even within populated garment industry areas, by being kept behind closed doors or having their mobile phones or other means of commu- 3.3. Trade unions nication confiscated, to prevent them from having contact with their families and seeking help.’ 3.4. Social exclusion based on ILO forced labour indicators108 caste and tribal background 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu Most of the workers in the Tamil Nadu spinning mills live in a state of quite strict social isolation. They come from 4. Normative framework - fun- faraway places, such as the outlying districts of Tamil Nadu, or are from other states in India. Interstate and in- damental labour standards trastate migrant workers often live in mill-run hostels where interaction with the outside world, even with family 4.1. International guidelines for responsible business conduct or friends, is limited. Workers who live on the factory premises are not allowed to leave, and outsiders are not 4.2. Fundamental labour standards permitted entry. 4.3. Forced labour indicators Hostels have phones, but workers can only use the phone in the presence of the hostel warden or supervisor, 4.4. Legislative developments and which clearly hinders personal or intimate conversations. Some workers have mobile phones and can use these to existing supply chain legislation 4.5. Labour laws in India and contact their families. Tamil Nadu 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 51 research

Executive summary

1. Introduction 1.1. Why this research? Workers also have to cope with obvious language barriers. Tamil is spoken in Tamil Nadu, but many interstate 1.2. Reading guide workers do not speak Tamil. Migrants from the northern states speak Hindi (the most important language in the 2. Methodology 2.1. Selection of mills northern part of India) but many people in Tamil Nadu the migrant workers might come into contact with – such 2.2. Field research as employers, supervisors, co-workers, hostel wardens, market sellers, etc. - do not speak Hindi. With so many 2.3. Supply chain research workers coming from so many different places and with so many different mother tongues, there is no common 2.4. Company review language to connect workers. Each spinning mill is, in effect, a small Tower of Babel. The Tamil-speaking workers 2.5. Methodological challenges share dormitories, but other workers are placed in mixed-language dormitories. This segregation, along with the 3. Context 3.1. India’s cotton, textile, and language barriers, prevents solidarity developing among the workers. garment industry 3.2. The Tamil Nadu textile and The isolation of hostel workers is exacerbated by the location of the spinning mills, which are often in remote garment industry places, far away from urban or residential areas, and with few facilities nearby. 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background We found that many workers were also not aware of the existence of organisations that could offer support or 3.5. Caste discrimination in garment advice, such as trade unions or labour rights organisations, which only increased their sense of isolation. factories and spinning mills in Tamil Nadu In practice, there are hardly any trade unions present in the spinning mills we looked at. Most of these mills also 4. Normative framework - fun- damental labour standards did not have a Works Committee (WC) or an Internal Complaints Committee (ICC), despite the legal obligations for 4.1. International guidelines for employers to have such committees in place. Alternatively, if such committees did exist, workers were not aware responsible business conduct of them (for more information on this topic see paragraph 5.3.10). 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and Some respondents said that workers had been explicity told by management not to speak with outsiders about existing supply chain legislation the conditions in the mill and hostel. Moreover, in three of the four mills where focus group discussions were held, 4.5. Labour laws in India and respondents said that when government officials visited (possibly hostel inspectors), workers were instructed by Tamil Nadu supervisors on how to answer questions. 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and Workers also worked very long days, with little time left for leisure or personal development. And, in the rare risk of forced labour moments that workers did have to themselves, they found there was not much to do. This is especially true for 5.1. Characteristics of the 29 workers who live on the mills’ compounds. Some respondents said that if they wanted to watch television in the investigated mills hostel, they had to pay. They cannot afford to pay, so they don’t watch. 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO This all adds up to a situation that leaves workers lonely and unsupported in their hardships. indicators 6. Impact of COVID-19 6.1 Loss of income ‘After joining this mill, I am totally depressed due to heavy work. My body gets 6.2 Emergency relief weaker day by day. The hardship of the mill is not to be told outside.’ 6.3 After lockdown Male worker (age 23), mill 17. 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links ‘We definitely want to complain and seriously think about it, but due to fear we do 7.2 Links established based on not.’ corporate disclosure FGD, mill 11. 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with ‘I am not happy, I am feeling trapped. This makes me feel angry. I feel like running regard to the draft research away from this place but I can not leave due to my family’s situation.’ findings 8. Conclusions: Beyond 29 mills Female worker (age 24), mill 15. – a sector risk analysis 8.1 Incidence and risks of forced ‘Most of my life I am spending with the machines. There is absolutely no outside labour 8.2 Lack of publicly available data and exposure. I am working here to take care of my family. Other than taking care of my supply chain transparency family I feel no happiness.’ 8.3 Flawed business model Male worker (age 32), mill 20. 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing ‘Here there is no safety. The management told us not to share outside about 8.6 Frontrunners bear the brunt what is happening here.’ 8.7 Leverage 9. Recommendations Male worker (age 31), from Bihar, mill 13. 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 52 research

Executive summary

1. Introduction 1.1. Why this research? 5. Physical and sexual violence 1.2. Reading guide 2. Methodology 2.1. Selection of mills 2.2. Field research 2.3. Supply chain research Physical and sexual violence 2.4. Company review ‘Forced labourers, their family members and close associates may be subjected to actual phys- 2.5. Methodological challenges ical or sexual violence. Violence can include forcing workers to take drugs or alcohol so as to 3. Context have greater control over them. Violence can also be used to force a worker to undertake tasks 3.1. India’s cotton, textile, and that were not part of the initial agreement, such as to have sex with the employer or a family garment industry 3.2. The Tamil Nadu textile and member or, less extreme, to undertake obligatory domestic work in addition to their ‘normal’ garment industry tasks. Physical abduction or kidnapping is an extreme form of violence which can be used to 3.3. Trade unions take a person captive and then force them to work. As violence is not acceptable as a discipli- 3.4. Social exclusion based on nary measure under any circumstances, it is a very strong indicator of forced labour.’ caste and tribal background 109 3.5. Caste discrimination in garment ILO forced labour indicators factories and spinning mills in Tamil Nadu 4. Normative framework - fun- damental labour standards 4.1. International guidelines for Although it is a sensitive topic, female workers did bring up the issue of sexual and physical abuse, during both the responsible business conduct individual interviews and the focus group discussions. They provided anecdotal evidence of abuses that align with 4.2. Fundamental labour standards behaviours categorised by the ILO as physical and sexual abuse. 4.3. Forced labour indicators 4.4. Legislative developments and Women workers at a number of mills reported physical and sexual abuse. Mill 16 specifically stood out as an ex- existing supply chain legislation 4.5. Labour laws in India and treme, negative example. The majority of the interviewed female workers who worked at mill 16 used words such Tamil Nadu as ‘slavery’ and ‘torture’ to describe the general working environment. 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 53 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide ‘Here, we are working like slaves. Here, I am losing my life and must bear all kinds of 2. Methodology 2.1. Selection of mills torture.’ 2.2. Field research Female worker (age 23) 2.3. Supply chain research 2.4. Company review 2.5. Methodological challenges A focus group discussion which included eight female workers from mill 16, reaffirmed their implicit comments 3. Context about physical abuse. Women reported feeling uneasy inside the mill’s premises. They said that, under the cover 3.1. India’s cotton, textile, and of loud machine noises, their male co-workers and supervisors directed sexualised comments at them. Women garment industry workers also gave examples of how management, mill personnel and co-workers, approached them - pulling hands, 3.2. The Tamil Nadu textile and garment industry inappropriate touching, pinching, expressing inappropriate feelings of love – all without the women’s consent. 3.3. Trade unions This abusive atmosphere extended to the hostels, where women workers also reported feeling ill at ease. Because 3.4. Social exclusion based on of this tension and atmosphere, women workers worked and moved around in groups, trying to stay safe by sur- caste and tribal background rounding themselves by female colleagues. 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu The focus group discussions on these topics, conducted with workers of mills 11 and 22, exposed similar patterns 4. Normative framework - fun- of sexual harassment and physical abuse. damental labour standards 4.1. International guidelines for responsible business conduct ‘We do not feel at ease. There is a threatening atmosphere. There are several 4.2. Fundamental labour standards disturbances; sexual harassment is one of them. We do not feel safe when we are 4.3. Forced labour indicators alone in the workplace or in the hostel or when we are using the toilet. We are not 4.4. Legislative developments and existing supply chain legislation safe.’ 4.5. Labour laws in India and FGD, mill 22. Tamil Nadu 4.6. Caste-based discrimination legislation in India ‘There is no safety here. At night, when we are sleeping, anyone can come in.’ 5. Evidence of existence and Female worke, mill 11. risk of forced labour 5.1. Characteristics of the 29 Similar comments were made by workers at several other investigated mills: investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced ‘There is no safety here like at home. During the night I am afraid. Being here is a labour – using the 11 ILO struggle.’ indicators 6. Impact of COVID-19 Female worker (age 22), mill 17. 6.1 Loss of income 6.2 Emergency relief ‘I am frightened here. My co-workers are being scolded often and sometime beaten 6.3 After lockdown 6.4 Safety measures inadequate up also. Even I have been scolded and was getting beaten. If I want go off from here, 7. Responses to the research due to family situation I am forced to be here. My father will scold me if leave the 7.1 Supply chain links work.’ 7.2 Links established based on corporate disclosure Female worker (age 23), mill 13. 7.3 Links with vertically integra- ted manufacturers At the mill 4 male-only focus group discussion, workers said they had not experienced any form of sexual or phys- 7.4 Company responses with ical violence. regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 54 research

Executive summary

1. Introduction 1.1. Why this research? 6. Intimidation and threats 1.2. Reading guide 2. Methodology 2.1. Selection of mills 2.2. Field research 2.3. Supply chain research Intimidation and threats 2.4. Company review ‘Victims of forced labour may suffer intimidation and threats when they complain about their 2.5. Methodological challenges conditions or wish to quit their jobs. In addition to threats of physical violence, other com- 3. Context mon threats used against workers include denunciation to the immigration authorities, loss of 3.1. India’s cotton, textile, and wages or access to housing or land, sacking of family members, further worsening of working garment industry 3.2. The Tamil Nadu textile and conditions or withdrawal of ‘privileges’ such as the right to leave the workplace. Constantly garment industry insulting and undermining workers also constitutes a form of psychological coercion, designed 3.3. Trade unions to increase their sense of vulnerability. 3.4. Social exclusion based on caste and tribal background 3.5. Caste discrimination in garment The credibility and impact of the threats must be evaluated from the worker’s perspective, factories and spinning mills in taking into account his or her individual beliefs, age, cultural background and social and eco- Tamil Nadu nomic status.’ 4. Normative framework - fun- ILO Forced Labour Indicators110 damental labour standards 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 55 research

Executive summary

1. Introduction 1.1. Why this research? Both in individual interviews, and during the focus group discussions, workers reported experiencing all kinds of 1.2. Reading guide intimidation and threats from managers and supervisors. These threats and warnings do not necessarily always 2. Methodology 2.1. Selection of mills materialise in concrete actions, but the workers described them as unsettling. The threats and intimidation create 2.2. Field research an atmosphere in which workers do not feel at ease or comfortable. Instead, they are literally afraid to speak their 2.3. Supply chain research mind, ask questions, disagree with decisions made for them, let alone protest. This system of intimidation is most 2.4. Company review effective when it comes to making workers work overtime. From respondents’ detailed answers it became clear 2.5. Methodological challenges that workers have internalised the threats and believe that they will be carried out. Workers believe they have no 3. Context 3.1. India’s cotton, textile, and choice but to shut up and walk the line. garment industry 3.2. The Tamil Nadu textile and Scolding is very common in the spinning mills. It seems to happen non-stop. Most of the interviewed workers de- garment industry scribed being scolded and shouted at in all kinds of situations. 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background Common threats made to workers include financial penalties. More than one-third of the interviewed workers 3.5. Caste discrimination in garment (270) believed that if they left their job, they would suffer a financial penalty, though whether this threat could be factories and spinning mills in carried out is unclear. Workers are also unsure about the threat, but they are frightened. This fear was mentioned Tamil Nadu by workers from mills 2, 3, 4, 6, 7, 8, 9, 11, 14, 15, 16, 17, 18, 19, 20, 21, 25, and 27111. 4. Normative framework - fun- damental labour standards 4.1. International guidelines for Other sources, including the south Indian labour rights organisations involved in this research, confirm the workers responsible business conduct experiences. The organisations say that employers threaten to withhold workers’ monthly wages or cancel their 4.2. Fundamental labour standards end-of-term lump sum, if workers quit their jobs112. These threats are, apparently, one of the ways employers try 4.3. Forced labour indicators 4.4. Legislative developments and to reduce labour turnover. existing supply chain legislation 4.5. Labour laws in India and Workers from mills 11, 16 and 22 said that if they complained, disagreed or made any negative comment, they Tamil Nadu risked incurring a penalty, compulsory night shifts or dismissal. This was discussed in detail in the focus groups. 4.6. Caste-based discrimination legislation in India The pressure on textile workers to work overtime is obvious. In 20 of the 29 mills that we investigated workers 5. Evidence of existence and reported this. Workers from mills 1, 2, 3, 5, 7, 8, 9, 11, 12, 13, 14, 15, 16, 17, 19, 20, 21, 22, 24, and 25 described risk of forced labour the potential repercussions of refusing overtime work. The main forms of pressure applied to workers to make 5.1. Characteristics of the 29 them work overtime include: scolding (mentioned 185 times), dismissal (mentioned 74 times), delaying payment of investigated mills wages (mentioned 68 times), and making deducations from wages (mentioned 32 times). 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO Workers also mentioned other types of retaliation that might be used should they dare to refuse overtime, ranging indicators from ‘refusing us to pick overtime hours of our own choice in the future’ (mentioned 25 times, by workers from 6. Impact of COVID-19 mills 1 and 17) and getting ‘a black mark in our record’, to ‘having to complete an extra shift’, ‘beatings and fights’ 6.1 Loss of income 6.2 Emergency relief (mill 11), and ‘blackmailing’. 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research ‘We cannot take leave, if we take leave our wages will be delayed. We have to keep 7.1 Supply chain links 7.2 Links established based on working for the sake of our families and most of us have debts and loans to pay off. corporate disclosure We bear all these for the sake of family, for our children’s education.’ 7.3 Links with vertically integra- ted manufacturers FGD, mill 4. 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 56 research

Executive summary

1. Introduction 1.1. Why this research? 7. Retention of identity documents 1.2. Reading guide 2. Methodology 2.1. Selection of mills 2.2. Field research 2.3. Supply chain research Retention of identity documents 2.4. Company review ‘The retention by the employer of identity documents or other valuable personal possessions is 2.5. Methodological challenges an element of forced labour if workers are unable to access these items on demand and if they 3. Context feel that they cannot leave the job without risking their loss. In many cases, without identity 3.1. India’s cotton, textile, and documents, the worker will not be able to obtain other jobs or access essential services, and garment industry 3.2. The Tamil Nadu textile and may be afraid to ask for help from authorities or NGOs.’ garment industry ILO forced labour indicators113 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu 4. Normative framework - fun- damental labour standards 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and According to the research team, it is not uncommon for management to keep the identity documents of their supply chain transparency 8.3 Flawed business model workers, leaving the workers themselves with only a copy. Participants from the four mills (mills 4, 11, 16 and 22) 8.4 Restricted interpretation of where focus group discussions were held, confirmed that this practice takes place. They told how they had to hand supply chain responsibility over their original identity document to the recruiting agent, and how it is now with mill management. According 8.5 Reliance on social auditing to the research team, management keep hold of workers’ original identity documents to prevent workers leaving. 8.6 Frontrunners bear the brunt 8.7 Leverage Holding on to workers’ original identity documents is another way of tying workers to the mill. 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 57 research

Executive summary

1. Introduction 1.1. Why this research? 8. Withholding of wages 1.2. Reading guide 2. Methodology 2.1. Selection of mills 2.2. Field research 2.3. Supply chain research Withholding of wages 2.4. Company review ‘Workers may be obliged to remain with an abusive employer while waiting for the wages that 2.5. Methodological challenges are owed to them. The fact of irregular or delayed payment of wages does not automatically 3. Context imply a forced labour situation. But when wages are systematically and deliberately withheld 3.1. India’s cotton, textile, and as a means to compel the worker to remain, and deny him or her of the opportunity to change garment industry 3.2. The Tamil Nadu textile and employer, this points to forced labour.’ garment industry ILO forced labour indicators114 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background 3.5. Caste discrimination in garment factories and spinning mills in A total of 105 workers, at six spinning mills, said that money was being deducted from their salary towards an ‘end Tamil Nadu of term lump sum’. In the absence of a contract, appointment letter or any other written confirmation, it is not clear 4. Normative framework - fun- what the end of employment terms are. Meanwhile, the prospect of an end of term lump sum is an effective means damental labour standards of ensuring workers stay at the mill. 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards The mills deducting money from workers’ salaries towards an ‘end of term lump sum’ are 15, 16, 17, 20, 24 and 27. 4.3. Forced labour indicators Because there is no written confirmation of employment terms, workers are at a high risk of not receiving the 4.4. Legislative developments and money they anticipate. existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 58 research

Executive summary

1. Introduction 1.1. Why this research? 9. Debt bondage 1.2. Reading guide 2. Methodology 2.1. Selection of mills 2.2. Field research 2.3. Supply chain research Debt bondage 2.4. Company review ‘Forced labourers are often working in an attempt to pay off an incurred or sometimes even 2.5. Methodological challenges inherited debt. The debt can arise from wage advances or loans to cover recruitment or trans- 3. Context port costs or from daily living or emergency expenses, such as medical costs. Debts can be 3.1. India’s cotton, textile, and compounded as a result of manipulation of accounts, especially when workers are illiterate. garment industry 3.2. The Tamil Nadu textile and Debt bondage may also arise when children are recruited in exchange for a loan given to their garment industry parents or relatives. Employers or recruiters make it difficult for workers to escape from the 3.3. Trade unions debt, by undervaluing the work performed or inflating interest rates or charges for food and 3.4. Social exclusion based on housing. caste and tribal background 3.5. Caste discrimination in garment factories and spinning mills in Debt bondage – or bonded labour – reflects an imbalance in power between the - work Tamil Nadu er-debtor and the employer-creditor. It has the effect of binding the worker to the em- 4. Normative framework - fun- ployer for an unspecified period of time, anything from a single season, to years, oreven damental labour standards 4.1. International guidelines for successive generations. It bears no resemblance to taking a ‘normal’ loan from a bank or responsible business conduct other independent lender, for repayment on mutually agreed and acceptable terms.’ 4.2. Fundamental labour standards ILO forced labour indicators115 4.3. Forced labour indicators 4.4. Legislative developments and

existing supply chain legislation 4.5. Labour laws in India and There are indications that some workers may be in debt bondage. Tamil Nadu At mills 11 and 12, some of the interviewed workers spoke about having to pay a fee to the agent who recruited 4.6. Caste-based discrimination them to work there. When the worker is recruited, the employer, via the recruiting agent, ‘gives’ the worker’s fami- legislation in India ly a sum of money which may be between INR 10,000 and 20,000 (€ 115.13 – 230.25). This money is given on the 5. Evidence of existence and risk of forced labour provision that the amount is then deducted from the worker’s salary at the end of her or his employment term. It 5.1. Characteristics of the 29 transpires, however, that workers must work for years to pay off this debt. investigated mills 5.2. Respondents’ profiles ‘We also have to pay some fee to the agent. At the same time, we too have received some advances from them 5.3. Risks and evidence of forced such as INR 10,000 (€ 115.13), INR 15,000 (€ 172.69), INR 20,000 (€ 230.25).’ labour – using the 11 ILO indicators FGD, mill 22. 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model ‘When they agent recruited us, 8.4 Restricted interpretation of he gave a fee of INR 20,000 supply chain responsibility 8.5 Reliance on social auditing (€ 230.25). We signed an 8.6 Frontrunners bear the brunt agreement that this amount will 8.7 Leverage 9. Recommendations be paid back by us.’ 9.1 Corporate actors FGD, mill 11. 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 59 research

Executive summary

1. Introduction 1.1. Why this research? 10. Abusive working and living conditions 1.2. Reading guide 2. Methodology 2.1. Selection of mills 2.2. Field research 2.3. Supply chain research Abusive working and living conditions 2.4. Company review ‘Forced labour victims are likely to endure living and working conditions that workers would 2.5. Methodological challenges never freely accept. Work may be performed under conditions that are degrading (humiliating 3. Context or dirty) or hazardous (difficult or dangerous without adequate protective gear), and in severe 3.1. India’s cotton, textile, and breach of labour law. Forced labourers may also be subjected to substandard living conditions, garment industry 3.2. The Tamil Nadu textile and made to live in overcrowded and unhealthy conditions without any privacy. garment industry 3.3. Trade unions Extremely bad working and living conditions alone do not prove the existence of forced labour; 3.4. Social exclusion based on unfortunately, people may sometimes ‘voluntarily’ accept bad conditions because of the lack caste and tribal background 3.5. Caste discrimination in garment of any alternative jobs. However, abusive conditions should represent an ‘alert’ to the possible factories and spinning mills in existence of coercion that is preventing the exploited workers from leaving the job.’ Tamil Nadu ILO forced labour indicators116 4. Normative framework - fun- damental labour standards 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards ‘We do not get proper sleep. We always have to work. We often have to work two 4.3. Forced labour indicators shifts and sometimes even three shifts. This makes us feel tired and drowsy. But we 4.4. Legislative developments and cannot take any rest.’ existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu ‘When the food is not good, we work with an empty stomach. We feel nauseous 4.6. Caste-based discrimination and sometimes we need to vomit.’ legislation in India 5. Evidence of existence and FGD, mill 16. risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 60 research

