Towards sustainable cocoa supply chains: Regulatory options for the EU - PDF Document

0 Views
Presentation Transcript

  1. Towards sustainable cocoa supply chains: Regulatory options for the EU

  2. 2 Towards sustainable cocoa supply chains: Regulatory options for the EU Author: Duncan Brack Photographs by Francisco Sierra, Maartje de Graaf, Tropenbos International and the following Flickr accounts: CIFOR, jbdodane, Ecosystem Services for Poverty Alleviatio and MinIsterio de Agricultura y Ganadería de Ecuador Cover photo by Maartje de Graaf Design: Constantin Nimigean June 2019 Acknowledgements Many thanks to all the individuals and organisations who contributed to the preparation of this paper, through interviews, meetings and comments on drafts; your help and input was invaluable. Special thanks to Julia Christian at Fern. While the paper was commissioned by Fern, Tropenbos International, Fairtrade International and the Fair trade Advocacy Office, the content and views expressed within, and any inadvertent errors, are solely the responsibility of the author. Fern UK, 1C Fosseway Business Centre, Stratford Road, Moreton in Marsh, GL56 9NQ, UK Fern Brussels, Rue d’Édimbourg, 26, 1050 Brussels, Belgium www.fern.org Tropenbos International Lawickse Allee 11 6701 AN Wageningen, Netherlands www.tropenbos.org The Fair Trade Advocacy Office Rue Fernand Bernier 15 1060 Brussels, Belgium www.fairtrade-advocacy.org This publication has been produced with the assistance of the Arcus Foundation, the European Union, and the UK Department for International Development. The contents of this publication are the sole responsibility of the author and can in no way be taken to reflect the views of the funders.

  3. 3 Contents Executive summary 4 1 Introduction 9 2 Background: cocoa and its impacts 10 2.1 2.2 2.3 2.4 2.5 Poverty, incomes and productivity Child labour Deforestation Governance and illegality Future challenges 11 12 13 14 16 3 Background: existing initiatives 17 3.1 3.2 3.3 3.4 Certification and standards Multi-stakeholder and producer-country initiatives Consumer-country commitments Are existing initiatives enough? 17 18 19 21 4 The need for EU regulation 23 4.1 4.2 Business, human rights and the environment The case for EU action 23 24 5 Standards for EU regulation 26 6 Producer-country-focused options 28 6.1 6.2 Option A: Bilateral agreements with cocoa-producing countries Option B: ‘Carding’ system 28 32 7 EU-market-focused options 34 7.1 7.2 Option C: Public procurement policy Option D: Regulation of sales in the EU market 34 35 8 Business-focused options 37 8.1 8.2 8.3 8.4 Option E: Competition law reform Option F: Corporate reporting for cocoa supply chains Option G: A due diligence regulation for cocoa placed on the EU market Option H: A general corporate due diligence requirement 37 39 41 45 9 Conclusions and recommendations 48 A possible package Impacts 48 50

  4. 4 Executive summary This paper is designed to inform the debate around potential European Union (EU)-wide regulation of cocoa and cocoa products, including chocolate, entering the EU market. It describes a range of options for possible action and discusses their advantages and disadvantages. Background Cocoa is an important source of income and employment for rural populations, particularly for the five to six million small-scale farmers who grow over 90 per cent of the world’s cocoa. In total, close to 50 million people are dependent on cocoa for their livelihoods. Nevertheless, most cocoa growers live in poverty, and the cocoa poverty trap has led to the widespread use of child labour. In addition, cocoa is a major driver of deforestation, particularly in Côte d’Ivoire and Ghana, which between them account for about two-thirds of global production and two-thirds of EU imports. In both countries, poor standards of governance and weak law enforcement also underlie many of the problems; at least 30–40 per cent of cocoa produced in Côte d’Ivoire is thought to be illegal. There have been many voluntary initiatives to tackle these problems. The use of certification schemes for cocoa is more common than for most agricultural commodities, a new International Organisation for Standardisation (ISO) standard for sustainable cocoa has been published, and many companies have their own programmes. Organisations and initiatives have been set up to tackle child labour and deforestation, and a number of EU Member States have announced programmes to address the sustainability of the cocoa sector. There is increasing acknowledgement, however, that while these current initiatives have had some positive impacts, they have not succeeded, and are not likely to succeed, in tackling low prices and poverty, child labour, deforestation and illegality across the whole cocoa sector. This is the position taken by many Non-Governmental Organisations (NGOs) and also, in recent months, by a number of chocolate and cocoa trading and processing companies. Certification schemes and company programmes do not cover the majority of cocoa farmers and have no direct control over some issues essential to farmers’ livelihoods, including access to infrastructure, inputs and support services. Not all companies active in the EU markets have adopted commitments or programmes to tackle the problems. And while companies may be able to improve standards where they buy cocoa beans directly from farms, this is much more difficult in the widespread indirect supply chains featuring traders and middlemen. National traceability systems are also inadequate or lacking. More broadly, issues which have major impacts on conditions in the cocoa sector such as land and forest governance and law enforcement; land-use and cocoa sector planning; the determination of national cocoa prices and annual cocoa production levels; and education policy, are all government responsibilities. This makes them difficult for external stakeholders to influence and means they cannot be addressed by certification or company schemes applying at the farm or supply chain level alone. Perhaps most importantly, systemic weaknesses in governance and law enforcement, including the prevalence of corruption, undermine many efforts to achieve sustainable production at the national level.

  5. 5 The case for and objectives of EU action The European Commission is already considering options for EU-wide action to reduce the impact of EU consumption on forests globally, potentially adapting the approach taken in the EU Forest Law Enforcement, Governance and Trade (FLEGT) Action Plan agreed in 2003. There is also growing interest in measures to limit the negative impacts of business activities on human rights, labour standards and the environment, along the lines set out in the United Nations (UN) Guiding Principles on Business and Human Rights. This has led, so far, to legislation of various types in France, the Netherlands and the United Kingdom (UK), as well as some jurisdictions outside the EU. Since the EU is the world’s largest importer of cocoa and cocoa products, EU consumption, and any standards it imposes on imports, has the potential to affect the conditions of production in the countries of origin. There is, accordingly, a strong case that EU-wide government action or regulation should be considered, but as yet there is no consensus on what it might be. This paper discusses eight possible options for EU action (and, in principle, by other consumer countries), in most cases drawing on the experience of similar measures applied to other commodities or products. Most of these options are not mutually exclusive, and various combinations