Executive summary

1. Introduction 1.1. Why this research? Low wages 1.2. Reading guide 2. Methodology Minimum wages in India are set at national, state, sector, and skill/occupational level. In Tamil Nadu, however, 2.1. Selection of mills there is no minimum wage set for the textile sector, only a minimum wage for apprentices in textile mills. 2.2. Field research 2.3. Supply chain research 2.4. Company review Apprentices earn a basic daily minimum wage of INR 110 (€ 1.27). Along with a daily dearness allowance of INR 2.5. Methodological challenges 213, this brings the daily minimum wage to INR 322 (€ 3.71). Based on this daily rate, a monthly wage would be 3. Context INR 8,390.20 (€ 96.59) (assuming six working days a week of eight hours in length). Shockingly, there are workers 3.1. India’s cotton, textile, and at all the spinning mills bar one (mill 7), whose daily wage is below the minimum wage for apprentices in textile garment industry 3.2. The Tamil Nadu textile and mills. According to Wage Indicator, a monthly living wage for a family of four (two parents with two children) in garment industry Tamil Nadu in 2020 is INR 23,000 (€ 264.79).117 3.3. Trade unions 3.4. Social exclusion based on Table 7 provides information on the daily wages and monthly take-home wages of workers at the 29 investigated caste and tribal background 3.5. Caste discrimination in garment spinning mills. The monthly wage is the sum of the daily wages, including overtime wages and monthly bonuses (if factories and spinning mills in any), minus deductions and penalties (if any). Tamil Nadu 4. Normative framework - fun- The majority of the interviewed workers (561) said they received a festival bonus; 25 workers118 said they received damental labour standards 4.1. International guidelines for an attendance bonus. responsible business conduct 4.2. Fundamental labour standards 4.3. Forced labour indicators Financial penalties 4.4. Legislative developments and At 18 spinning mills, respondents said financial penalties might be issued for, for example, arriv- existing supply chain legislation 4.5. Labour laws in India and ing late to work, being absent from work, damaged or broken machines, machine parts or tools. Tamil Nadu 4.6. Caste-based discrimination Interviewed workers at the following mills made mention of financial penalties: 1, 2, 3, 4, 5, 6, 7, 9, 11, 12, 13, 15, legislation in India 16, 17, 19, 20, 21, and 26. 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 Table 7: Daily and monthly wages (lowest and highest) as reported by interviewed workers investigated mills 5.2. Respondents’ profiles Mill Lowest daily wage Highest daily wage Lowest monthly Highest monthly 5.3. Risks and evidence of forced (INR) (INR) wage (INR) wage (INR) labour – using the 11 ILO indicators 1. 290 350 6,000 10,000 6. Impact of COVID-19 2. 200 450 6,000 10,000 6.1 Loss of income 6.2 Emergency relief 3. 290 290 8,500 8,500 6.3 After lockdown 4. 210 450 8,000 8,500 6.4 Safety measures inadequate 7. Responses to the research 5. 200 350 6,000 9,000 7.1 Supply chain links 6. 250 320 6,500 9,000 7.2 Links established based on corporate disclosure 7. 330 330 10,000 10,000 7.3 Links with vertically integra- 8. 250 350 No answer No answer ted manufacturers 9. Don’t know/no Don’t know/no answer 9,000 Don’t know 7.4 Company responses with regard to the draft research answer findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 11. 200 300 5,400 9,000 8.1 Incidence and risks of forced 12. 200 310 5,000 9,000 labour 8.2 Lack of publicly available data and 13. 200 320 6,000 9,600 supply chain transparency 14. 200 350 7,500 10,000 8.3 Flawed business model 15. 200 350 6,500 10,000 8.4 Restricted interpretation of supply chain responsibility 16. 210 320 6,000 9,600 8.5 Reliance on social auditing 17. 200 350 6,000 10,000 8.6 Frontrunners bear the brunt 8.7 Leverage 18. 300 320 No answer No answer 9. Recommendations 19. 200 300 7,500 9,000 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 61 research

Executive summary

1. Introduction 1.1. Why this research? 20. 200 300 8,000 9,000 1.2. Reading guide 2. Methodology 21. 200 300 8,000 9,500 2.1. Selection of mills 22. 260 300 8,000 9,000 2.2. Field research 2.3. Supply chain research 23. 330 350 8,000 8,000 2.4. Company review 24. 250 250 7,500 7,500 2.5. Methodological challenges 3. Context 25. 200 200 8,000 8,000 3.1. India’s cotton, textile, and 26. 120 120 3,600 3,600 garment industry 27. 250 250 7,500 7,500 3.2. The Tamil Nadu textile and garment industry 28. 400 470 12,600 18,000 3.3. Trade unions 29. 380 450 10,000 14,000 3.4. Social exclusion based on caste and tribal background 30. 270 380 10,000 10,000 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu 4. Normative framework - fun- Health impacts damental labour standards 4.1. International guidelines for Four focus group discussions involving workers from four spinning mills (4, 11, 16 and 22) focused on aspects of responsible business conduct health and safety. 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and The interviewed workers spoke of ways their work, and work environment, affected their health. existing supply chain legislation 4.5. Labour laws in India and Workers reported feeling fatigued because of long working hours, lack of sleep and lack of proper food. The ma- Tamil Nadu chines are always running, so workers can never take a rest. 4.6. Caste-based discrimination legislation in India Work in spinning mills requires workers to be on their feet the entire time which, the workers said, increased their 5. Evidence of existence and exhaustion. Standing all day also causes leg pains. risk of forced labour 5.1. Characteristics of the 29 Weight loss was also a common concern. This issue arose in all four focus group discussions, as well as in individ- investigated mills 5.2. Respondents’ profiles ual worker interviews. Workers said they had been in good health when they arrived at the mill but had since lost 5.3. Risks and evidence of forced weight, which they attributed to the demanding work and lack of proper food. labour – using the 11 ILO indicators All the participants in the four focus group discussions described the environment in the spinning mills as dusty, 6. Impact of COVID-19 6.1 Loss of income hot, and moist. The air in spinning mills is deliberately kept hot and humid to prevent threads breaking. Workers 6.2 Emergency relief said they are provided with mouth caps but feel they cannot breathe properly. 6.3 After lockdown Headaches and coughs were reported, and some workers also said they were anaemic. They attributed these 6.4 Safety measures inadequate illnesses and conditions to their low food consumption. 7. Responses to the research At the four spinning mills where focus group discussions on occupational health and safety took place, participants 7.1 Supply chain links 7.2 Links established based on said there was no such thing as paid sick leave. If workers took sick leave, they would not be paid for their days corporate disclosure away from work. As a result, workers continued working even when they were ill. 7.3 Links with vertically integra- ted manufacturers Some medication is available in the spinning mills. Workers said that painkillers are available if workers need them, 7.4 Company responses with regard to the draft research but those in the focus group discussions said that, most of the time, they were charged for medication. Conse- findings quently, workers do not ask for painkillers even when they feel unwell. If workers are seriously ill, supervisors will 8. Conclusions: Beyond 29 mills allow them to go home but this leave would be unpaid. – a sector risk analysis 8.1 Incidence and risks of forced labour Injuries 8.2 Lack of publicly available data and supply chain transparency According to workers taking part in the focus group discussions, common injuries include cuts to fingers, hands 8.3 Flawed business model and wrists caused by handling sharp cotton threads. At mill 16, participants spoke of a co-worker who had lost a 8.4 Restricted interpretation of finger due to an accident in the spinning mill. supply chain responsibility Male workers also talked about injuries caused by having to handle heavy loads. 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt Workers said there were some first-aid materials in the factory, for minor injuries. Workers with serious injuries 8.7 Leverage were sent to a clinic for treatment. 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 62 research

Executive summary

1. Introduction 1.1. Why this research? Restrictions in toilet use 1.2. Reading guide 2. Methodology Interviewed workers described restrictions imposed by management on the use of the toilet. During working 2.1. Selection of mills hours, workers had to ask the supervisor for permission to use the toilet. Participants in the focus group discus- 2.2. Field research 2.3. Supply chain research sions said they were only allowed to use the toilet two or three times a day, and each visit had a time limit. These 2.4. Company review restrictions are particularly problematic for women workers, especially when they are menstruating. 2.5. Methodological challenges 3. Context ‘There are rules like; during shift hours we are allowed to use toilets only twice or 3.1. India’s cotton, textile, and garment industry thrice, with time limit of 5 to 10 minutes only. Even during our period time. The 3.2. The Tamil Nadu textile and supervisor has to permit us to go for toilets.’ garment industry FGD, mill 22. 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background ‘The supervisor has to permit us to go for toilets. If used more times, we get scolding 3.5. Caste discrimination in garment or will be sent out.’ factories and spinning mills in Tamil Nadu FGD, mill 11. 4. Normative framework - fun- damental labour standards 4.1. International guidelines for responsible business conduct Absence of workplace trade unions 4.2. Fundamental labour standards Interviewed workers had very limited awareness of labour rights and trade unions. 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation We asked all 725 respondents if there was a trade union at the mill where they worked. Only 93 workers said 4.5. Labour laws in India and there was a trade union at their workplace. Respondents from two mills (mills 14 and 22) answered the question Tamil Nadu by speaking about the “ICC committee”, appearing to mistake the Internal Complaints Committee (ICC) for a trade 4.6. Caste-based discrimination union (indicating that workers do not know the difference between a trade union and an ICC). legislation in India 5. Evidence of existence and risk of forced labour Workers at five spinning mills did actually name the union active in their mill (AITUC and, once, CITU). Only three 5.1. Characteristics of the 29 of the 725 respondents said they were a member of a trade union (one worker from mill 16, and two workers from investigated mills mill 27). 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO Table 8: Spinning mills where (a number of) workers said there was a union: indicators 6. Impact of COVID-19 Mill no. No of workers indicating there was a trade union in their mill Name of the union 6.1 Loss of income 6.2 Emergency relief 11 10 AITUC 6.3 After lockdown 6.4 Safety measures inadequate 12 4 AITUC, CITU 7. Responses to the research 13 5 AITUC 7.1 Supply chain links 14 12 ICC Committee 7.2 Links established based on corporate disclosure 16 8 AITUC 7.3 Links with vertically integra- 17 4 AITUC ted manufacturers 7.4 Company responses with 20 1 Don’t know regard to the draft research 21 1 Don’t know findings 22 6 ICC Committee 8. Conclusions: Beyond 29 mills – a sector risk analysis 24 20 Don’t know 8.1 Incidence and risks of forced 27 22 Don’t know labour 8.2 Lack of publicly available data and Total 93 supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility ‘There is no labour union active here so workers’ problems are not rectified.’ 8.5 Reliance on social auditing Male worker (age 21), from Odisha, mill 2. 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 63 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide ‘Here there are many problems. They are not giving us bonus. They are not giving us 2. Methodology 2.1. Selection of mills proper food. Everything is very costly but our salary is very low. They are not allowing 2.2. Field research labour unions here.’ 2.3. Supply chain research 2.4. Company review Male worker (age 30), from Odisha, mill 2. 2.5. Methodological challenges 3. Context Various workers from mill 6, reported that management did not allow them to join a labour union. Among the pri- 3.1. India’s cotton, textile, and orities for improvement, workers from this mill said they wanted to have an active labour union at the workplace. garment industry 3.2. The Tamil Nadu textile and garment industry Non-functioning or non-existent complaint committees 3.3. Trade unions 3.4. Social exclusion based on Internal Complaints Committees (ICC) had been established at eight spinning mills (mills 11, 12, 13, 14, 16, 19, 20 caste and tribal background and 21). At four of these mills, workers said they knew who the chair of the committee was and, at two of the mills, 3.5. Caste discrimination in garment workers said the committees had dealt with complaints. factories and spinning mills in Tamil Nadu 4. Normative framework - fun- Works Committees had been established at four spinning mills (mills 11, 12, 13 and 16) and the workers could damental labour standards name the chair of the committee. At two of the mills, workers said complaints had been submitted to this commit- 4.1. International guidelines for tee. At mill 13, a complaint about sexual harassment had been submitted to the committee. It is not clear how often responsible business conduct 4.2. Fundamental labour standards complaints were taken seriously and/or resolved. There are both positive and negative signals. 4.3. Forced labour indicators In some cases, workers were able to reach out to NGOs. 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu ‘The Committee is there, but complaints are not taken seriously. It is up to the 4.6. Caste-based discrimination supervisor to handle the situation.’ legislation in India Worker, mill 13. 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 ‘Supervisor and warden atrocities were solved.’ investigated mills Worker, mill 16. 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO Workers from mill 16 said that when they had reached out, on occasion, to NGOs, the NGO had helped resolve indicators some issues. 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief ‘We cannot report to anyone. We share some minor issues to the hostel warden. But 6.3 After lockdown the issues such as related to sexual harassment we do not share with anyone. There 6.4 Safety measures inadequate is a committee, but we do not complain. It is often inactive.’ 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on ‘But sometimes we share with NGOs, and they have talked to the factory and solved corporate disclosure some issues.’ 7.3 Links with vertically integra- ted manufacturers FGD, mill 16. 7.4 Company responses with regard to the draft research findings Living conditions 8. Conclusions: Beyond 29 mills As described previously, many workers have no choice but to stay in the mill-run hostel, and management often – a sector risk analysis 8.1 Incidence and risks of forced require workers to stay in such hostels. labour 8.2 Lack of publicly available data and Hostel workers described their accommodation as basic. There are no separate bedrooms and living rooms. Work- supply chain transparency ers have to share rooms with many people, often upto ten or 20 others, and sometimes with more than 20. 8.3 Flawed business model 8.4 Restricted interpretation of Workers on different shifts use the same room; while those on one shift are at work, those on the other shift use supply chain responsibility the room. The rooms are small, hot, stuffy, and badly ventilated. The electricity supply is irregular. Workers are 8.5 Reliance on social auditing provided with just a simple sleeping mat and pillow, no other bedding or furniture. The toilets and bathroom are 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 64 research

Executive summary

1. Introduction 1.1. Why this research? shared between 20 or 30 people. Workers said the hostels were untidy and the facilities, unhygienic. Workers are 1.2. Reading guide responsible for keeping the hostel tidy, no cleaning service is provided, but given their long working hours, it is 2. Methodology 2.1. Selection of mills no surprise that workers have little time or energy for chores. The job of cleaning the toilets often falls to female 2.2. Field research workers at the bottom of the social hierarchy, such as those with a migration or Scheduled Castes background. 2.3. Supply chain research 2.4. Company review ‘The hostel is overcrowded. Management is not concerned about this. They are not 2.5. Methodological challenges 3. Context concerned about workers’ health and wellbeing.’ 3.1. India’s cotton, textile, and Worker, mill 1. garment industry 3.2. The Tamil Nadu textile and garment industry ‘No hygiene is maintained here. We are not given good food. For the name sake they 3.3. Trade unions are giving food and accommodation. They are not treating us as human beings.’ 3.4. Social exclusion based on Worker, mill 1. caste and tribal background 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu 4. Normative framework - fun- 11. Excessive overtime damental labour standards 4.1. International guidelines for responsible business conduct Excessive overtime 4.2. Fundamental labour standards ‘Forced labourers may be obliged to work excessive hours or days beyond the limits prescribed 4.3. Forced labour indicators by national law or collective agreement. They can be denied breaks and days off, having to 4.4. Legislative developments and take over the shifts and working hours of colleagues who are absent, or by being on call 24 existing supply chain legislation 4.5. Labour laws in India and hours a day, 7 days a week. Tamil Nadu 4.6. Caste-based discrimination The determination of whether or not overtime constitutes a forced labour offence can be quite legislation in India complex. As a rule of thumb, if employees have to work more overtime than is allowed under 5. Evidence of existence and national law, under some form of threat (e.g. of dismissal) or in order to earn at least the min- risk of forced labour 5.1. Characteristics of the 29 imum wage, this amounts to forced labour.’ investigated mills ILO forced labour indicators119 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 65 research

Executive summary

1. Introduction 1.1. Why this research? Excessive, and forced, overtime is a pervasive issue in the Tamil Nadu spinning mills. The picture painted by work- 1.2. Reading guide ers is of working days regularly lasting 12 hours or more. The issue of excessive overtime provoked a lot of re- 2. Methodology 2.1. Selection of mills sponses. Workers spoke in detail, and with vehemence, about the hardship of working such long hours. They said 2.2. Field research they were never able to properly rest because of the cumulative effect of working ridiculously long standard shifts, 2.3. Supply chain research forced overtime, and being around noisy machines all the time. 2.4. Company review 2.5. Methodological challenges Spinning mills operate 24 hours a day. Mills usually have a 3 x 8-hour shift system in place to accommodate this. 3. Context 3.1. India’s cotton, textile, and 170 workers, working at 10 of the 29 investigated spinning mills, said that in their mills there were only two shifts – garment industry a day shift and a night-shift – of 12 hours each. At six of these 10 mills (mills 3, 4, 5, 28, 29, and 30), all the workers 3.2. The Tamil Nadu textile and were obliged to work 12-hour shifts. At the other four mills (mills 1, 2, 6, and 11), it appeared that some workers garment industry worked 8-hour shifts, and other workers worked 12-hour shifts. 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background Indian law prescribes that workers should get one day of rest a week, but 283 of our respondents, working for 16 3.5. Caste discrimination in garment different mills (mills 1, 2, 3, 4, 5, 11, 12, 13, 15, 16, 19, 20, 22, 28, 29, and 30), said they often had to work seven factories and spinning mills in days in a row. Tamil Nadu 4. Normative framework - fun- damental labour standards 90 workers said they had to work 12 hours a day, for seven consecutive days (a weekly working total of a stagger- 4.1. International guidelines for ing 84 hours). This issue concerned all workers interviewed at mills 28, 29 and 30, and a number of those inter- responsible business conduct viewed at mills 1, 2 and 6120. Some workers even spoke of working 24 hours straight. 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and ‘We complete our shift and if some workers of the next shift do not come in time we existing supply chain legislation have to continue. Sometimes we go at 6 a.m. and come back only at 6 a.m. next day. 4.5. Labour laws in India and Tamil Nadu I continue to do this since I am poor.’ 4.6. Caste-based discrimination Female worker (age 33), mill 13. legislation in India 5. Evidence of existence and ‘They are making us do the day and night shift continuously, without any rest. There risk of forced labour 5.1. Characteristics of the 29 is even no time to go to the toilet. We are always working. We are even not allowed investigated mills to go home for holidays.’ 5.2. Respondents’ profiles 5.3. Risks and evidence of forced Female worker, mill 13. labour – using the 11 ILO indicators ‘Both day and night shifts are there. Most of the days we have to do both the shifts. 6. Impact of COVID-19 6.1 Loss of income The machine will keep running, so we can’t stop the work. If we stop the supervisor 6.2 Emergency relief will scold. We cannot take leave, if we take leave, our wages will be delayed. We 6.3 After lockdown have to keep working for the sake of our families and most of us are in debt. We bear 6.4 Safety measures inadequate 7. Responses to the research all these for the sake of family, for our children’s education.’ 7.1 Supply chain links FGD, mill 4. 7.2 Links established based on corporate disclosure Workers at mills 11, 16 and 22, who participated in focus group discussions, said that they sometimes had to work 7.3 Links with vertically integra- ted manufacturers two double shifts. 7.4 Company responses with regard to the draft research Over half the workers interviewed said they could not refuse to work overtime, in addition to their regular shifts findings of eight or 12 hours. 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced In our sample of 29 spinning mills, only two (mills 26 and 27) seemed to comply with Indian law which prescribes labour 60 working hours a week (six days of eight hours = 48 hours + 12 hours overtime), as the maximum. Respondents 8.2 Lack of publicly available data and at these two mills said they did not work any overtime, and that their working week consisted of 48 hours (six days supply chain transparency of an 8-hour shift). 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility Workers living in hostels on the mill compound were particularly at risk of being forced to work excessive hours. 8.5 Reliance on social auditing Respondents explained how management misused the workers’ all-day presence at the compound to call them to 8.6 Frontrunners bear the brunt work whenever the need arose. At one-third of the spinning mills (1, 2, 3, 4, 5, 6, 9, 12, 13, and 21), workers had 8.7 Leverage 9. Recommendations been pulled out of bed in the night in order to work. 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 66 research

Executive summary

1. Introduction 1.1. Why this research? Some hostel workers said they had to work extra shifts if local workers did not show up for their shift. 1.2. Reading guide 2. Methodology 2.1. Selection of mills ‘The machines are on 24 hours per day. I get no proper sleep. They are continuously 2.2. Field research asking us to work both the day and night shift which is very hard.’ 2.3. Supply chain research Female worker (age 23), mill 13. 2.4. Company review 2.5. Methodological challenges 3. Context ‘Even when in the hostel we should be prepared to work anytime. Since there is no 3.1. India’s cotton, textile, and rain in our home town, farming is not possible and we have no other choice then to garment industry 3.2. The Tamil Nadu textile and work in this hard situation.’ garment industry Female worker (age 23), mill 13. 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background ‘Since we are staying in the hostel we should be ready to work in any shift.’ 3.5. Caste discrimination in garment Worker, mill 22. factories and spinning mills in Tamil Nadu 4. Normative framework - fun- ‘Since I am staying in the hostel, the management is making us to do all shifts.’ damental labour standards Worker, mill 25. 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards ‘I do not like to stay in the hostel because they will make us to do more work every 4.3. Forced labour indicators day. They will not give rest, not even during menstruation period.’ 4.4. Legislative developments and Female workers, mill 12. existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu ‘We are pressured to do overtime at night hours. This makes us very tired.’ 4.6. Caste-based discrimination Worker, mill 5. legislation in India 5. Evidence of existence and risk of forced labour ‘We are called for the work at any time while sleeping, we cannot say no, if I say no, I 5.1. Characteristics of the 29 am frightened whether I will be sent off from the work.’ investigated mills 5.2. Respondents’ profiles Worker, mill 13. 5.3. Risks and evidence of forced labour – using the 11 ILO ‘We are working non-stop, without knowing whether it is day or night.’ indicators 6. Impact of COVID-19 Female worker (age 35), mill 20. 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 67 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 6. Impact of COVID-19 2. Methodology 2.1. Selection of mills 2.2. Field research In July 2020, we conducted additional interviews with 15 workers to hear first-hand from them how the corona 2.3. Supply chain research crisis had affected their lives. All the workers had been interviewed previously during the field research. Before 2.4. Company review the outbreak of the COVID-19 pandemic and subsequent lockdown, these workers were employed at eight of the 2.5. Methodological challenges investigated spinning mills; one worker from mill 19 and two workers each from mills 11, 12, 13, 15, 16, 17 and 22. 3. Context 3.1. India’s cotton, textile, and The eight spinning mills were all closed for prolonged periods during the lockdown. Apart from mill 12, which was garment industry closed for four months, the other mills were closed for three months (April to June 2020). 3.2. The Tamil Nadu textile and garment industry The interviewed workers hail from different districts within Tamil Nadu. All of them resided in factory hostels be- 3.3. Trade unions fore the COVID-19 outbreak. Two stayed in their hostel during the lockdown (one worker from mill 11 and mill 13, 3.4. Social exclusion based on caste and tribal background respectively). The other workers were sent home. Workers travelled home by truck or by walking, or both. Nine 3.5. Caste discrimination in garment workers said their employer arranged for transport home. factories and spinning mills in Tamil Nadu 4. Normative framework - fun- 6.1 Loss of income damental labour standards 4.1. International guidelines for During the lockdown period, none of the workers received their regular wages or were given the option of paid responsible business conduct leave. All said that they did not hold out any hope of receiving wages for the lockdown period. 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and 6.2 Emergency relief existing supply chain legislation 4.5. Labour laws in India and The only financial compensation that workers received was a one-off payment of INR 1,000 (€ 11.51). 12 workers Tamil Nadu said they had received a cash amount of INR 1,000 during the period their mill was closed. One worker (from mill 4.6. Caste-based discrimination 13) said she received INR 4,000 (€ 46.05). Two workers, both from mill 11, did not receive any money at all. These legislation in India cash payments seemed to be financial support from the Tamil Nadu government made to ration card holders. 5. Evidence of existence and risk of forced labour When the lockdown was announced at the end of March 2020, the Tamil Nadu state government promised to 5.1. Characteristics of the 29 issue an INR 1,000 cash payment to all ration card holders in the state, along with free rice, sugar, and other essen- investigated mills tial commodities. Despite this, six respondents said they had been denied food rations. Three of these respondents 5.2. Respondents’ profiles said they did not have a ration card and, without a card, could not receive any food relief or materials from the 5.3. Risks and evidence of forced labour – using the 11 ILO government. One worker (from mill 13), said her employer provided food during lockdown. 10 respondents said indicators they received some dry rations from NGOs or charities. One worker received no support whatsoever, while two 6. Impact of COVID-19 others did not answer the question about emergency relief. 6.1 Loss of income 6.2 Emergency relief The loss of income led to profound economic hardship for the workers, especially as their families also often de- 6.3 After lockdown 6.4 Safety measures inadequate pended on the income.The interviewees spoke about their struggles to survive and their inability to support their 7. Responses to the research families. 7.1 Supply chain links 7.2 Links established based on corporate disclosure ‘I have been without work for three months, there is no money. I cannot buy food or 7.3 Links with vertically integra- clothes for my children.’ ted manufacturers Worker (age 23), mill 15. 7.4 Company responses with regard to the draft research findings ‘We lost our livelihoods and we are struggling for survival.’ 8. Conclusions: Beyond 29 mills Worker (age 19), formerly mill 12, now working in another spinning mill. – a sector risk analysis 8.1 Incidence and risks of forced labour ‘I was sent away from the mill. My family has no own house so we were living in a 8.2 Lack of publicly available data and rented house. There is no money to pay the rent. The house owner asked us to leave. supply chain transparency 8.3 Flawed business model There is no money for expenses and we were denied ration rice [free rice from the 8.4 Restricted interpretation of government] because we have no ration card. We are surviving with old rice.’ supply chain responsibility Worker (age 24), mill 16. 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 68 research

Executive summary

1. Introduction 1.1. Why this research? ‘We are struggling for food.’ 1.2. Reading guide 2. Methodology Worker (age 19), formerly mill 13, now doing farm work. 2.1. Selection of mills 2.2. Field research ‘I am not able to support my family; my parents are in crisis. I am struggling to pay 2.3. Supply chain research 2.4. Company review my loan instalments.’ 2.5. Methodological challenges Worker (age 22), mill 13. 3. Context 3.1. India’s cotton, textile, and garment industry ‘More than COVID-19 I am worried that I will die of hunger.’ 3.2. The Tamil Nadu textile and Worker (age 27), mill 19. garment industry 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background 6.3 After lockdown 3.5. Caste discrimination in garment factories and spinning mills in All respondents said that, after lockdown, the workforce at their mill was reduced by about half, and that wom- Tamil Nadu en, young workers, migrants, and workers with a Scheduled Caste or Scheduled Tribes background were among 4. Normative framework - fun- those most likely to lose their jobs. Two respondents (both from mill 12) said that dismissed workers had received damental labour standards 4.1. International guidelines for compensation of INR 6,000 (€ 69.08). The other 13 workers said that dismissed workers had not received any responsible business conduct compensation. 4.2. Fundamental labour standards 4.3. Forced labour indicators The interviewees said that the workforce had been reduced because of a dramatic decrease in demand for prod- 4.4. Legislative developments and ucts, and problems with delivery of inputs (cotton) since all business had closed down. existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu Four respondents no longer worked for their former employer. Two respondents worked at a mill (mill 12) that 4.6. Caste-based discrimination had not reopened at the time of the interview, so had begun working at another mill. The other two respondents legislation in India – aged 19 and 20, respectively, with a Scheduled Castes background - were laid off as their mills (mills 11 and 13) 5. Evidence of existence and risk of forced labour reduced their workforce. One of them had found work in another mill, and the other had taken on farm work. She 5.1. Characteristics of the 29 now usually works seven days a week and nine hours a day, depending on the work that needs to be done. investigated mills 5.2. Respondents’ profiles Workers that had returned to their mills, and those who had found work in other mills, reported that they now 5.3. Risks and evidence of forced worked less days in a week but more hours in a day, and that their salary had been reduced. labour – using the 11 ILO indicators 6. Impact of COVID-19 Working weeks have gone down from six or seven days (see Chapter 5) to between two and four days. Four 6.1 Loss of income workers (one from mill 11, two from mill 15, and one worker who changed to another mill) now work four days a 6.2 Emergency relief week. Five workers work three days a week (mills 11, 16, 19, 22 and one worker who changed to another mill), 6.3 After lockdown 6.4 Safety measures inadequate and another five workers said they now work two or three days a week (workers from mills 13, 16, and 22, and 2 7. Responses to the research workers from mill 17). 7.1 Supply chain links 7.2 Links established based on While the number of working days has decreased, the length of the working day has been extended at all mills, corporate disclosure 7.3 Links with vertically integra- usually to 14 hours a day. Five workers mentioned days of between 12 and 14 hours, and nine workers mentioned ted manufacturers days of 14 hours. 7.4 Company responses with regard to the draft research The interviewed workers also said they were no longer paid for overtime, and their daily wage had decreased. Eight findings workers from mills 15, 16, 17 and 22, and one worker from mill 11, said they now received daily wages of INR 170 8. Conclusions: Beyond 29 mills – a sector risk analysis – 200 (€ 1.96 – 2.30) compared to their pre-corona daily wage of INR 200 – 300 (€ 2.30 - 4.03). One worker (from 8.1 Incidence and risks of forced mill 13) said that while her daily wage had stayed the same, the number of working hours in her day had increased. labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 69 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 6.4 Safety measures inadequate 2. Methodology The 14 respondents who were still working in a spinning mill said that the following safety measures had been 2.1. Selection of mills introduced at their workplace: 2.2. Field research 2.3. Supply chain research 2.4. Company review • Screening body temperature 2.5. Methodological challenges 3. Context • Use of face mask 3.1. India’s cotton, textile, and • Provision of hand sanitiser garment industry 3.2. The Tamil Nadu textile and • Additional handwashing facilities garment industry 3.3. Trade unions • Disinfection of work spaces 3.4. Social exclusion based on caste and tribal background • Measures to maintain social distancing inside the factory. 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu Nevertheless, the interviewed workers had serious doubts about the adequacy of these measures. Seven workers 4. Normative framework - fun- (at mills 15, 16, 17, 19 and 22) said there was no way to maintain social distancing in the hostel and that they were damental labour standards afraid of contracting the virus there. One worker (mill 11) said she was afraid that outside workers (non-hostel 4.1. International guidelines for workers) would bring the virus to the factory. Only one worker (from mill 13) thought the measures taken by her responsible business conduct 4.2. Fundamental labour standards employer were adequate. 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 70 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 7. Responses to the research 2. Methodology 2.1. Selection of mills 2.2. Field research In December 2020, SOMO and Arisa shared draft chapters of this report with garment factories, brands and retail- 2.3. Supply chain research ers that we understood to be directly or indirectly sourcing from the spinning mills under investigation, based on 2.4. Company review information we retrieved from public supplier lists and shipment records extracted from Panjiva. In this report the 2.5. Methodological challenges names of the mills are not given, instead we have numbered the mills. The companies that we contacted for the 3. Context 3.1. India’s cotton, textile, and review were, however, informed of the names and addresses of the mills under investigation. garment industry 3.2. The Tamil Nadu textile and We reached out to the Dutch Agreement on Sustainable Garments and Textile (AGT), the German Partnership for garment industry Sustainable Textiles (PST), Ethical Trading Initiative (ETI), Fair Wear Foundation (FWF), Fair Labor Association (FLA), 3.3. Trade unions amfori, Better Cotton Initiative (BCI), and the Global Organic Textile Standard (GOTS) asking them to figure out 3.4. Social exclusion based on caste and tribal background whether any of their corporate members had direct or indirect sourcing relations with the 29 mills under investiga- 3.5. Caste discrimination in garment tion (or with the vertically integrated companies that mills belong to). Draft research findings were made available. factories and spinning mills in In a later stage, SOMO and Arisa had direct contact with corporate members of AGT, FWF and amfori, as a result Tamil Nadu of interventions by these initiatives’ secretariats/staff. 4. Normative framework - fun- damental labour standards 4.1. International guidelines for SOMO and Arisa reached out to a number of brands and retailers that we found to have sourcing relationships responsible business conduct with four garment factories in Bangladesh, China, and Sri Lanka which in turn sourced yarns from one or more of 4.2. Fundamental labour standards the mills under investigation. 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation This chapter summarises the responses to our research that we received from companies, MSIs, RBC initiatives 4.5. Labour laws in India and and certification schemes. Tamil Nadu 4.6. Caste-based discrimination legislation in India 7.1 Supply chain links 5. Evidence of existence and risk of forced labour Cotton yarn and fabrics produced in Tamil Nadu are processed in the Tamil Nadu and Indian garment industry 5.1. Characteristics of the 29 before being exported to their final destination – often the European or US market - in the form of clothing and investigated mills household textiles. Yarn itself is often exported to countries such as Bangladesh, China and Sri Lanka, where 5.2. Respondents’ profiles clothes and textile items are produced, mainly for export. 5.3. Risks and evidence of forced labour – using the 11 ILO indicators In this research, it has been difficult to establish links between the investigated spinning mills, fabric manufactur- 6. Impact of COVID-19 ers, garment manufacturers and the brands and retailers who are the ultimate clients. 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown In the Panjiva database there were no recent export records, or sometimes no export records at all, for almost half 6.4 Safety measures inadequate of the mills being investigated, suggesting that these mills supply yarn to other Indian factories who may export to 7. Responses to the research international clients. The lack of transparency surrounding this process makes it difficult to establish links between 7.1 Supply chain links the Tamil Nadu yarn producers and the Indian garment manufacturers. With a few exceptions, most of those in- 7.2 Links established based on corporate disclosure volved in the supply chain - spinning mills, garment factories, clothing brands and retailers - are not forthcoming 7.3 Links with vertically integra- with information about their supplier and client base. While an increasing number of brands and retailers are dis- ted manufacturers closing their supplier lists, this transparency tends to be limited to the suppliers of end-products. 7.4 Company responses with regard to the draft research We are aware that we have only been able to establish a limited number of supply chain links. There are many other findings 8. Conclusions: Beyond 29 mills corporate actors involved besides the ones mentioned in this report. – a sector risk analysis 8.1 Incidence and risks of forced In summary, brands and retailers may be directly or indirectly linked to one or more of the investigated spinning labour units in various ways. Below, we indicate which scenario applies to the companies that we contacted as part of the 8.2 Lack of publicly available data and supply chain transparency review procedure for this research report. 8.3 Flawed business model 8.4 Restricted interpretation of SOMO and Arisa use the term ‘direct link’ to mean cases where yarn and/or fabric used in the brand/retailer’s supply chain responsibility products comes from one of the investigated mills. This does not mean that the brand/retailer has a contractual 8.5 Reliance on social auditing relationship with the investigated spinning mill; there may be one or more suppliers and/or agents in between (for 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 71 research

Executive summary

1. Introduction 1.1. Why this research? example, garment factories). We speak of indirect links between brands/retailers and the mills under investigation, 1.2. Reading guide in cases where a brand/retailer sources from a CMT-unit that is part of the same vertically integrated company as 2. Methodology 2.1. Selection of mills the mill we investigated but yarn/ fabrics do not come from the investigated mill but from another spinning unit 2.2. Field research within this company; or even from a spinning unit outside this company. 2.3. Supply chain research 2.4. Company review Scenario 1. (direct link) The investigated spinning units are part of the brand/retailer’s supply chain because 2.5. Methodological challenges the brand/retailer uses a sourcing model to identify the spinning mills that should supply yarn and/or fabrics 3. Context 3.1. India’s cotton, textile, and (known as ‘nominated’ or ‘preferred’ suppliers) for the production of end-products. garment industry Example: Tesco (mill 14) 3.2. The Tamil Nadu textile and garment industry 3.3. Trade unions 3.4. Social exclusion based on Scenario 2. (direct link) A brand/retailer is linked to one of the investigated spinning units through a gar- caste and tribal background menting factory (based in, or outside, India) that sources yarn and/or fabrics from one of the investigated 3.5. Caste discrimination in garment units. The brand/retailer has outsourced the responsibility to select yarn/fabric suppliers to its first-tier factories and spinning mills in Tamil Nadu supplier (such as a garmenting/Cut Make Trim (CMT) factory) or an agent. 4. Normative framework - fun- Examples: Zeeman (mill 6), WE Fashion (mill 3) damental labour standards 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards Scenario 3. (direct link) A brand/retailer is sourcing from a CMT unit in a vertically integrated company, and 4.3. Forced labour indicators the investigated spinning unit is part of the same company. Yarn and/or fabrics are supplied by the investi- 4.4. Legislative developments and gated unit. existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination Scenario 4. (indirect link) A brand/retailer is sourcing from a CMT unit of a vertically integrated company, legislation in India and the investigated spinning unit is part of the same company. The yarn and/or fabrics, however, are sup- 5. Evidence of existence and plied by a different unit or company. It is impossible for SOMO and Arisa to verify these claims. risk of forced labour 5.1. Characteristics of the 29 Examples: IKEA (mill 19), Gap (mill 7), Next (mill 7), Sainsbury’s (mill 7), The Cookie Company (mill 27), Marc O’Polo investigated mills (mill 14) 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators In scenarios 1, 2 and 3, brands and retailers are directly linked to the investigated unit(s), because yarn and/or 6. Impact of COVID-19 fabrics produced by the mills are used in the production of their goods. 6.1 Loss of income 6.2 Emergency relief In scenario 4, the yarn and/or fabrics are supplied by other units or companies. 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research Orders for brands and retailers are, supposedly, processed in designated units. While these units may be audited, 7.1 Supply chain links other units in the same building or which are part of the same company, may not be audited by the brand/retailer. If 7.2 Links established based on non-compliances are found in the audited unit, then corrective actions may be taken, but non-compliances in other corporate disclosure units will go unnoticed and uncorrected. As described in Chapter 5, many factories in Tamil Nadu focus on various 7.3 Links with vertically integra- ted manufacturers production processes and may also house multiple spinning units. Most of the brands and retailers we spoke with 7.4 Company responses with did not provide details on how they ascertain which unit processes the yarns and fabrics they order. regard to the draft research findings SOMO and Arisa are of the opinion that also in the latter case, there is a responsibility to conduct ‘heightened 8. Conclusions: Beyond 29 mills – a sector risk analysis human rights due diligence’. If labour rights issues become apparent, for example by means of CSO reports, in units 8.1 Incidence and risks of forced belonging to the same company as where your products are made regardless of the direct or indirect relation to the labour problematic unit, then these are signals to be taken seriously and require the buyer to investigate and take action. 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model SOMO and Arisa also contacted a number of brands and retailers that source from garment factories based in 8.4 Restricted interpretation of Bangladesh, China, and Sri Lanka, which in turn source from one or more of the investigated spinning units. Some supply chain responsibility companies stated that the yarn used in their products comes from other producers. If this is the case, there is no 8.5 Reliance on social auditing link between the brand/retailer and the investigated spinning unit, which creates another scenario: 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 72 research

Executive summary

1. Introduction 1.1. Why this research? Scenario 5. A brand/retailer is sourcing from a garment factory that is sourcing from one of the investigated 1.2. Reading guide units, but the yarns/fabrics used in the products of the brand/retailer come from another supplier 2. Methodology 2.1. Selection of mills Though some of the companies that responded to SOMO and Arisa, explained their sourcing relation- 2.2. Field research ship with the investigated companies, the nature of that relationship still sometimes remained un- 2.3. Supply chain research 2.4. Company review clear. It is impossible for SOMO and Arisa to verify claims about the origin of yarn used in these com- 2.5. Methodological challenges panies products. A number of companies stated that they sourced from CMT units that are part of the 3. Context same company as the investigated units, but that the yarn used for their goods is produced elsewhere. 3.1. India’s cotton, textile, and The Cookie Company Group (TCCG) , with regard to mill 27, shared blockchain-based documentation which listed garment industry the different producers and suppliers involved in the production of TCCG goods linked to the CMT unit that we 3.2. The Tamil Nadu textile and garment industry identified. In this case, yarns used for the products of TCCG did not come from mill 27 but from another spinning 3.3. Trade unions unit within that larger company. 3.4. Social exclusion based on caste and tribal background Mill 7 in our report is part of a vertically integrated company. Carrefour informed us that they previously had 3.5. Caste discrimination in garment factories and spinning mills in a sourcing relationship with that company, but that the relationship had since ended. Carrefour did not clarify Tamil Nadu whether there had been a sourcing relationship directly with mill 7, despite our request for clarification. 4. Normative framework - fun- damental labour standards In other cases, SOMO and Arisa only have the word of the companies we questioned. 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards 7.2 Links established based on corporate disclosure 4.3. Forced labour indicators 4.4. Legislative developments and The Dutch Agreement on Sustainable Garments and Textile (AGT) is a broad coalition of businesses, business as- existing supply chain legislation sociations, trade unions, NGOs, and the Dutch government. The five stakeholder groups are represented on the 4.5. Labour laws in India and steering committee and have equal votes. The aim of the AGT is to improve working conditions, prevent pollution, Tamil Nadu 121 4.6. Caste-based discrimination and promote animal welfare in production countries. The AGT currently has 81 corporate signatories . legislation in India 5. Evidence of existence and Corporate members of the Dutch AGT are required to conduct due diligence and progressively map their supply risk of forced labour chains. The AGT discloses an aggregated list of suppliers that are part of AGT members’ supply chains. This list 5.1. Characteristics of the 29 122 investigated mills includes suppliers of yarn and fabrics . To date, the Dutch AGT is the only initiative that discloses supplier infor- 5.2. Respondents’ profiles mation beyond the first tier. 5.3. Risks and evidence of forced labour – using the 11 ILO Parties in the Dutch AGT have developed a number of collective projects. One of these projects aims to improve indicators 6. Impact of COVID-19 working conditions in Tamil Nadu. The following themes are central to this project: discrimination and gender, 6.1 Loss of income child labour, forced labour, freedom of association, living wage, and health and safety in the workplace. The project 6.2 Emergency relief began in September 2020 and will run for three years123. It includes a training programme for worker committees 6.3 After lockdown at 75 factories and spinning mills, and training courses for the residents of 35 hostels. The regional helpline that 6.4 Safety measures inadequate factory workers can call to report malpractices will also be upgraded and linked to the complaints and dispute 7. Responses to the research 124 7.1 Supply chain links mechanism under the Agreement . 7.2 Links established based on corporate disclosure At the time of our research, five of the investigated mills were included in the AGT’s supplier list, which means that 7.3 Links with vertically integra- at least one AGT member has sourcing relationships with these mills. Following our request for further information, ted manufacturers 7.4 Company responses with the AGT Secretariat consulted their internal database and identified the AGT member(s) in question. As a next step, regard to the draft research the AGT Secretariat reached out to these member companies and encouraged them to contact SOMO and Arisa. findings Three companies did contact us: 8. Conclusions: Beyond 29 mills – a sector risk analysis • The Cookie Company Group (TCCG) 8.1 Incidence and risks of forced labour • Zeeman 8.2 Lack of publicly available data and supply chain transparency • WE Fashion 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 73 research

Executive summary

1. Introduction 1.1. Why this research? Fair Wear Foundation (FWF) describes itself as a ‘partnership for a more ethical garment industry’. FWF brings 1.2. Reading guide together different stakeholders; the board consist of 50 per cent business associations and 50 per cent trade 2. Methodology 125 2.1. Selection of mills unions and NGOs . With more than 140 member brands and staff in production countries across Asia, Europe, 126 2.2. Field research and Africa, FWF works towards ‘improving the lives of garment workers around the world’ , focusing on garment 2.3. Supply chain research production - specifically sewing – and the cutting and trimming processes127. 2.4. Company review FWF discloses an aggregated list of CMT factories that supply its members. Following a request from us, FWF 2.5. Methodological challenges checked the available supply chain information in their system and reached out to member companies that source 3. Context 3.1. India’s cotton, textile, and in India to check whether any links existed between the investigated mills and FWF members. As a result, FWF garment industry member Marc O’Polo contacted SOMO and Arisa. Zeeman is also a member of FWF, but contact with Zeeman was 3.2. The Tamil Nadu textile and facilitated by the AGT. garment industry 3.3. Trade unions 3.4. Social exclusion based on amfori is a business initiative that brings together over 2,400 retailers, importers, brands, and associations from caste and tribal background more than 40 countries. Social compliance auditing is an important element of amfori’s approach. amfori also 3.5. Caste discrimination in garment provides a toolbox of human rights due diligence instruments, such as capacity building, stakeholder engagement, factories and spinning mills in and advocacy. The amfori platform is a single point for all supply chain performance information, and amfori audits Tamil Nadu can also be accessed by amfori members through this platform. These audits mean that the amfori Secretariat has 4. Normative framework - fun- damental labour standards access to supply chain information such as the relationships between amfori members and their first-tier suppliers. 4.1. International guidelines for Based on supply chain information in the amfori database, on behalf of SOMO and Arisa, amfori reached out to a responsible business conduct number of their member companies that could have direct or indirect links with one or more of the facilities under 4.2. Fundamental labour standards investigation. Because of amfori’s focus on first-tier suppliers (CMT units), we expected links could be established 4.3. Forced labour indicators 4.4. Legislative developments and between amfori members and vertically integrated manufacturers in Tamil Nadu, or between amfori members and existing supply chain legislation the four garment factories in Bangladesh, China, India, and Sri Lanka respectively that we had found to source from 4.5. Labour laws in India and the Tamil Nadu spinning mills under investigation. Tamil Nadu 4.6. Caste-based discrimination legislation in India Responding to our call, seven amfori members reached out to us. We have not been able to establish a full picture 5. Evidence of existence and of the possible direct or indirect links between the seven amfori members and the 29 mills under investigation. risk of forced labour 5.1. Characteristics of the 29 Ethical Trade Initiative (ETI) is an ‘alliance of companies, trade unions and NGOs that promotes respect for workers’ investigated mills rights around the globe’128. With a combined turnover of over £166 bn, ETI company members include supermar- 5.2. Respondents’ profiles 5.3. Risks and evidence of forced kets, fashion retailers, department stores, and stone sourcing companies, as well as major suppliers to retailers of labour – using the 11 ILO food and drink, flowers, clothing, shoes, homewear, promotional and other products. Many ETI company members indicators are based in the UK, but there are also members in Australia, Germany, Spain, Sweden, and the US. At the time 6. Impact of COVID-19 of writing, ETI had 68 full members129. All corporate members of ETI are expected to adopt the ETI Base Code of 6.1 Loss of income 130 6.2 Emergency relief Labour Practice . 6.3 After lockdown 6.4 Safety measures inadequate SOMO and Arisa asked the ETI Secretariat to help identify ETI company members that may source from the inves- 7. Responses to the research tigated spinning units. SOMO and Arisa shared the names of the investigated units with ETI. ETI agreed to reach 7.1 Supply chain links 7.2 Links established based on out to member companies and request they share the names of their yarn suppliers in Tamil Nadu. ETI would then corporate disclosure check this information against our list of investigated spinning mills, and facilitate contact between relevant mem- 7.3 Links with vertically integra- bers and SOMO/Arisa. It is not clear, at the time of writing, what steps ETI has taken. We did not hear back from ted manufacturers ETI, or from any ETI members, as a result of the intervention of the ETI Secretariat. 7.4 Company responses with regard to the draft research findings Fair Labor Association (FLA) is a collaborative effort of universities, civil society organisations, and socially responsi- 8. Conclusions: Beyond 29 mills ble companies dedicated to protecting workers’ rights around the world131. FLA has 60 company members, most of – a sector risk analysis them based in the US. As affiliates of FLA, companies agree to subject their supply chains to independent assess- 8.1 Incidence and risks of forced ments and monitoring. FLA monitoring is, however, limited to first-tier suppliers, and company members are only labour 8.2 Lack of publicly available data and required to disclose their first-tier suppliers to the FLA secretariat. Because of this, FLA indicated that they could supply chain transparency not be of assistance with regard to establishing links between the investigated spinning units and FLA company 8.3 Flawed business model members. 8.4 Restricted interpretation of supply chain responsibility The German Partnership for Sustainable Textiles (PST)was founded in 2014 ‘to improve the social and environmental 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt conditions within global textile production’. The PST brings together industry stakeholders (companies and asso- 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 74 research

Executive summary

1. Introduction 1.1. Why this research? ciations), NGOs, trade unions, standards organisations, and the German Government. ‘The Partnership strives to 1.2. Reading guide improve conditions in global textile supply networks — from the production of raw materials to the disposal of 2. Methodology 2.1. Selection of mills textiles.’ 2.2. Field research 2.3. Supply chain research 2.4. Company review PST member companies commit to undertaking risk assessments, and to use the results of these assessments to 2.5. Methodological challenges set targets for the following two years. In addition to prevention and mitigation, member companies aim to provide 3. Context 132 3.1. India’s cotton, textile, and remedial action and compensation in the case of negative effects . There are no requirements with regard to garment industry supply chain mapping and public disclosure. 3.2. The Tamil Nadu textile and garment industry PST includes three collective programmes, one of which aims to bring about systemic improvements to working 3.3. Trade unions 3.4. Social exclusion based on conditions in the textile and garment industry in Tamil Nadu, with a focus on the women and girls working in spin- caste and tribal background ning mills. As part of this programme, a training trajectory works towards the establishment of complaint commit- 3.5. Caste discrimination in garment tees in 200 spinning mills and factories133. factories and spinning mills in Tamil Nadu PST members do not disclose supply chain information to the PST Secretariat so the Secretariat has been unable 4. Normative framework - fun- damental labour standards to help us establish potential links between the investigated spinning units and PST company members. 4.1. International guidelines for Better Cotton Initiative (BCI)tracks BCI-certified cotton on its journey from the fields to the final customers: brand responsible business conduct and retailers. A number of the investigated spinning mills process BCI-certified cotton, and are members of BCI. 4.2. Fundamental labour standards BCI has information that would have helped us establish links between the investigated spinning units and brands 4.3. Forced labour indicators 4.4. Legislative developments and and retailers, but it told us that the information was confidential and could not be shared. We asked BCI to reach existing supply chain legislation out to those brands and retailers that it could identify, and encourage them to contact us, but this did not mate- 4.5. Labour laws in India and rialise. Tamil Nadu 4.6. Caste-based discrimination legislation in India Global Organic Textile Standard (GOTS) told us they do not have information about the customers of GOTS-certified 5. Evidence of existence and spinning mills, and therefore could not help us establish supply chain links. GOTS did say, however, that they were risk of forced labour willing to take action with regard to the findings of this research (see below). 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced 7.3 Links with vertically integrated manufacturers labour – using the 11 ILO As explained in Chapter 5, a number of the investigated spinning mills are part of vertically integrated companies indicators 6. Impact of COVID-19 that are also involved in the manufacture of end-products, such as garments and home textiles. Thanks to the 6.1 Loss of income accessibility of Indian customs data (through paid databases), we established links between such companies and 6.2 Emergency relief importing companies. 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research Shipment records found in the databases, covering the research period, showed that the following companies 7.1 Supply chain links sourced from units that are part of the same company as the spinning facilities we investigated: 7.2 Links established based on corporate disclosure • IKEA 7.3 Links with vertically integra- ted manufacturers IKEA sourced, or sources, home textiles from a CMT-unit that is part of the same company as spinning mill 7.4 Company responses with 19 in this report. When asked for a response, IKEA told us that yarn produced in the investigated unit is not regard to the draft research used in IKEA products134. findings 8. Conclusions: Beyond 29 mills • GAP – a sector risk analysis 8.1 Incidence and risks of forced GAP sourced, or sources, from a garmenting unit that is part of the same company asspinning mill 7 in this labour report. GAP informed SOMO that ‘no GAP Inc. products are comprised of fabric from this facility’135. 8.2 Lack of publicly available data and supply chain transparency • NEXT 8.3 Flawed business model 8.4 Restricted interpretation of NEXT had, or has, clothing items made by a garmenting unit that is part of the same company as spinning supply chain responsibility mill 7 in this report. When asked for a response, NEXT told us that this garmenting unit uses yarns from 8.5 Reliance on social auditing different mills, not from spinning mill 7136. 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 75 research

Executive summary

1. Introduction 1.1. Why this research? • Sainsbury’s 1.2. Reading guide 2. Methodology Shipment records indicate that Sainsbury’s was sourcing from a garmenting unit that is part of the same 2.1. Selection of mills company as spinning mill 7 in this report. Sainsbury’s told us: ‘We do not source from the mill that you have 2.2. Field research referenced’. Sainsbury’s added that it was already aware of the issues described in this report and said it 2.3. Supply chain research 137 2.4. Company review would use this report in interactions with its suppliers . 2.5. Methodological challenges • Carrefour 3. Context 3.1. India’s cotton, textile, and When we reached out to Carrefour in December 2020, the company told us it had a commercial relation- garment industry ship with a CMT unit that is part of the same company as mill number 7 in this report. According to Carre- 3.2. The Tamil Nadu textile and four, this relationship no longer exists. Carrefour also stated that, after receiving the draft findings of our garment industry 3.3. Trade unions research, it asked to visit mill 7, but was denied access by the supplier. Carrefour wrote: ‘As a consequence, 3.4. Social exclusion based on we removed this supplier from our current and future supplier list and thus are cutting any future commer- caste and tribal background cial relations with them’138. 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu • The Cookie Company Group (TCCG) 4. Normative framework - fun- damental labour standards Links were established between The Cookie Company and two vertically integrated companies. In the 4.1. International guidelines for first case, TCCG had a sourcing relation with a CMT unit that is part of the same company as mill 27 in responsible business conduct this report. TCCG shared documentation with SOMO and Arisa listing production facilities involved in the 4.2. Fundamental labour standards 4.3. Forced labour indicators production of TCCG-branded goods. The yarns for TCCG did not come from the investigated spinning unit 4.4. Legislative developments and but from another spinning unit part of the same vertically integrated company. existing supply chain legislation 4.5. Labour laws in India and In the second case, we found a link between TCCG and a CMT unit that is part of the same company as mill Tamil Nadu 9 in this report. TCCG informed us that the relation with this company was terminated by the end of 2019. 4.6. Caste-based discrimination TCCG told us this was the result of a lack of supply chain transparency provided by this manufacturer. legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 Table 9: Investigated mills investigated mills 5.2. Respondents’ profiles Facility num- Certification/ International export destina- Connection with brands/ retailers and 5.3. Risks and evidence of forced labour – using the 11 ILO ber in report membership tions MSIs indicators improvement 6. Impact of COVID-19 initiatives 6.1 Loss of income 1 no information found no information found 6.2 Emergency relief 2 GOTS no information found NEXT (ETI, BCI) 6.3 After lockdown 6.4 Safety measures inadequate 3 BCI, GOTS factories in Peru, Vietnam, Bangladesh, WE Fashion (Dutch AGT) 7. Responses to the research etc. 7.1 Supply chain links 4 factories in China no information found 7.2 Links established based on corporate disclosure 5 no information found NEXT (ETI, BCI) 7.3 Links with vertically integra- 6 no information found Zeeman (Dutch AGT, FWF, BCI) ted manufacturers 7 BCI garmenting division exporting to Carrefour (amfori), GAP (BCI), NEXT (ETI, BCI), 7.4 Company responses with regard to the draft research brands and retailers in the US and Sainsburys (ETI, BCI) findings Europe 8. Conclusions: Beyond 29 mills 8 BCI factories in Vietnam, Germany, China, NEXT (ETI, BCI) – a sector risk analysis 8.1 Incidence and risks of forced etc. labour 9 garmenting division exports to com- The Cookie Company (Dutch AGT, amfori) 8.2 Lack of publicly available data and panies in Germany, the Netherlands, supply chain transparency 8.3 Flawed business model France, etc. 8.4 Restricted interpretation of 11 no information found no information found supply chain responsibility 12 no recent information found no information found 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 13 exporting to factories in China no information found 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 76 research

Executive summary

1. Introduction 1.1. Why this research? 14 BCI, GOTS exporting to factories in Indonesia, Tesco (ETI, BCI), Marc O’Polo (FWF), NEXT (ETI, 1.2. Reading guide 2. Methodology China, Singapore, Sri Lanka, etc BCI) 2.1. Selection of mills 15 exporting to factories in China, Viet- no information found 2.2. Field research nam, Bangladesh, etc. 2.3. Supply chain research 2.4. Company review 16 no information found no information found 2.5. Methodological challenges 17 no recent information found no information found 3. Context 18 no information found no information found 3.1. India’s cotton, textile, and garment industry 19 exporting to companies in the US, IKEA (BCI) 3.2. The Tamil Nadu textile and Germany, China, Italy, etc. garment industry 20 no information found no information found 3.3. Trade unions 3.4. Social exclusion based on 21 BCI exporting to factories in Peru and NEXT (ETI, BCI) caste and tribal background Bangladesh 3.5. Caste discrimination in garment 22 no information found no information found factories and spinning mills in Tamil Nadu 23 no information found no information found 4. Normative framework - fun- 24 BCI exporting to factories in Sri Lanka no information found damental labour standards (most recent is August 2019) 4.1. International guidelines for responsible business conduct 25 no information found no information found 4.2. Fundamental labour standards 26 no information found no information found 4.3. Forced labour indicators 27 BCI exporting to factories in the US, Sri The Cookie Company (Dutch AGT, amfori), NEXT 4.4. Legislative developments and existing supply chain legislation Lanka, South Korea, Bangladesh, etc. (ETI, BCI) 4.5. Labour laws in India and 28 no information found no information found Tamil Nadu 29 no recent information found no information found 4.6. Caste-based discrimination legislation in India 30 no information found no information found 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 7.4 Company responses with regard to the draft research findings 5.2. Respondents’ profiles 5.3. Risks and evidence of forced Responding to the draft findings of our research, most companies focused on the nature of their supply chain links labour – using the 11 ILO with the investigated spinning mills (or the absence of such links). Our investigation shows that companies often indicators have a direct sourcing relationship with the CMT-unit of a vertically integrated enterprise, but source yarn from 6. Impact of COVID-19 other suppliers. The lack of detailed responses to the working, living and employment conditions described in this 6.1 Loss of income 6.2 Emergency relief report, suggests that companies do not feel these issues are their responsibilities. 6.3 After lockdown 6.4 Safety measures inadequate Some companies responded to our findings and took action: 7. Responses to the research Tesco confirmed that mill 14 in this report is part of their supply chain. Tesco works with a sourcing model based 7.1 Supply chain links on ‘preferred spinning mills’ and ‘declarations of traceability’. Therefore Tesco knows where the yarns that are used 7.2 Links established based on 139 corporate disclosure in Tesco products originated from. Tesco currently uses 60 per cent BCI cotton and 40 per cent organic cotton . 7.3 Links with vertically integra- Tesco’s top 20 yarn and fabric suppliers are located in India. Tesco told us it has begun including these facilities in ted manufacturers monitoring activities. 7.4 Company responses with regard to the draft research findings The company initiated an investigation with regard to mill 14 in this report, and detected several ‘critical issues’ in 8. Conclusions: Beyond 29 mills relation to wages and benefits, and verbal intimidation. Tesco is now working with the mill on a set of corrective – a sector risk analysis actions. 8.1 Incidence and risks of forced labour and were also identified as having links to and said to be 8.2 Lack of publicly available data and NEXT Marc O’Polo mill 14. NEXT, Marc O’Polo Tesco supply chain transparency interested to undertake interest for jointaction. SOMO and Arisa facilitated contact between the three companies. 8.3 Flawed business model Zeeman and Carrefour undertook steps to start investigations, and enter into dialogue, with the spinning units in 8.4 Restricted interpretation of their supply chains. However, the spinning mills in question mills( 6 and 7) refused to have conversations about the supply chain responsibility issues raised. In the case of Zeeman, the spinning mill even stated that it would no longer supply yarn to Zeeman’s 8.5 Reliance on social auditing 140 8.6 Frontrunners bear the brunt agent Vogueserv . 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 77 research

Executive summary

1. Introduction 1.1. Why this research? Carrefour informed SOMO and Arisa that, because it was denied access to the facility, it removed the mill as a 1.2. Reading guide current and future supplier141. Zeeman stated that, as remediation was impossible due to the unwillingness of the 2. Methodology 142 2.1. Selection of mills mill to enter into dialogue, the company had decided to end its relationship with the spinning mill . 2.2. Field research 2.3. Supply chain research WE Fashion sources yarn from mill 3 in this report, via a garment manufacturer based in Tamil Nadu. In response 2.4. Company review to the draft findings, WE Fashion told us that, in 2020, mill 3 had taken part in a training programme where a local 2.5. Methodological challenges NGO provided sessions on labour rights for workers, middle management, and human resources staff. WE Fashion 3. Context 3.1. India’s cotton, textile, and wrote the following: ‘We are investigating if the same issues that have been reported on by Somo (-), have come up garment industry during the NGO training too, and if this training has changed the situation in this mill’. WE Fashion also stated that it 3.2. The Tamil Nadu textile and would be joining AGT’s Factory Support Programme in Tamil Nadu, which began in 2020. This programme consists garment industry of a worker committees training programme, and focuses on improving worker-management dialogue143. 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background A number of companies said that, while they had a sourcing relationship with a CMT unit that is part of the same 3.5. Caste discrimination in garment company, the yarn/fabrics they use for their products do not come from the investigated mill. Some of these com- factories and spinning mills in panies said that they were concerned by the research findings and are undertaking further action. Marc O’Polo, Tamil Nadu for example, uses144 yarn from a spinning unit that is part of the same company as mill 14 (this company operates 4. Normative framework - fun- damental labour standards several spinning units). Marc O’Polo writes: ‘since the issue reflected in your research is very grave, we want to under- 4.1. International guidelines for stand the exact situation of workers in the unit where our supplier sourced from, so we will commission a monitoring audit responsible business conduct with Fair Wear Foundation in this unit and based on the findings take remediation actions’. 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and In contrast, some companies appear to see no reason for taking action. IKEA, for example, stated: ‘IKEA has no existing supply chain legislation business relationship with mill 19 and therefore the mill is not part of the IKEA supplier programs and IWAY, the IKEA 4.5. Labour laws in India and supplier code of conduct’. The company said that it did not have ‘the legal right to enforce IWAY, conduct audits at mill Tamil Nadu no 19 or refer to mill no 19 as an IKEA supplier’145. 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and NEXT risk of forced labour 5.1. Characteristics of the 29 Following conversations with SOMO and Arisa, NEXT expressed interest in receiving the list of investigated investigated mills units, so the company could investigate if any of the mills were part of its supply chain. We honoured this re- 5.2. Respondents’ profiles quest, and NEXT told us that six of the 29 mills were part of its supply chain. A follow-up conversation between 5.3. Risks and evidence of forced NEXT, SOMO. and Arisa is planned146. labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income GOTS 6.2 Emergency relief 6.3 After lockdown GOTS told us they had contacted the GOTS Certification Bodies (CB) responsible for the certification ofthe 6.4 Safety measures inadequate GOTS-certified mills in our research, and asked them to comment on the findings of this report. At the time of 7. Responses to the research writing, we have not yet received this feedback. GOTS has also asked the concerned CBs to share the most recent 7.1 Supply chain links audit reports, and told us they may ask the CBs to start conducting unannounced audits in these facilities to ensure 7.2 Links established based on that all GOTS requirements are being met147. corporate disclosure 7.3 Links with vertically integra- ted manufacturers A number of companies that we reached out to did not respond at all, including the four garment factories in 7.4 Company responses with Bangladesh, China, and Sri Lanka. regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 78 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 8. Conclusions: Beyond 29 mills – a sector 2. Methodology 2.1. Selection of mills 2.2. Field research risk analysis 2.3. Supply chain research 2.4. Company review This research is based on interviews with 725 workers at 29 mills, which we consider to be a very considerable 2.5. Methodological challenges sample. We found evidence of the existence of human rights and labour rights violations in the 29 researched mills, 3. Context including forced labour. While our findings directly concern the 29 mills we investigated, we feel justified, on the 3.1. India’s cotton, textile, and garment industry basis of this sample, to make extrapolations and present a risk analysis for the Tamil Nadu textile sector as a whole. 3.2. The Tamil Nadu textile and garment industry 3.3. Trade unions 8.1 Incidence and risks of forced labour 3.4. Social exclusion based on For this research, we used the 11 indicators of forced labour, developed by the ILO. Forced labour, according to the caste and tribal background 3.5. Caste discrimination in garment ILO’s definition, is a result of a combination of indicators. On the basis of the evidence we found, we feel justified factories and spinning mills in saying that there are increased risks of human rights and labour rights violations in the Tamil Nadu textile industry, Tamil Nadu including a considerable risk of forced labour. 4. Normative framework - fun- damental labour standards 4.1. International guidelines for The indicators of most concern to us are: abuse of vulnerability (indicator 1), deception (indicator 2), intimidation responsible business conduct and threats (indicator 6), abusive working and living conditions (indicator 10), and excessive overtime (indicator 4.2. Fundamental labour standards 11).These were the indicators most present in researched mills. 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation 1. Abuse of vulnerability 4.5. Labour laws in India and Tamil Nadu This research shows that a considerable proportion of workers in the Tamil Nadu textile industry are labour mi- 4.6. Caste-based discrimination grants. One third of the workers we interviewed hailed from outside Tamil Nadu, with many originating from legislation in India Assam, Bihar, Odisha, and Uttar Pradesh. Poverty, indebtedness, and lack of economic prospects drive workers to 5. Evidence of existence and take jobs in spinning mills in faraway Tamil Nadu. Because of a lack of a shared common language, migrant workers risk of forced labour 5.1. Characteristics of the 29 in Tamil Nadu face barriers when communicating with their employers. There is also a language barrier between investigated mills workers from different states and regions. The research shows that textile workers have little knowledge of labour 5.2. Respondents’ profiles laws and social security regulations. Many textile workers are young; ten respondents were below the age of 18 5.3. Risks and evidence of forced when interviewed. More than one-third of the interviewed workers said there were workers below the age of 18 labour – using the 11 ILO indicators in their mill, incriminating 21 mills. In five of these spinning mills, workers told us that the youngest workers were 6. Impact of COVID-19 14 years of age. Workers also, generally, have little educational background. Many workers are young women, 6.1 Loss of income bound by culture to gendered codes. And, in addition, textile workers often belong to marginalised caste or tribal 6.2 Emergency relief communities. 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research All these factors contribute to a worker’s vulnerability. Workers lack the language, the education, the social-eco- 7.1 Supply chain links nomic status, and the maturity to communicate on an equal footing with management, supervisors, and hostel 7.2 Links established based on staff. This makes them extremely vulnerable to misunderstandings and abuse. corporate disclosure 7.3 Links with vertically integra- ted manufacturers 2. Deception 7.4 Company responses with regard to the draft research For the majority of workers, the information they received during recruitment about the employment and living findings conditions in their prospective jobs turned out to be incorrect. Once they had started the job, and were housed 8. Conclusions: Beyond 29 mills in the hostel, they often found wages to be lower than they had been promised, were working longer hours than – a sector risk analysis 8.1 Incidence and risks of forced promised, and were not entitled to paid annual leave. They also found that money was deducted from their wages labour for food and accommodation, contrary to what they had been told during recruitment. This situation amounts to 8.2 Lack of publicly available data and deceptive recruitment. supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of 3. Restriction of movement supply chain responsibility 8.5 Reliance on social auditing The majority of interviewed workers – 481 out of 725 - lived in hostels on the mills’ premises, and some described 8.6 Frontrunners bear the brunt severe limitations to their freedom. When they are not working, workers must remain in the dormitories. They 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 79 research

Executive summary

1. Introduction 1.1. Why this research? are not free to mix with one another. Hostel wardens were described as very strict, and workers feel constantly 1.2. Reading guide watched. 2. Methodology 2.1. Selection of mills 2.2. Field research Because of their extremely long working hours, workers have very little free time so are unable to pursue their 2.3. Supply chain research own interests. 2.4. Company review 2.5. Methodological challenges While workers are allowed to make trips to town, to the market, on a weekly or monthly basis, they must do so 3. Context 3.1. India’s cotton, textile, and in the presence, and under the supervision, of wardens. There is no other affordable housing for workers in the garment industry vicinity of the mill, so workers have little choice but to live in hostels on the factory premises. These restrictions of 3.2. The Tamil Nadu textile and movement are especially true for women workers. garment industry 3.3. Trade unions 3.4. Social exclusion based on 4. Isolation caste and tribal background 3.5. Caste discrimination in garment Workers living in hostels are not usually allowed visitors. Hostel wardens are tasked to oversee and supervise the factories and spinning mills in workers that live there. Making private phone calls is not easy; phone usage is restricted, and wardens eavesdrop Tamil Nadu on conversations. 4. Normative framework - fun- damental labour standards 4.1. International guidelines for As workers come from various states and speak various languages, there are communication barriers among workers, responsible business conduct which limit social interaction between workers and prevent them organising themselves on the work floor and in hostels. 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and We found that many workers were not aware of the existence of organisations that could offer support or advice. existing supply chain legislation Very few workers were aware of trade unions or labour rights organisations. 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination 5. Physical and sexual violence legislation in India 5. Evidence of existence and Women workers described an abusive atmosphere, both on the work floor and in the hostels. Male managers, risk of forced labour supervisors, hostel staff, and even co-workers, engage in inappropriate touching (such as holding hands and pinch- 5.1. Characteristics of the 29 ing), make sexualised comments, or profess insincere and inappropriate feelings of affection. These behaviours investigated mills often take place under cover of loud machine noises. Sleeping quarters in the hostels are not closed, and women 5.2. Respondents’ profiles described a lack of privacy and safety there. 5.3. Risks and evidence of forced labour – using the 11 ILO indicators Anecdotal as it may seem, the evidence presented in this report should be rigorously considered in the context of 6. Impact of COVID-19 a global culture of sexual harassment in the workplace that is significantly underreported. 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6. Intimidation and threats 6.4 Safety measures inadequate 7. Responses to the research All the interviewed workers described being under huge pressure to work overtime, to meet their employer’s 7.1 Supply chain links schedule and demands. This pressure takes all forms: scolding, getting ‘a black mark on our record’; having to do 7.2 Links established based on an extra shift; threat of dismissal; delayed payment of wages; deductions from wages. Some workers even spoke corporate disclosure 7.3 Links with vertically integra- of ‘beatings and fights’ and ‘blackmailing’. Workers also described experiencing this kind of pressure when they ted manufacturers asked for leave. 7.4 Company responses with regard to the draft research What stood out in these testimonies were the threats, rather than the details about actual penalties or disciplinary findings measures. More than one third of the interviewed workers believed, for example, they would face a financial pen- 8. Conclusions: Beyond 29 mills – a sector risk analysis alty if they left their job. 8.1 Incidence and risks of forced It seems fair to conclude that there is a climate of fear at work in the spinning mills. Workers have internalised labour the threats, and have come to think that whatever their employers and superiors are saying to them will actually 8.2 Lack of publicly available data and happen. Workers see no choice but to shut up and walk the line. supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 7. Retention of identity documents 8.5 Reliance on social auditing Some workers spoke of management confiscating identity documents, leaving them with only a copy of their ID. 8.6 Frontrunners bear the brunt Holding on to workers’ original identity documents is an effective way of tying workers to their employer. 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 80 research

Executive summary

1. Introduction 1.1. Why this research? 8. Withholding of wages 1.2. Reading guide 2. Methodology Workers from six mills said that money was regularly deducted from their salary to go towards an ‘end-of-term 2.1. Selection of mills lump sum’. As many workers do not have an employment contract, appointment letter or any other written confir- 2.2. Field research 2.3. Supply chain research mation of their employment, the precise date of the end of their term of employment is often unclear. 2.4. Company review 2.5. Methodological challenges 3. Context 9. Debt bondage 3.1. India’s cotton, textile, and Some of the interviewed workers spoke about having to pay a fee to the employer’s recruiting agent who found garment industry 3.2. The Tamil Nadu textile and them their job at the mill. During recruitment, the worker’s family were ‘given’ money by the employer via the garment industry agent. This money was given on the proviso that it would be deducted from the worker’s salary at the end of her or 3.3. Trade unions his term of employment. It appears, however, that workers must work for years to pay off this debt. 3.4. Social exclusion based on caste and tribal background 3.5. Caste discrimination in garment 10. Abusive working and living conditions factories and spinning mills in Tamil Nadu Low wages constitute part of the abusive working conditions. These low wages are compounded by financial pen- 4. Normative framework - fun- alties imposed by employers as punishment for various types of misconduct, such as arriving late to work, being damental labour standards absent from work, damaging machines, or losing machine parts or tools. 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards On the basis of the 725 interviews we conducted, we concluded that working in the spinning mills has potentially 4.3. Forced labour indicators severe negative health impacts. The machines are always running and workers work extremely long hours. As a 4.4. Legislative developments and result of these crazy working hours there is little time for proper, restorative night rest. The majority of interviewed existing supply chain legislation workers spoke about being tired and feeling weak. 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination Because workers are on their feet for long hours, many mentioned pain in their legs. The combination of heat, hu- legislation in India midity, and cotton fluff creates an unhealthy work environment and causes respiratory problems. Workers spoke of 5. Evidence of existence and headaches and coughing as common ailments. Everyday injuries included cuts to fingers, and on hands and wrists. risk of forced labour 5.1. Characteristics of the 29 Use of the toilet is restricted to two or three times a day, and each visit has a strict time limit. This is particularly investigated mills problematic for women workers, especially when they are menstruating. 5.2. Respondents’ profiles 5.3. Risks and evidence of forced Workers also spoke about the lack of proper food, and the poor quality of food served in the factory canteens. labour – using the 11 ILO Many reported loss of weight and anaemia. indicators 6. Impact of COVID-19 6.1 Loss of income Hostel accommodation is very basic. Workers sleep and stay in dormitories. Workers share dormitories with ten 6.2 Emergency relief or twenty people, sometimes more. Workers from different shifts use the same dormitories; while one shift is at 6.3 After lockdown work, the other shift uses the room. Workers’ quarters are small, badly ventilated, hot, and stuffy. Electricity in the 6.4 Safety measures inadequate 7. Responses to the research hostels is irregular. Workers have to make do with a basic sleeping mat and pillow, and are not provided with any 7.1 Supply chain links other bedding or furniture. 7.2 Links established based on corporate disclosure The toilets and bathrooms in the hostels are shared by upto thirty people. Workers described the hostels as untidy 7.3 Links with vertically integra- and the facilities as unhygienic. ted manufacturers 7.4 Company responses with regard to the draft research The level of unionisation throughout the Tamil Nadu spinning mills is pathetic, despite that freedom of association findings and the right to collective bargaining are key enabling rights. Only three of the 725 interviewed workers were 8. Conclusions: Beyond 29 mills union members. Workers in only five mills could name the union. None of the investigated mills had a collective – a sector risk analysis 8.1 Incidence and risks of forced bargaining agreement in place. labour Workers were also not guaranteed access to remedy. Internal Complaints Committees (ICC) were established at 8.2 Lack of publicly available data and only eight mills. Workers at only four of these mills could name the chair of the committee, and, at only two mills, supply chain transparency workers indicated that the committees had dealt with complaints. 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 81 research

Executive summary

1. Introduction 1.1. Why this research? 11. Excessive overtime 1.2. Reading guide 2. Methodology Spinning mills operate continuously. Most mills have a system in place of 3 x 8-hour shifts, though a 2 x 12-hour 2.1. Selection of mills shift system is also very common. At one third of the investigated spinning mills workers said that they are regularly 2.2. Field research 2.3. Supply chain research woken up in the middle of the night in order to work. 2.4. Company review 2.5. Methodological challenges 3. Context 8.2 Lack of publicly available data and supply chain transparency 3.1. India’s cotton, textile, and This research was hampered by a serious lack of publicly available information on the spinning mill industry. Data garment industry 3.2. The Tamil Nadu textile and on the Tamil Nadu textile industry, business structures, and supply chains, is either outdated or not available. garment industry 3.3. Trade unions Reliable data simply on the number of spinning mills in Tami Nadu are not available. Furthermore, information on 3.4. Social exclusion based on the size of the mills (number of workers), the type of mill (vertically integrated or stand-alone), their precise ad- caste and tribal background dresses, the names of the enterprises, ownership etc, is hard to come by. 3.5. Caste discrimination in garment factories and spinning mills in Tamil Nadu There is also a lack of information about supply chains. While a growing number of brands and retailers now dis- 4. Normative framework - fun- close supplier lists, these public lists are mostly limited to first-tier suppliers, and spinning mills are usually regarded damental labour standards as third-tier suppliers. Public transparency about third-tier suppliers is rare. Moreover, many brands and retailers 4.1. International guidelines for responsible business conduct do not even map their supply chain up to the level of spinning mills. 4.2. Fundamental labour standards 4.3. Forced labour indicators The public availability of up-to-date supply chain information is essential for workers, labour unions, labour rights 4.4. Legislative developments and organisations, as well as corporate actors and improvement initiatives, to make a concerted response to labour existing supply chain legislation 4.5. Labour laws in India and rights violations. This kind of response is only possible if all those involved in the supply chain make themselves Tamil Nadu known. Transparency about supply chain relationships - volumes and values of orders, purchasing practices, con- 4.6. Caste-based discrimination tracts, prices etc. - is essential to identify the responsibility level of buyers. legislation in India 5. Evidence of existence and Most yarns and fabrics produced in Tamil Nadu are exported to other countries, such as Bangladesh and China, risk of forced labour 5.1. Characteristics of the 29 where they are processed into branded goods. Products made in Bangladesh and China are exported to the EU investigated mills and the US but, because of the lack of export data, it is impossible to tell which brands are involved, and data from 5.2. Respondents’ profiles Panjiva or OAR are inconclusive. 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 8.3 Flawed business model 6. Impact of COVID-19 6.1 Loss of income The business model of the global garment and textile industry has often been characterised as loose, flexible, sup- 6.2 Emergency relief ply chains and complex webs of dozens of intermediaries. The supply chains of brands and retailers often include 6.3 After lockdown many different actors, like buying houses, agents, middle-men and procurement service providers. This research 6.4 Safety measures inadequate 7. Responses to the research confirms that the model of outsourcing production, together with unfair purchasing practices, is at the root of the 7.1 Supply chain links poor practices experienced by workers, such as a lack of living wage, child labour, and structural overtime. 7.2 Links established based on corporate disclosure As part of the review procedure we contacted a group of brands and retailers that we identified as having direct or 7.3 Links with vertically integra- indirect relations with the investigated companies. We shared with them two draft chapters of this report which ted manufacturers 7.4 Company responses with describe the research findings. regard to the draft research findings We found that brands and retailers often outsource the procuring of yarns and fabrics to the garment factories 8. Conclusions: Beyond 29 mills that make their end-products. Brands and retailers do not necessarily have any relationship with the spinning mills – a sector risk analysis 8.1 Incidence and risks of forced that supply yarns and fabrics. In fact, it appears it is not even common practice for brands and retailers to ask their labour first-tier suppliers which spinning mills they work with, let alone ask them about the working conditions at those 8.2 Lack of publicly available data and yarn and fabric suppliers. It is, therefore, very easy for brands and retailers to say that they have no access to, and supply chain transparency power over, this link of the supply chain. 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility We found, however, that while some brands and retailers do not have a direct contractual relationship with the 8.5 Reliance on social auditing spinning mills in their supply chain, they still may enter into dialogue with these facilities when necessary. 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 82 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide Other business models also exist where, for example, an agent, rather than the garmenting factory, is responsible 2. Methodology 2.1. Selection of mills for choosing the yarns and fabrics manufacturers. In another model, brands and retailers work with so-called nom- 2.2. Field research inated or preferred mills to determine where yarns/fabrics are sourced and the production capacity of the mill. In 2.3. Supply chain research the latter case, brands and retailers do have a relationship with fabric manufacturers, making the labour conditions 2.4. Company review in these mills, and the need to improve them where necessary, part of the brand/retailer’s responsibility. 2.5. Methodological challenges 3. Context 3.1. India’s cotton, textile, and In this research, we identified a number of garment factories in Bangladesh, China, India and Sri Lanka, , that are garment industry direct customers of the mills we investigated. We have asked these garment factories about their relationships 3.2. The Tamil Nadu textile and with the mills but have not received any responses. garment industry 3.3. Trade unions 3.4. Social exclusion based on In some cases, brands and retailers reached out to the garment factories in their supply chain, and asked them to caste and tribal background relay our concerns and queries to the mills in their supply chain. 3.5. Caste discrimination in garment factories and spinning mills in While most of the companies that responded to SOMO and Arisa’s review request, explained their business model Tamil Nadu and sourcing relationship with the investigated companies, some of these relationships remained unclear. A num- 4. Normative framework - fun- damental labour standards ber of companies said they did source from garment units that are part of the same company as the investigated 4.1. International guidelines for mills, but that the yarn used in their goods was produced elsewhere. Some companies provided information to responsible business conduct substantiate this claim (such as the names of spinning mills in Tamil Nadu that supply yarns and fabrics to their 4.2. Fundamental labour standards garment suppliers) but others did not. 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and 8.4 Restricted interpretation of supply chain responsibility Tamil Nadu Despite international norms (UNGP/OECD Guidelines) clearly stating that companies are responsibile for condi- 4.6. Caste-based discrimination tions throughout their supply chain, very few of those involved (brands/retailers, MSIs, etc) assumed responsibility legislation in India 5. Evidence of existence and for conditions higher up the supply chain. risk of forced labour 5.1. Characteristics of the 29 Some brands and retailers did assume responsibility for conditions in the units they sourced from but very few took investigated mills responsibility for the sub-standard conditions in units belonging to the same company, if they did not have a direct 5.2. Respondents’ profiles 5.3. Risks and evidence of forced contractual or commercial relationship with that unit. labour – using the 11 ILO indicators Orders for brands and retailers are supposedly processed in designated units. While these units may be audit- 6. Impact of COVID-19 ed, other units in the same building or belonging to the same company are not audited by the brand/retailer. If 6.1 Loss of income non-compliant practices are found in the audited unit, action to correct them might be taken but non-compliant 6.2 Emergency relief 6.3 After lockdown practices in adjacent units will go unnoticed and uncorrected. And, as described in Chapter 3, many textile facto- 6.4 Safety measures inadequate ries in Tamil Nadu have more than one, or even multiple, spinning units. 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on 8.5 Reliance on social auditing corporate disclosure 7.3 Links with vertically integra- Some of the brands and retailers, in their responses to Arisa and SOMO, referred to their standard audit practices ted manufacturers and the outcomes of recent audits, as well as to ad hoc inspections and extra audits undertaken in response to 7.4 Company responses with the findings and allegations we had presented to them. A number of audit firms were mentioned in this context, regard to the draft research such as Intertek. This is not hopeful. The social auditing undertaken by brands and retailers to detect risks and findings 8. Conclusions: Beyond 29 mills violations in their supply chain has long been discarded because of its ineffectiveness. It is frustrating, therefore, to – a sector risk analysis see brands and retailers still relying on these outdated tools. Only one of the brands and retailers that responded 8.1 Incidence and risks of forced to our review request explicitly mentioned getting in touch with local labour rights organisations and other stake- labour holders, which should be a part of human rights due diligence (HRDD). 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of 8.6 Frontrunners bear the brunt supply chain responsibility SOMO and Arisa are aware that companies that are frontrunners in terms of supply chain transparency are more 8.5 Reliance on social auditing easily linked to to human rights and labour rights violations in their supply chains, while their less transparent 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 83 research

Executive summary

1. Introduction 1.1. Why this research? competitors who are very likely to be linked to the same issues stay under the radar. SOMO and Arisa are, never- 1.2. Reading guide theless, of the opinion that it is important to disclose the names of brands and retailers linked to the investigated 2. Methodology 2.1. Selection of mills spinning units, and highlight any efforts these companies make to address the labour rights issues occurring in the 2.2. Field research lower tiers of their supply chain. Such transparency enables collective approaches. In addition, according to the 2.3. Supply chain research UNGP and OECD Guidelines, companies must communicate how adverse human rights impacts are addressed, 2.4. Company review and disclose the relevant details (such as findings on human rights risks and abuses occurring in their supply chains) 2.5. Methodological challenges to show their procedures are adequate. 3. Context 3.1. India’s cotton, textile, and garment industry 3.2. The Tamil Nadu textile and Enhanced transparency needed garment industry 3.3. Trade unions 148 3.4. Social exclusion based on The Transparency Pledge coalition lists the following arguments in favour of supply chain transparency : caste and tribal background 3.5. Caste discrimination in garment • Supply chain transparency can help workers gain faster access to redress for human rights abuses; factories and spinning mills in • Supply chain transparency is central to conducting effective human rights due diligence; Tamil Nadu • Publishing supply chain information builds trust. 4. Normative framework - fun- damental labour standards 4.1. International guidelines for The GoodElectronics Network, SOMO, and the Business, Human Rights and the Environment Re- responsible business conduct search Group (BHRE) also call for disclosure based on the worker’s right to know. Companies are 4.2. Fundamental labour standards responsible for disclosing all information that may impact on, or is necessary to realise, workers’ rights, 4.3. Forced labour indicators 4.4. Legislative developments and lives and livelihoods, to primary rights-holders and to other related parties such as worker represent- existing supply chain legislation atives and representative organisations. States are obliged to ensure that companies do disclose this 4.5. Labour laws in India and information and that workers are able to access and engage meaningfully with such information. This Tamil Nadu makes states and business enterprises ‘duty bearers’149. 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 8.7 Leverage 5.2. Respondents’ profiles 5.3. Risks and evidence of forced The review process clearly showed that many companies are still using a very weak interpretation of HRDD. labour – using the 11 ILO Brands and retailers easily claim that they have no leverage over a spinning facility, either because they have no indicators direct relationship with that facility, and/or because their buying power is limited. 6. Impact of COVID-19 6.1 Loss of income Performance of MSIs/RBIs /certification schemes varies greatly 6.2 Emergency relief 6.3 After lockdown For this research we contacted a number of improvement initiatives relevant to the global garment and textile 6.4 Safety measures inadequate industry: the Dutch AGT, amfori, BCI, ETI, FLA, FWF, GOTS, and the German PST. 7. Responses to the research 7.1 Supply chain links We learned that all these initiatives have different standards and practices when it comes to the supply chain 7.2 Links established based on corporate disclosure transparency of their corporate members. 7.3 Links with vertically integra- ted manufacturers The AGT stands out in a positive way as it requires its member brands and retailers, from the third year of their 7.4 Company responses with membership, to map their supply up to the level of spinning mills, and share this information with the AGT Secre- regard to the draft research findings tariat. A number of AGT members have mapped their supply chain to the level of spinners, and the AGT is disclos- 150 8. Conclusions: Beyond 29 mills ing this information in an aggregated format via the OAR . This approach is to be commended and should serve – a sector risk analysis as an example for other brands and retailers, and also for multi-stakeholder and corporate improvement initiatives. 8.1 Incidence and risks of forced The German PST, FLA, and ETI sit at the other end of the spectrum. The member brands of these initiatives are not labour required to map or disclose their supply chains beyond first-tier suppliers (end-products). FLA and ETI say supply 8.2 Lack of publicly available data and supply chain transparency chain transparency is discussed within their respective membership bases, but there have been no clear-cut poli- 8.3 Flawed business model cies or requirements formulated. 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 84 research

Executive summary

1. Introduction 1.1. Why this research? 1.2. Reading guide 9. Recommendations 2. Methodology 2.1. Selection of mills 2.2. Field research 9.1 Corporate actors 2.3. Supply chain research 2.4. Company review We ask all corporate actors along the garment supply chain, to: 2.5. Methodological challenges 3. Context • Adhere to international laws and standards, Indian laws, and Tamil Nadu state regulations with regard to 3.1. India’s cotton, textile, and garment industry human rights and labour rights; 3.2. The Tamil Nadu textile and • Conduct heightened Human Rights Due Diligence (HRDD), based on a strong interpretation of the current garment industry 3.3. Trade unions international framework on due diligence. 3.4. Social exclusion based on • This includes, amongst other things, incorporating HRDD policy throughout all business operations, caste and tribal background 3.5. Caste discrimination in garment including all departments and phases of product development: from design to materials selection, from factories and spinning mills in procurement to quality control, as well as in the upstream supply chain; Tamil Nadu 4. Normative framework - fun- • This requires ensuring that rights holders, and their representative organisations, have a decisive voice damental labour standards in HRDD processes, in particular those in vulnerable worker categories, such as: women; migrant 4.1. International guidelines for workers; workers from Scheduled Castes, including Dalits, Other Backward Classes; and Scheduled responsible business conduct Tribes (ST); 4.2. Fundamental labour standards 4.3. Forced labour indicators • Engage in rigorous supply chain mapping. This mapping should extend beyond first-tier suppliers and encom- 4.4. Legislative developments and pass both the units that make the end-products, and the facilities that produce yarns and fabrics, as well as existing supply chain legislation all those involved in the process of procuring materials and producing end-products, such as buying houses, 4.5. Labour laws in India and Tamil Nadu agents, and providers of procurement services; 4.6. Caste-based discrimination • Design and operate systems that facilitate the ongoing collection and management of supply chain data, to legislation in India 5. Evidence of existence and facilitate combining data/insights on the origin of materials and supply chain links/production phases. Block risk of forced labour chain technology may be helpful; 5.1. Characteristics of the 29 investigated mills • Publicly disclose supply chain information, in accordance with the highest prevailing standard in this field; 5.2. Respondents’ profiles • Actively commit to furtheringfreedom of association and collective bargaining. Facilitate the establishment 5.3. Risks and evidence of forced labour – using the 11 ILO and functioning of democratically elected, independent, factory-level, trade unions; indicators • Facilitate workers’ access to credible judicial and non-judicial grievance mechanisms. At an operational level, 6. Impact of COVID-19 6.1 Loss of income trade unions should be able to play a legitimate role in addressing labour-related disputes. Grievance redressal 6.2 Emergency relief mechanisms, prescribed by Indian law, should be established and functional at supplier level. Operation- 6.3 After lockdown al-level grievance mechanisms can complement freedom of association and collective bargaining, but are no 6.4 Safety measures inadequate substitute for either; 7. Responses to the research 7.1 Supply chain links • Increase focus on the existence and risk of forced labour: embrace the ILO normative framework with regard 7.2 Links established based on to forced labour and address the 11 ILO forced labour indicators both individually and together; corporate disclosure 7.3 Links with vertically integra- • Increase focus on the existence and risk of child labour: do not accept workers below 18 years in any facility ted manufacturers in the supply chain; 7.4 Company responses with regard to the draft research • When identifying new suppliers, in particular spinning units, select those that have all the legally required findings committees in place, and are open to organising and unionisation; 8. Conclusions: Beyond 29 mills – a sector risk analysis • Engage in good faith, and meaningful dialogue, and seek collaboration with human rights and labour rights 8.1 Incidence and risks of forced organisations at local, state, national, and international level; labour 8.2 Lack of publicly available data and • Actively seek to increase leverage over suppliers. This can be done in a number of ways: by building long- supply chain transparency term, stable buyer-supplier relationships; entering into clear and fair contracts; publicly disclosing business 8.3 Flawed business model relationships; and/or announcing that if there is repeated non-compliance, the business relationship may be 8.4 Restricted interpretation of supply chain responsibility terminated; 8.5 Reliance on social auditing • Manufacturers and brands must ensure that spinning mills in their supply chain do not work with labour 8.6 Frontrunners bear the brunt 8.7 Leverage contractors/agents that make false promises or use deception, as a way of recruiting workers. 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 85 research

Executive summary

1. Introduction 1.1. Why this research? The Apparel and Footwear Supply Chain Transparency Pledge provides a common minimum 1.2. Reading guide standard: 2. Methodology 2.1. Selection of mills • The full name of all authorised production units and processing facilities (printing, embroidery, 2.2. Field research laundry, etc.); 2.3. Supply chain research • The site addresses; 2.4. Company review • The parent company of the business at the site; 2.5. Methodological challenges • Type of products made (apparel, footwear, home textile, accessories, etc.); 3. Context 3.1. India’s cotton, textile, and • Worker numbers at each site (less than 1,000 workers; 1,001 to 5,000 workers; 5,001 to 10,000 garment industry workers; more than 10,000 workers). 3.2. The Tamil Nadu textile and Information to be published in a spreadsheet or other searchable format.151 garment industry 3.3. Trade unions 3.4. Social exclusion based on caste and tribal background Advanced supply chain disclosure involves sharing more data on: 3.5. Caste discrimination in garment • The corporate structure of supplier facilities; factories and spinning mills in • The type of business relationship, including information on the type of contract and the contract Tamil Nadu terms (in as far as disclosure of such terms is not prohibited by law); 4. Normative framework - fun- damental labour standards • Sourced products: HS-code, volume, Free on Board (FoB) value; 4.1. International guidelines for • The labour force of the supplier facilities, in terms of demographic characteristics (gender, age, responsible business conduct migratory status, etc.), employment relationship (directly employed or indirectly employed); 4.2. Fundamental labour standards • Whether, and to what extent, supplier facilities enable freedom of association and the right to 4.3. Forced labour indicators 4.4. Legislative developments and collective bargaining, including whether there is a trade union active in the facility, if collective existing supply chain legislation bargaining takes place and whether a CB Agreement is implemented. This would also include in- 4.5. Labour laws in India and formation on the presence and functioning of the legally required factory-level committees, such Tamil Nadu as Works Committee (WC), Grievance Redressal Committee (GRC) and Internal Complaints Com- 4.6. Caste-based discrimination legislation in India mittee (ICC); 5. Evidence of existence and • The availability and quality of operational-level grievance mechanisms for workers in the supply risk of forced labour chain, both on supplier level, brand-operated mechanisms and mechanisms operated by MSIs, 5.1. Characteristics of the 29 certification schemes and RBC initiatives, including information on complaints received, taken up investigated mills and resolved; 5.2. Respondents’ profiles 5.3. Risks and evidence of forced • Factory-level HRDD processes, including (depending what approaches are taken): risk assessments, labour – using the 11 ILO compliance monitoring, verification, inspections, and/or social audits. Disclosure must encompass: indicators policy, methodology, details on audit firms or other commissioned partners, findings and results, 6. Impact of COVID-19 decisions regarding prioritisation, corrective action plans, and the outcomes of such actions.- Fur 6.1 Loss of income 6.2 Emergency relief thermore, companies must investigate, and share, how their purchasing practices might adversely 6.3 After lockdown affect the human rights of workers in the supply chain; 6.4 Safety measures inadequate • Such information must be shared in the public domain and be made bilaterally available to relevant 7. Responses to the research CSOs and trade unions. 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- Additional format requirements for supply chain disclosure: ted manufacturers • Present information in a language that will be understood by those rights-holders who are directly 7.4 Company responses with involved; regard to the draft research findings • The information must be available free of charge; 8. Conclusions: Beyond 29 mills • Provide regular updates, at least annually; – a sector risk analysis • Ensure that previous versions of disclosed data are archived but remain publicly accessible; 8.1 Incidence and risks of forced • Align published factory lists with the Open Data Standard for the Apparel Sector and submit the labour 8.2 Lack of publicly available data and information to the Open Apparel Registry. supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 86 research

Executive summary

1. Introduction 1.1. Why this research? We specifically ask the Tamil Nadu textile industry to: 1.2. Reading guide • Commit to fair recruitment practices, particularly in relation to migrant workers. This means ensuring that 2. Methodology 2.1. Selection of mills prospective workers receive correct information on their working and living conditions; that employers live up 2.2. Field research to the promises made to recruited workers; that workers are not asked to make advance payments; and that 2.3. Supply chain research workers are formally registered at the relevant administrative authorities; 2.4. Company review 2.5. Methodological challenges • Provide all workers with written employment contracts (EC) and monthly payslips in a format that workers 3. Context can understand. Workers should be given a copy of their EC to keep for their own administration; 3.1. India’s cotton, textile, and garment industry • Implement a standard working week of 48 hours (for example, 3 x 8-hour shifts a day); 3.2. The Tamil Nadu textile and • Pay all categories of workers, from apprentices to skilled workers, a wage that enables them to cover their garment industry 152 3.3. Trade unions basic needs and have a discretionary income . Pay wages via bank transactions as per the law. Discontinue 3.4. Social exclusion based on the practice of withholding wages and/or benefits and unfair, burdensome, loan schemes; caste and tribal background 3.5. Caste discrimination in garment • Make all the legally required social security contributions to the Employees’ Provident Fund (EPF) and the factories and spinning mills in Employees’ State Insurance (ESI); Tamil Nadu 4. Normative framework - fun- • Facilitate the establishment and functioning of democratically elected, independent, factory-level trade un- damental labour standards ions. Engage in good faith social dialogue with these unions with a view to producing collective bargaining 4.1. International guidelines for agreements that are respected and implemented, and regularly renegotiated as per the law; responsible business conduct 4.2. Fundamental labour standards • Facilitate the establishment and functioning of legally required factory-level committees, such as the Works 4.3. Forced labour indicators Committee (WC), Grievance Redressal Committee (GRC), Internal Complaints Committee (ICC) and Safety 4.4. Legislative developments and Committee (SC); existing supply chain legislation 4.5. Labour laws in India and • Engage in good faith dialogue and look to collaborate with human rights and labour organisations, includ- Tamil Nadu ing trade union federations, labour rights organisations, civil society organisations, and multi stakeholder 4.6. Caste-based discrimination initiatives, at a local, state, national, and international level. Train workers, supervisors, and management on legislation in India 5. Evidence of existence and labour rights. Involve workers and labour rights organisations in monitoring working conditions, and grievance risk of forced labour redressal. 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles We ask international garment brands and retailers to: 5.3. Risks and evidence of forced • Acknowledge and address the power imbalance between: brands and retailers, upstream members of labour – using the 11 ILO the supply chain and factory management, middle management and workers. Brands and retailers should indicators 6. Impact of COVID-19 refrain from unilaterally imposing requirements and demands on their suppliers. Balanced buyer-supplier re- 6.1 Loss of income lationships require a two-way approach; 6.2 Emergency relief • 6.3 After lockdown • Refrain from Unfair Trading Practices and adhere to responsible purchasing practices that enable those 6.4 Safety measures inadequate involved in all tiers of the supply chain to be decent employers, offer employment contracts, and pay a living 7. Responses to the research wage to their workers. It is not acceptable for buyers to pressure their suppliers to meet standards on human 7.1 Supply chain links rights without providing them with the means to do so. More details below; 7.2 Links established based on corporate disclosure • Consolidate their supply to muster sufficient leverage over suppliers, including spinning units. Depending on 7.3 Links with vertically integra- the type of production model, possible strategies are: ted manufacturers 7.4 Company responses with • Simplify/shorten supply chains: reconsider the role of agents, middle-men, and buying houses; regard to the draft research findings • Build long term relationships with suppliers: 8. Conclusions: Beyond 29 mills • Take responsibility for identifying yarn and fabric suppliers in the supply chain; – a sector risk analysis 8.1 Incidence and risks of forced • Work with ‘nominated’ spinning mills; labour 8.2 Lack of publicly available data and • Have proper and consistent forecasting (booking production capacity in advance, for example); supply chain transparency 8.3 Flawed business model • Reward mills that respect labour rights by giving them ‘preferred supplier’ status, increasing procure- 8.4 Restricted interpretation of ment prices, and guaranteeing a minimum volume of orders; supply chain responsibility 8.5 Reliance on social auditing • Address issues in cooperation with companies that source from the same mill; 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 87 research

Executive summary

1. Introduction 1.1. Why this research? • Base HRDD policies and practices on a progressive interpretation of what constitutes the production 1.2. Reading guide network. If non-compliant practices become apparent, for example by means of CSO reports, in units 2. Methodology 2.1. Selection of mills belonging to the same company as where your products are made regardless of the direct or indirect 2.2. Field research relation to the problematic unit, then these are signals to be taken seriously and require the buyer to 2.3. Supply chain research investigate and take action. 2.4. Company review 2.5. Methodological challenges Note, total monopolisation of a supplier makes a supplier dependent on the business performance of a single 3. Context buyer. This is not a sustainable model. 3.1. India’s cotton, textile, and garment industry • Monitor and remediate labour issues when they occur at all levels along the supply chain, in collaboration, 3.2. The Tamil Nadu textile and where possible, with first-tier suppliers and/or other buyers. Monitoring should include off-site interviews garment industry with workers; 3.3. Trade unions 3.4. Social exclusion based on • Join credible Responsible Business Conduct initiatives caste and tribal background 153 3.5. Caste discrimination in garment • that are of a binding nature upon the brand members ; factories and spinning mills in • have MSI features; Tamil Nadu 4. Normative framework - fun- • that facilitate enhanced supply chain transparency, and include further tiers, beyond end-product damental labour standards factories – such as yarn and textile producers - in their improvement programmes; 4.1. International guidelines for responsible business conduct • that structurally and meaningfully engage rights-holders and their representative organisations. 4.2. Fundamental labour standards 4.3. Forced labour indicators • No ‘cut & run’: Buyers should refrain from discontinuing relations with suppliers on the basis of non-compli- 4.4. Legislative developments and ance. To ‘cut and run’ is not an effective way of impressing upon suppliers that they should improve on hu- existing supply chain legislation man rights and labour conditions. As well as adhering to fair purchasing practices, buyers must seek to apply 4.5. Labour laws in India and Tamil Nadu leverage, in collaboration with other buyers, to steer suppliers towards improvements. If nothing works, then 154 4.6. Caste-based discrimination responsible disengagement is a last resort . legislation in India 5. Evidence of existence and risk of forced labour Fair purchasing practices include: 5.1. Characteristics of the 29 • Fair contractual terms; investigated mills 5.2. Respondents’ profiles • Fair prices based on fair and ‘open costing’; 5.3. Risks and evidence of forced labour – using the 11 ILO • Mutual agreement between buyer and supplier on production planning and lead times; indicators 6. Impact of COVID-19 • Reasonable payment terms (60 days); 6.1 Loss of income • No unilateral suspension of cancelling of contracts; 6.2 Emergency relief 6.3 After lockdown • Transparency about purchasing practices and costing methodologies. Contracts between buyer and 6.4 Safety measures inadequate supplier should not contain clauses that deny parties the right to publicly disclose such information. 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- EU Directive on business-to-business relationships in the agricultural and food supply chain (2019), ted manufacturers bans the following ten Unfair Trading Practices155: 7.4 Company responses with 1. Payment later than 30 days for perishable agricultural and food products; regard to the draft research 2. Payment later than 60 days for other agri-food products; findings 8. Conclusions: Beyond 29 mills 3. Short-notice cancellations of perishable agri-food products; – a sector risk analysis 4. Unilateral contract changes by the buyer; 8.1 Incidence and risks of forced 5. Payments not related to a sale of agricultural and food products; labour 6. Risk of loss and deterioration transferred to the supplier; 8.2 Lack of publicly available data and 7. Refusal of a written confirmation of a supply agreement by the buyer, despite request of the sup- supply chain transparency 8.3 Flawed business model plier; 8.4 Restricted interpretation of 8. Misuse of trade secrets by the buyer; supply chain responsibility 9. Commercial retaliation by the buyer; 8.5 Reliance on social auditing 10. Transferring the costs of examining customer complaints to the supplier. 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 88 research

Executive summary

1. Introduction 1.1. Why this research? Six further practices are banned unless they are provided for in clear and unambiguous terms in a 1.2. Reading guide supply agreement: 2. Methodology 2.1. Selection of mills 11. Return of unsold products; 2.2. Field research 12. Payment of the supplier for stocking, display, and listing; 2.3. Supply chain research 13. Payment of the supplier for promotion; 2.4. Company review 14. Payment of the supplier for marketing; 2.5. Methodological challenges 15. Payment of the supplier for advertising; 3. Context 3.1. India’s cotton, textile, and 16. Payment of the supplier for staff of the buyer, fitting out premises. garment industry 3.2. The Tamil Nadu textile and garment industry We ask Responsible Business Conduct initiatives, Multi Stakeholder Initiatives and certification schemes to: 3.3. Trade unions 3.4. Social exclusion based on • Include civil society stakeholders, including trade unions, on an equal footing, in decision-making, consulta- caste and tribal background tions, programmes, and projects; 3.5. Caste discrimination in garment factories and spinning mills in • Be transparent about the timelines and substantive requirements imposed upon member companies; Tamil Nadu • Require member companies to map their supply chain and publicly disclose supply chain information and 4. Normative framework - fun- damental labour standards HRDD processes, in line with the requirements formulated above. Mapping and disclosure could be a phased 4.1. International guidelines for process, with disclosure to the initiative’s secretariat and aggregated public disclosure as an intermediate step. responsible business conduct • Publish the assessment of member companies, as well as the assessment framework and criteria used; 4.2. Fundamental labour standards 4.3. Forced labour indicators • Publish lists of member brands that source in specific countries or regions. 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 9.2 Government actors 4.6. Caste-based discrimination legislation in India To the governments of countries where brands and retailers are headquartered: 5. Evidence of existence and • Actively enforce existing legal instruments with regard to respect for human rights in international supply risk of forced labour chains, such as the UK Modern Slavery Act (2015), the French Corporate Duty of Vigilance Law (2017), the 5.1. Characteristics of the 29 Dutch Child Labour (Duty of Care) Act (‘Wet Zorgplicht Kinderarbeid’) (2019); and the German HRDD law investigated mills (2021); 5.2. Respondents’ profiles 5.3. Risks and evidence of forced • Develop and adopt HRDD legislation – at a national level and in EU and international settings - that obliges labour – using the 11 ILO companies, to conduct HRDD in line with the OECD Guidelines; that can be applied to companies of all sizes indicators 6. Impact of COVID-19 and types, including Small and Medium Enterprises (SMEs); and that covers the full supply chain; 6.1 Loss of income • Support, also with financial means, credible national, international and/or sectoral RBIs/MSIs improvement 6.2 Emergency relief 6.3 After lockdown initiatives that aim to improve working conditions in the global garment and textile industry, allowing for a 6.4 Safety measures inadequate ‘smart mix’ of binding regulations and improvement initiatives.Credible means in line with all aspects of the 7. Responses to the research UNGP/OECD Guidelines and Guidance; 7.1 Supply chain links 7.2 Links established based on • Develop national and/or EU-level regulation to address unfair trading practices UTPs( ) in business-to-busi- corporate disclosure ness relationships in the garment and textile supply chain, similar to the EU directive on business-to-business 7.3 Links with vertically integra- relationships in the agricultural and food supply chain, adopted in 2019156; ted manufacturers 7.4 Company responses with • In the context of political and trade relations with India, both bilateral and in EU/international settings, raise regard to the draft research the issue of forced labour in the south Indian textile industry; findings 8. Conclusions: Beyond 29 mills • EU and EU member states amend EU custom regulations to allow further public disclosure of imports of – a sector risk analysis yarns, fabrics, apparel items and household textiles, with a minimum requirement that the following be dis- 8.1 Incidence and risks of forced closed: name and address of manufacturers and importers, HS code, product type/description, volume, FoB labour 8.2 Lack of publicly available data and value of goods; supply chain transparency • Ensure that only companies that apply enhanced HRDD and operate fully in line with OECD Guidelines and 8.3 Flawed business model 8.4 Restricted interpretation of Guidance and the UNGPs have access to governmental support and contracts. Exclude companies with an supply chain responsibility insufficient HRDD track record, from public procurement, official trade delegations, state aid, etc. 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 89 research

Executive summary

1. Introduction 1.1. Why this research? To Indian government authorities, at national and state level: 1.2. Reading guide • Urgently ratify the relevant ILO Conventions, in particular the ILO Conventions on Freedom of Association 2. Methodology 2.1. Selection of mills and Collective Bargaining, Child Labour, and the Protocol to the Forced Labour Convention, and transpose 2.2. Field research these conventions into national law; 2.3. Supply chain research 2.4. Company review • Develop and enact HRDD legislation; 2.5. Methodological challenges • Ensure that only companies that apply enhanced HRDD, and operate fully in line with OECD Guidelines and 3. Context Guidance and the UNGPs, have access to government support and contracts. Exclude companies with an 3.1. India’s cotton, textile, and garment industry insufficient HRDD track record from public procurement, official trade delegations, state aid, etc. 3.2. The Tamil Nadu textile and • Enforce all central and state laws and regulations that protect the rights and interests of textile industry garment industry 3.3. Trade unions workers in Tamil Nadu, in particular intrastate and interstate migrant workers, giving special regard to those in 3.4. Social exclusion based on vulnerable worker categories such as young women, and people with a Dalit or tribal background, and migrant caste and tribal background and hostel workers. 3.5. Caste discrimination in garment factories and spinning mills in • Develop and pass legislation on a minimum wage for all types of workers, working under all types of con- Tamil Nadu tracts, in the garment and textile industry; 4. Normative framework - fun- damental labour standards • Take measures to ensure that the rights of migrant workers are protected, especially those in vulnerable 4.1. International guidelines for worker categories such as young women, and people with a Dalit or tribal background. These measures would responsible business conduct include: regulating and monitoring government-sponsored and other recruiters; providing workers and their 4.2. Fundamental labour standards 4.3. Forced labour indicators families and communities with adequate information about their rights and what they should realistically 4.4. Legislative developments and expect of their new workplaces; ensuring that job opportunities provided under government-funded skills existing supply chain legislation training programmes meet Indian labour laws and international criteria for decent work; 4.5. Labour laws in India and Tamil Nadu • Ensure judicial grievance mechanisms are in place for workers; 4.6. Caste-based discrimination • Support and protect civic space, in particular the civil society organisations, labour movement, grassroots -or legislation in India 5. Evidence of existence and ganisations, and human rights defenders in Tamil Nadu, that look into, and report on, the plight of vulnerable risk of forced labour workers and the abuses they suffer in the garment and textile sector; 5.1. Characteristics of the 29 investigated mills • Improve the accessibility and credibility of existing judicial and non-judicial grievance redressal mechanisms 5.2. Respondents’ profiles for workers. 5.3. Risks and evidence of forced labour – using the 11 ILO indicators Specific points of attention in relation to fighting discrimination on the basis of caste and tribal 6. Impact of COVID-19 background: 6.1 Loss of income All stakeholders that have a role and a stake in the Indian textile sector – manufacturers, buyers, MSIs 6.2 Emergency relief 6.3 After lockdown and other improvement initiatives, business associations, Indian and international civil society organi- 6.4 Safety measures inadequate sations, government representatives - need to acknowledge that discrimination on the basis of caste 7. Responses to the research and tribal background is a highly problematic issue that needs to be urgently addressed. There are 7.1 Supply chain links many ways of doing this. Suggested ways forward are: 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- • Include discrimination on the basis of caste and tribal background, in corporate two-way codes of ted manufacturers conduct and monitoring programmes. There are various helpful tools that companies, MSIs and 7.4 Company responses with RBC initiatives can use to help with this, such as the Dalit discrimination check (IDSN, 2008), ETI regard to the draft research Guidance on Caste in Global Supply Chains (ETI, 2019), and the Ambedkar Principles and Guide- findings 8. Conclusions: Beyond 29 mills lines to address Caste Discrimination in the Private Sector (ISDN, 2009); – a sector risk analysis • Ensure workers with a tribal or low-caste background, including Dalits, are aware of their rights; 8.1 Incidence and risks of forced Educate all workers, supervisors, and management on the root causes and manifestations of caste labour and tribal discrimination; 8.2 Lack of publicly available data and supply chain transparency • Work towards a balanced representation in supervisor and management roles of people with dif- 8.3 Flawed business model ferent caste, tribal, gender, and social backgrounds; 8.4 Restricted interpretation of • The OECD should include discrimination on the basis of caste, and other similar forms of discrim- supply chain responsibility ination, in the OECD Guidelines for Multinationals; 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 90 research

Executive summary

1. Introduction 1.1. Why this research? • The problem of discrimination based on caste and tribal background in the garment and textile 1.2. Reading guide industry should be the subject of discussions between India and the countries where international 2. Methodology 2.1. Selection of mills buying companies are headquartered. These discussions should take place in the context of overall 2.2. Field research trade relations, as an important aspect of the business and human rights agenda; 2.3. Supply chain research • EU Member states should work with the European Commission to create an EU-wide policy on 2.4. Company review Dalits and Business and Human Rights, both as part of the wider policy of raising the issue of caste 2.5. Methodological challenges discrimination in relationships with India, and as part of its business and human rights policies. 3. Context 3.1. India’s cotton, textile, and garment industry 3.2. The Tamil Nadu textile and garment industry 9.3 Need for further independent research 3.3. Trade unions While we are proud of this research, we also acknowledge that a number of topics have not received the attention 3.4. Social exclusion based on caste and tribal background they deserve. More research is required, undertaken by independent parties, to establish, in particular, the corre- 3.5. Caste discrimination in garment lation of various factors such as caste and tribal background, poverty, labour migration, wages, housing conditions, factories and spinning mills in conditions on the work floor, etc. It is, in this context, that we call upon government and public funders to make Tamil Nadu this possible, and, at the same time we ask that corporate actors collaborate with follow-up research. 4. Normative framework - fun- damental labour standards 4.1. International guidelines for responsible business conduct 4.2. Fundamental labour standards 4.3. Forced labour indicators 4.4. Legislative developments and existing supply chain legislation 4.5. Labour laws in India and Tamil Nadu 4.6. Caste-based discrimination legislation in India 5. Evidence of existence and risk of forced labour 5.1. Characteristics of the 29 investigated mills 5.2. Respondents’ profiles 5.3. Risks and evidence of forced labour – using the 11 ILO indicators 6. Impact of COVID-19 6.1 Loss of income 6.2 Emergency relief 6.3 After lockdown 6.4 Safety measures inadequate 7. Responses to the research 7.1 Supply chain links 7.2 Links established based on corporate disclosure 7.3 Links with vertically integra- ted manufacturers 7.4 Company responses with regard to the draft research findings 8. Conclusions: Beyond 29 mills – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 91 research

Executive summary

1. Introduction 1.1. Why this research? case of the Tamil Nadu textile and garment industry’, March 2013 1.2. Reading guide Endnotes and SOMO and 2. Methodology 1. We cannot mention the name of this anthropologist, as we fear ICN, ‘Case closed, problems persist - Grievance mechanisms of ETI 2.1. Selection of mills being linked to this research might cause him to have problems and SAI fail to benefit young women and girls in the South Indian textile industry’, June 2018 https://www.somo.nl/case-closed- 2.2. Field research when applying for a visa for India. problems-persist/ 2.3. Supply chain research 2. See for instance: SOMO and ICN, ‘Captured by Cotton - Exploit- ed Dalit girls produce garments in India for European and US 9. Government of Tamil Nadu, Handlooms, Handicrafts, Textiles & 2.4. Company review markets’, May 2011 https://www.somo.nl/nl/captured-by-cotton/; Khadi Department, ‘Tamil Nadu New Integrated Textile Policy 2.5. Methodological challenges SOMO and ICN, ‘Maid in India - Young Dalit Women Continue 2019’ http://www.spc.tn.gov.in/policy_doc/TN_Textile_Poli- 3. Context to Suffer Exploitative Conditions in India’s Garment Industry’, cy_2019.pdf 3.1. India’s cotton, textile, and April 2012 https://www.somo.nl/nl/maid-in-india/; SOMO and 10. For more details on the different types of mills, see chapter 3. garment industry ICN, ‘Flawed Fabrics - The abuse of girls and women workers in 11. We cannot mention the name of this anthropologist, as we fear 3.2. The Tamil Nadu textile and the South Indian textile industry’, October 2014 https://www. being linked to this research might cause him problems when garment industry somo.nl/flawed-fabrics/?noredirect=en_GB and; ICN, ‘Fabric of applying for a visa for India. 3.3. Trade unions Slavery – Large-scale forced (child) labour in South India’s spinning 12. Open Apparel Registry website: https://openapparel.org/ 13. A paid database which includes detailed shipment information for 3.4. Social exclusion based on mills’, December 2016 https://arisa.nl/wp-content/uploads/Fab- a limited number of countries (amongst others including India and caste and tribal background ricOfSlavery.pdf 3. See for instance: SOMO and ICN, ‘Time for Transparency - The the US) https://panjiva.com/ (subscription based). 3.5. Caste discrimination in garment case of the Tamil Nadu textile and garment industry’, March 2013 14. A paid database which includes detailed shipment information for factories and spinning mills in https://www.somo.nl/time-for-transparency/ and SOMO and ICN, a limited number of countries (amongst others including India and Tamil Nadu ‘Case closed, problems persist - Grievance mechanisms of ETI the US) https://panjiva.com/ 4. Normative framework - fun- and SAI fail to benefit young women and girls in the South Indian 15. The Ministry of Textiles, Government of India, ‘Annual Report damental labour standards textile industry’, June 2018 https://www.somo.nl/case-closed- 2019-2020’ http://texmin.nic.in/sites/default/files/AR_MoT_2019- 4.1. International guidelines for problems-persist/ 20_English.pdf responsible business conduct 4. See for instance: 16. Ibid. 4.2. Fundamental labour standards * Anti-Slavery International, ‘Slavery on the high street – Forced 17. Fair Wear Foundation, ‘Country Study India 2019’ https://api.fair- wear.org/wp-content/uploads/2019/06/CS-INDIA-2019.pdf 4.3. Forced labour indicators labour in the manufacture of garments for international brands’, 18. Government of Tamil Nadu, Handlooms, Handicrafts, Textiles & 4.4. Legislative developments and June 2012; Khadi Department, ‘Tamil Nadu New Integrated Textile Policy existing supply chain legislation * Solidaridad-South & South East Asia & Fair Labor Association, ‘Understanding the Characteristics of the Sumangali Scheme 2019’ http://www.spc.tn.gov.in/policy_doc/TN_Textile_Poli- 4.5. Labour laws in India and in Tamil Nadu Textile & Garment Industry and Supply Chain cy_2019.pdf Tamil Nadu Linkages’, May 2012; 19. Ibid. 4.6. Caste-based discrimination * Homeworkers Worldwide, ‘Forced Labour in Tamil Nadu’s textile 20. Ibid. legislation in India and garment industry:An investigation into the working condi- 21. Ibid. 5. Evidence of existence and tions of young women making clothes for the UK high street’, 22. bid. risk of forced labour November 2014; 23. Fair Wear Foundation, ‘Country Study India 2019’ < https://api. 5.1. Characteristics of the 29 * Fair Wear Foundation, ‘FWF Guidance for members: The Suman- fairwear.org/wp-content/uploads/2019/06/CS-INDIA-2019.pdf> 24. Verité, ‘Help Wanted: Hiring, Human Trafficking and Modern-Day investigated mills gali Scheme and India’s bonded labour system’, March 2015; Slavery in the Global Economy’, 2010 https://www.verite.org/ 5.2. Respondents’ profiles * Burns, D. et al., Institute of Development Studies, ‘Patterns and wp-content/uploads/2016/11/Help_Wanted_2010.pdf. 5.3. Risks and evidence of forced Dynamics of Bonded Labour and Child Labour in the Spinning Mills of Tamil Nadu: Findings From Life Story Analysis’, Septem- 25. ITUC, ‘2020 ITUC Global Rights Index – The World’s Worst Coun- labour – using the 11 ILO ber 2016; tries for Workers’, < https://www.ituc-csi.org/IMG/pdf/ituc_global- indicators * International Labour Organization, ‘Working Conditions of Mi- rightsindex_2020_en.pdf> 6. Impact of COVID-19 grant Garment Workers in India – A literature review, 2017’; 26. Ibid. 6.1 Loss of income * Partners in Change & READ, ‘Business reinvent servitude - Un- 27. Human Right Watch, ‘Caste discrimination: A global concern’, 6.2 Emergency relief derstanding The Status Of Female Migrant Labour From Odisha chapter 4, 6.4 Safety measures inadequate * Fashion Revolution & Tamil Nadu Declaration, ‘Out of Sight: a 28. International Dalit Solidarity Network, ‘Caste discrimination’, 7. Responses to the research call for transparency from field to fabric, October 2020’. 29. Although in India manual scavenging is prohibited by law through 7.1 Supply chain links 5. The Indian counterparts and interview team that we worked with the Prohibition of Employment as Manual Scavengers and Their 7.2 Links established based on cannot be mentioned by name for security reasons. 6. Although the caste system, caste-based classification and caste Rehabilitation Act, 2013 , IDSN estimates 7.3 Links with vertically integra- law in India, these practices are unfortunately still wide-spread. that about 1,3 million Dalits are still depending on manual scav- ted manufacturers Arisa and SOMO use the caste framework reluctantly, only to enging for their livelihood . 7.4 Company responses with reveal what is truly going on in society, not to revive an outdated 30. Ethical Trading Initiative, ‘Caste in global supply chains’, page 4, regard to the draft research and rightfully abandoned societal hierarchy. 31. Department of Backward Classes, Most Backward Classes & – a sector risk analysis May 2011 < https://www.somo.nl/nl/captured-by-cotton/>; Minorities Welfare, ‘About us’, labour to Suffer Exploitative Conditions in India’s Garment Industry’, April 2012 ; SOMO and 32. 31 Goi Monitor, ‘History of reservation in India’, supply chain transparency the South Indian textile industry’, October 2014 and; ICN, ‘Fabric of 104,545,716 Scheduled Tribes. Source: supply chain responsibility mills’, December 2016 International Conference Series 4: Examining Stigmatization of 8.6 Frontrunners bear the brunt 8. See for instance: SOMO and ICN, ‘Time for Transparency - The Leather Industry: By focusing on the Labor Forms of Dalits and 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 92 research

Executive summary

1. Introduction 1.1. Why this research? Baraku’, page 17-19, Tokyo University of Foreign Studies (TUFS), 8f04efb208f10/1591193532010/At+risk+of+Forced+labour+%- 1.2. Reading guide http://www.tufs.ac.jp/ts/society/findas/wordpress/wp-content/ 28download%29.pdf> 2. Methodology uploads/2020/03/2019-FINDAS-International-WS-proceed- 54. European Parliament, ‘European Parliament resolution of 10 March 2.1. Selection of mills ings-dalits.pdf 2021 with recommendations to the Commission on corporate 35. Ethical Trading Initiative, ‘Caste in global supply chains’, https:// due diligence and corporate accountability (2020/2129(INL))’, 10 2.2. Field research www.ethicaltrade.org/sites/default/files/shared_resources/ March 2021 https://www.europarl.europa.eu/doceo/document/ 2.3. Supply chain research ETI%20Base%20Code%20guidance%2C%20caste%20in%20glob- TA-9-2021-0073_EN.html 2.4. Company review al%20supply%20chains.pdf 55. Thomson Reuters Foundation News, ‘Germany to fine companies 2.5. Methodological challenges 36. Toshie Awaya, Tsutomu Tomotsune and Maya Suzuki, ‘FINDAS whose suppliers abuse rights’, 12 February 2021 3.1. India’s cotton, textile, and Leather Industry: By focusing on the Labor Forms of Dalits and 56. SOMO, ‘The Next Step for Corporate Accountability in the Neth- garment industry Baraku’, page 18-20, Tokyo University of Foreign Studies (TUFS), erlands: The New Bill for Responsible and Sustainable Interna- 3.2. The Tamil Nadu textile and http://www.tufs.ac.jp/ts/society/findas/wordpress/wp-content/ tional Business Conduct’, 18 March 2021. https://www.somo.nl/ garment industry uploads/2020/03/2019-FINDAS-International-WS-proceed- the-next-step-for-corporate-accountability-in-the-netherlands/ 3.3. Trade unions ings-dalits.pdf 57. Law number 2017-399 from 27 March 2017 concerning the duty 37. Human Rights Advocacy and Research Foundation, ‘30 year Tamil of vigilance of parent companies and ordering companies. 3.4. Social exclusion based on Nadu scorecard’, page 43-44, http://hrf.net.in/2020poa/ 58. European Coalition for Corporate Justice, ‘French Corporate Duty caste and tribal background 38. International Dalit Solidarity Network, ‘Caste, Gender and Forced Of Vigilance Law – Frequently Asked Questions’, March 2017. 3.5. Caste discrimination in garment & Bonded Labour’, http://idsn.org/wp-content/uploads/2015/06/ 59. In May 2019, the Dutch Senate voted to adopt this law. Several factories and spinning mills in Caste-Gender-and-Modern-Slavery1.pdf aspects of interpretation and especially implementation of the law Tamil Nadu 39. Fair Wear Foundation, ‘India country study’, page 39-40, https:// are still to be determined via an instrument known as a General 4. Normative framework - fun- api.fairwear.org/wp-content/uploads/2019/06/CS-INDIA-2019. Administrative Order or GAO. Source: MVO Platform website, damental labour standards pdf “Update: Frequently Asked Questions about the new Dutch Child 4.1. International guidelines for 40. Fair Wear Foundation, ‘India country study’, page 29, https://api. Labour Due Diligence Law”, 3 June 2019 https://www.mvoplat- responsible business conduct fairwear.org/wp-content/uploads/2019/06/CS-INDIA-2019.pdf form.nl/en/frequently-asked-questions-about-the-new-dutch- 4.2. Fundamental labour standards 41. Fair Wear Foundation, ‘India country study’, page 18 https://api. child-labour-due-diligence-law/ fairwear.org/wp-content/uploads/2019/06/CS-INDIA-2019.pdf 60. MVO Platform, ‘Update: Frequently Asked Questions about the 4.3. Forced labour indicators 42. Fair Wear Foundation, ‘India country study’, page 40, https://api. new Dutch Child Labour Due Diligence Law’, 3 June 2019, < 4.4. Legislative developments and fairwear.org/wp-content/uploads/2019/06/CS-INDIA-2019.pdf https://www.mvoplatform.nl/en/frequently-asked-questions- existing supply chain legislation 43. Interview with caste discrimination expert of a Tamil Nadu based about-the-new-dutch-child-labour-due-diligence-law/> 4.5. Labour laws in India and civil society organisation, November 25 and 26, 2019. 61. Legislation.gov.uk, ‘UK Modern Slavery Act 2015, Provision 54, Tamil Nadu 44. The United Nations Human Rights Office of the High Com- Transarency in Supply Chains’, http://www.legislation.gov.uk/ 4.6. Caste-based discrimination missioner,‘Guiding principles on business and human rights. ukpga/2015/30/section/54/enacted legislation in India Implementing the United Nations “Protect, tesepect and Remedy” 62. Business and Human Rights in Law, ‘Key developments – Australia’, 5. Evidence of existence and Framework’, https://www.ohchr.org/documents/publications/guid- http://www.bhrinlaw.org/key-developments (accessed on 15 April risk of forced labour ingprinciplesbusinesshr_en.pdf 2021). 5.1. Characteristics of the 29 45. OECD Responsible business conduct, ‘OECD Guidelines for Multi- 63. State of California, Department of Justice, ‘The California Transpar- national Enterprises. 2011 edition’, http://mneguidelines.oecd.org/ ency in Supply Chains Act’, https://oag.ca.gov/SB657 (accessed on investigated mills mneguidelines/ 15 April 2021). 5.2. Respondents’ profiles 46. OECD Responsible business conduct, ‘OECD Due Diligence 64. International Labour Organisation, ‘The child labour (prohibition 5.3. Risks and evidence of forced Guidance for Responsible Business Conduct’, http://mneguidelines. and regulation) act, 2016’, 29 July 2016, https://www.ilo.org/dyn/ labour – using the 11 ILO oecd.org/due-diligence-guidance-for-responsible-business-con- natlex/docs/ELECTRONIC/105989/129748/F-2107857447/ indicators duct.htm IND105989.pdf 6. Impact of COVID-19 47. OECD Responsible business conduct, ‘OECD Due Diligence Guid- 65. Government of India, ‘The Constitution of India’, 1949, https:// 6.1 Loss of income ance for responsible Supply Chains in the Harment and Footwear www.india.gov.in/my-government/constitution-india 6.2 Emergency relief Sector’, https://mneguidelines.oecd.org/oecd-due-diligence-guid- 66. Government of India, ‘Bonded Labour System (Abolition) Act’, 6.3 After lockdown ance-garment-footwear.pdf 1976, https://pblabour.gov.in/Content/documents/pdf/rti/ 6.4 Safety measures inadequate 48. International Labour Organisation, ‘Conventions and Recommen- rti_chapter18.pdf, https://labour.gov.in/sites/default/files/The- dations’, https://www.ilo.org/global/standards/introduction-to-in- BondedLabourSystem(Abolition)Act1976.pdf 7. Responses to the research ternational-labour-standards/conventions-and-recommendations/ 67. Traidcraft, ‘An exploratory study into working conditions in the 7.1 Supply chain links lang--en/index.htm textile & apparel sector in the National Capital Region, India’, June 7.2 Links established based on 49. International Labour Organization, ‘NORMLEX Ratifications for 2020 https://static1.squarespace.com/static/59242ebc03596e- corporate disclosure India’ https://www.ilo.org/dyn/normlex/en/f?p=NORMLEX- 804886c7f4/t/5ed7afb58fb8f04efb208f10/1591193532010/ 7.3 Links with vertically integra- PUB:11200:0::NO::P11200_COUNTRY_ID:102691 (accessed on At+risk+of+Forced+labour+%28download%29.pdf ted manufacturers 13 November 2020). 68. Exchange rate of 2 November 2020 https://www.xe.com 7.4 Company responses with 50. International Labour Organization, ‘Declaration on Fundamental 69. Factories Act 1948, https://www.indiacode.nic.in/han- regard to the draft research Principles and Rights at Work (1998)’ https://www.ilo.org/declara- dle/123456789/1530?view_type=browse&sam_han- findings tion/thedeclaration/textdeclaration/lang--en/index.htm> dle=123456789/1362 8. Conclusions: Beyond 29 mills 51. International Dalit Solidarity Network, ‘United Nations: ILO’, 70. Government of India, ‘Inter-state migrant workmen (regulation of https://idsn.org/key-issues/caste-business/ilo-2/> employment and conditions of service) Act, 1979’. https://clc.gov. – a sector risk analysis 52. International Labour Organisation, ‘ILO indicators of Forced in/clc/acts-rules/inter-state-migrant-workmen 8.1 Incidence and risks of forced Labour’ https://www.ilo.org/global/topics/forced-labour/publica- 71. Factories Act 1948 https://www.indiacode.nic.in/han- labour tions/WCMS_203832/lang--en/index.htm > https://www.ilo.org/ dle/123456789/1530?view_type=browse&sam_han- 8.2 Lack of publicly available data and global/topics/forced-labour/publications/WCMS_203832/lang- dle=123456789/1362 supply chain transparency -en/index.htm#:~:text=These%20indicators%20are%20intend- 72. Government of India, ‘THE EMPLOYEES’ STATE INSURANCE ACT, 8.3 Flawed business model ed%20to,who%20may%20require%20urgent%20assistance. 1948’, https://www.esic.nic.in/attachments/actfile/ae036213fe- 8.4 Restricted interpretation of 53. ‘At risk of forced labour. An exploratory study into working c2e5ca97e2c9314a1fc9e9.pdf supply chain responsibility conditions in the textile & apparel sector in the Nation- 73. Fair Wear Foundation, ‘Country Study India 2019’, https://api. 8.5 Reliance on social auditing al Capital Region, India’, https://static1.squarespace.com/ fairwear.org/wp-content/uploads/2019/06/CS-INDIA-2019.pdf 8.6 Frontrunners bear the brunt static/59242ebc03596e804886c7f4/t/5ed7afb58fb- 74. The employees‘provident fund organisation (EPFO), ‘The 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 93 research

Executive summary

1. Introduction 1.1. Why this research? employees‘provident funds and miscellaneous provisions act, markets’, May 2011 https://www.somo.nl/nl/captured-by-cotton/; 1.2. Reading guide 1952’, to Suffer Exploitative Conditions in India’s Garment Industry’, 2.1. Selection of mills 75. https://www.indiacode.nic.in/handle/123456789/2104?view_ April 2012 https://www.somo.nl/nl/maid-in-india/ ;SOMO and type=browse&sam_handle=123456789/1362 ICN, ‘Flawed Fabrics - The abuse of girls and women workers in 2.2. Field research 76. Fair Wear Foundation, ‘India Country Study 2019’, < https://api. the South Indian textile industry’, October 2014 https://www. 2.3. Supply chain research fairwear.org/wp-content/uploads/2019/06/CS-INDIA-2019.pdf> somo.nl/flawed-fabrics/?noredirect=en_GB and; ICN, ‘Fabric of 2.4. Company review 77. Laws of India, ‘The Tamil Nadu Hostels and Homes for Women and Slavery – Large-scale forced (child) labour in South India’s spinning 2.5. Methodological challenges Children (Regulation) Act 2014’, ricOfSlavery.pdf 3.1. India’s cotton, textile, and 78. Government of India, ‘The Constitution of India’, 1949 worker did not indicate to work night shifts. The other respon- 3.2. The Tamil Nadu textile and 79. Fair Wear foundation, ‘India country study 2019’, page 52, 102. This question was asked to all 725 workers, at the 29 mills we 80. Fair Wear Foundation, ‘India Country Study 2019’, < https://api. investigated. Workers of 21 mills indicated that the youngest 3.4. Social exclusion based on fairwear.org/wp-content/uploads/2019/06/CS-INDIA-2019.pdf> workers in their mill were 14 years of age. This is very worrying caste and tribal background 81. Industrial Disputes (central) Rules 1957, sections 38-57 signal. To nuance this picture, we may need to add that in the case 3.5. Caste discrimination in garment 82. 81 Fair Wear Foundation, ‘Worker-management dialogue in of seven of these 21 mills, there were less than five respondents factories and spinning mills in Indian legislation – a guidance document’, industry deliver decent jobs for workers in Myanmar?’, February damental labour standards 83. Ibid. 2017, https://www.somo.nl/the-myanmar-dilemma/ 4.1. International guidelines for 84. Government of India, https://www.india.gov.in/sites/upload_files/ 104. Note that 54 workers did not answer the question about pay slips. responsible business conduct npi/files/coi_part_full.pdf 105. International Labour Organization, ‘ILO Indicators of Forced La- 4.2. Fundamental labour standards 85. International Dalit Solidarity Network, ‘Dalit Discrimination Check’, bour’, October 2012, https://www.ilo.org/global/topics/forced-la- 2008 https://idsn.org/wp-content/uploads/user_folder/pdf/ bour/publications/WCMS_203832/lang--en/index.htm 4.3. Forced labour indicators New_files/CSR/Dalit_Discrimination_Check.pdf 106. Noted by Dutch anthropologist who volunteered for SOMO, based 4.4. Legislative developments and 86. Ibid. on conversations with workers and recruiters. existing supply chain legislation 87. Ethical Trading Initiative, ‘Base Code Guidance: Caste in Global 107. One worker at mill 12 did not answer the question about annual 4.5. Labour laws in India and Supply Chains’, 2019, https://www.ethicaltrade.org/sites/ leave. Tamil Nadu default/files/shared_resources/ETI%20Base%20Code%20guid- 108. Ibid. 4.6. Caste-based discrimination ance%2C%20caste%20in%20global%20supply%20chains.pdf 109. Ibid. legislation in India 88. International Dalit Solidarity Network, ‘The Ambedkar Principles 110. Ibid. 5. Evidence of existence and – Principles and Guidelines to address Caste Discrimination in the 111. Ibid. risk of forced labour Private Sector’, September 2009 https://idsn.org/wp-content/up- 112. Workers from mills 4, 22, 23, 24, 26, 28, 29, 30 did not think they 5.1. Characteristics of the 29 loads/user_folder/pdf/New_files/IDSN/Ambedkar_Principles_bro- would be penalised if they quit their job. chure.pdf 113. Reported by the Dutch anthropologist who volunteered for investigated mills 89. Better Cotton Initiative, ‘homepage’, https://bettercotton.org/ SOMO, on the basis of conversations with one of the Tamil Nadu 5.2. Respondents’ profiles (accessed on 18 March 2021). labour rights organisations that assisted in the field research. 5.3. Risks and evidence of forced 90. Global Organic Textile Standard, ‘homepage’, https://global-stan- 114. Ibid. labour – using the 11 ILO dard.org/ (accessed on 18 March 2021). 115. Ibid. indicators 91. Global Organic Textile Standard, ‘Third-party certification’, https:// 116. Ibid. 6. Impact of COVID-19 global-standard.org/the-standard/gots-key-features/third-par- 117. Ibid. 6.1 Loss of income ty-certification 118. Wage Indicator, ‘Living wage data’, 2020, https://wageindicator. 6.2 Emergency relief 92. Global Organic Textile Standard, ‘GOTS response to SOMO/ Arisa org/salary/living-wage/list-of-country-region-living-wages-da- 6.3 After lockdown report’, 12 February 2021. ta-availability 6.4 Safety measures inadequate 93. Global Organic Textile Standard, ‘Complaints procedure’, https:// 119. All 25 interviewed workers at mill 24. global-standard.org/the-standard/protection/complaint-procedure 120. International Labour Organization, ‘ILO Indicators of Forced La- 7. Responses to the research 94. Global Organic Textile Standard, ‘GOTS detects evidence of bour’, October 2012 https://www.ilo.org/global/topics/forced-la- 7.1 Supply chain links Organic Cotton Fraud in India’, 30 October 2020, https://www. bour/publications/WCMS_203832/lang--en/index.htm 7.2 Links established based on global-standard.org/news/gots-press-release-gots-detects-evi- 121. We interviewed 25 workers in mills 1, 2 and 6. In each of these corporate disclosure dence-of-organic-cotton-fraud-in-india three mills, five workers spoke of 7-day week schedules. 7.3 Links with vertically integra- 95. Government of India, Ministry of Social Justice and Empower- 122. Dutch Agreement on Sustainable Garments and Textile, ‘signato- ted manufacturers ment, Department of Social Justice and Empowerment, ‘List of ries’, https://www.imvoconvenanten.nl/en/garments-textile/signa- 7.4 Company responses with Scheduled Castes’, http://socialjustice.nic.in/UserView/index- tories (accessed on 12 March 2021). regard to the draft research ?mid=76750, ‘State/UT-wise Number of Entries in the Central List 123. In their third year of membership, AGT members have to have findings of OBCs (as on 12.04.2018)’, http://socialjustice.nic.in/UserView/ mapped and disclosed suppliers beyond the first tier. 8. Conclusions: Beyond 29 mills index?mid=76674 and Ministry of Tribal Affairs, ‘State/Union 124. Dutch Agreement on Sustainable Garments and Textile, ‘Projects’, Territory-wise list of Scheduled Tribes in India’, https://tribal.nic.in/ https://www.imvoconvenanten.nl/en/garments-textile/agreement/ – a sector risk analysis ST/LatestListofScheduledtribes.pdf projects (accessed on 18 March 2021). 8.1 Incidence and risks of forced 96. International Labour Organization, ‘ILO Indicators of Forced La- 125. Dutch Agreement on Sustainable Garments and Textile, ‘Improving labour bour’, October 2012 https://www.ilo.org/global/topics/forced-la- working conditions in Tamil Nadu through cooperation’, https:// 8.2 Lack of publicly available data and bour/publications/WCMS_203832/lang--en/index.htm www.imvoconvenanten.nl/en/garments-textile/agreement/proj- supply chain transparency 97. Ibid. ects/tamil-nadu (accessed on 18 March 2021). 8.3 Flawed business model 98. Fair Wear Foundation, ‘India Country Study 2019’, https://api. 126. Fair Wear Foundation, ‘Get to know fair wear’, https://www. 8.4 Restricted interpretation of fairwear.org/wp-content/uploads/2019/06/CS-INDIA-2019.pdf fairwear.org/about-us/get-to-know-fair-wear (accessed on 12 supply chain responsibility 99. Workers of mill 28, interviewed by Dutch volunteer. March 2021). 8.5 Reliance on social auditing 100. See for instance: SOMO and ICN, ‘Captured by Cotton - Exploit- 127. Fair Wear Foundation, ‘Brands’, https://www.fairwear.org/pro- 8.6 Frontrunners bear the brunt ed Dalit girls produce garments in India for European and US grammes/brands (accessed on 12 March 2021). 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 94 research

Executive summary

1. Introduction 1.1. Why this research? 128. Fair Wear Foundation, ‘Get to know fair wear’, https://www. 1.2. Reading guide fairwear.org/about-us/get-to-know-fair-wear (accessed on 12 2. Methodology March 2021). 2.1. Selection of mills 129. Ethical Trading Initiative, ‘About ETI’, https://www.ethicaltrade.org/ about-eti (accessed on 12 March 2021). 2.2. Field research 130. Ethical Trading Initiative, ‘Our member’, https://www.ethicaltrade. 2.3. Supply chain research org/about-eti/our-members (accessed on 18 March 2021). 2.4. Company review 131. Ethical Trading Initiative, ‘What we do’, https://www.ethicaltrade. 2.5. Methodological challenges org/about-eti/what-we-do (accessed on 12 March 2021). 3. Context 132. Fair Labor Association, ‘About US’, https://www.fairlabor.org/ 3.1. India’s cotton, textile, and about-us (accessed on 12 March 2021). garment industry 133. Partnerships for Sustainable Textiles, ‘Supply Chain Responsibility’, 3.2. The Tamil Nadu textile and https://www.textilbuendnis.com/en/der-review-prozess/ (accessed garment industry on 18 March 2021). 3.3. Trade unions 134. Partnerships for Sustainable Textiles, ‘Partnership Initiative Tamil Nadu’, https://www.textilbuendnis.com/en/detailseite-tamil-nadu/ 3.4. Social exclusion based on (accessed on 18 March 2021). caste and tribal background 135. Ikea, ‘Response to: SOMO & Arisa report on Tamil Nadu spinning 3.5. Caste discrimination in garment mill Industry’, 12 February 2021. factories and spinning mills in 136. Email from GAP to SOMO and Arisa dated 27 January 2021. Tamil Nadu 137. Email from NEXT to SOMO and Arisa dated 8 January 2021. 4. Normative framework - fun- 138. Email from Sainsbury’s to SOMO and Arisa dated 2 February 2021. damental labour standards 139. Email from Carrefour to SOMO and Arisa dated 4 February 2021. 4.1. International guidelines for 140. Conversation between Tesco and SOMO & Arisa, 23 February responsible business conduct 2012. 4.2. Fundamental labour standards 141. Email from Zeeman to SOMO and Arisa dated 10 February 2021. 142. Email from Carrefour to SOMO and Arisa dated 4 February 2021. 4.3. Forced labour indicators 143. Email from Zeeman to SOMO and Arisa dated 10 February 2021. 4.4. Legislative developments and 144. WE Fashion, “Statement WE Fashion - in rection to SOMO – Arisa existing supply chain legislation report”, 9 February 2021. 4.5. Labour laws in India and 145. Marc O’Polo described the relationship as follows: our direct Tamil Nadu business relation is with an agency that handles a supplier which is 4.6. Caste-based discrimination sourcing from the spinning mill. 24 February 2021. legislation in India 146. Ikea, ‘Response to: SOMO & Arisa report on Tamil Nadu spinning 5. Evidence of existence and mill Industry’, 12 February 2021. risk of forced labour 147. Email from NEXT to SOMO and Arisa dated 22 February 2021. 5.1. Characteristics of the 29 148. GOTS, “GOTS Response to SOMO/ ARISA Report, 12 February 2021. investigated mills 149. Transparency Pledge, ‘Why disclose?’, https://transparencypledge. 5.2. Respondents’ profiles org/why-disclose 5.3. Risks and evidence of forced 150. GoodElectronics, ‘Beyond corporate transparency. The right to labour – using the 11 ILO know in the electronics industry’, April 2020, https://goodelectro- indicators nics.org/beyond-corporate-transparency/ 6. Impact of COVID-19 151. Open Apparel Registry, ‘Facilities’, https://openapparel.org/facili- 6.1 Loss of income ties?contributors=139 6.2 Emergency relief 152. Transparancy pledge, ‘The pledge’, https://transparencypledge.org/ 6.3 After lockdown the-pledge 6.4 Safety measures inadequate 153. The Asia Floor Wage Campaign provides a methodology for setting a living wage https://asia.floorwage.org/calculating-a-living-wage/ 7. Responses to the research More information on https://cleanclothes.org/poverty-wages 7.1 Supply chain links 154. Clean Clothes Campaign, ‘Binding Agreements’, https://clean- 7.2 Links established based on clothes.org/binding-agreements corporate disclosure 155. SOMO, ‘Should I stay or should I go? Exploring the role of disen- 7.3 Links with vertically integra- gagement in human rights due diligence’, April 2016, https://www. ted manufacturers somo.nl/should-i-stay-or-should-i-go-2/ 7.4 Company responses with 156. The European Commission, ‘The Directive on Unfair Trading Pracit- regard to the draft research ces in the agricultural and food supply chain’, 2019 https://ec.euro- findings pa.eu/info/sites/default/files/food-farming-fisheries/key_policies/ 8. Conclusions: Beyond 29 mills documents/brochure-utp-directive_en.pdf – a sector risk analysis 8.1 Incidence and risks of forced labour 8.2 Lack of publicly available data and supply chain transparency 8.3 Flawed business model 8.4 Restricted interpretation of supply chain responsibility 8.5 Reliance on social auditing 8.6 Frontrunners bear the brunt 8.7 Leverage 9. Recommendations 9.1 Corporate actors 9.2 Government actors 9.3 Need for further independent SPINNING AROUND WORKERS’ RIGHTS | 95 research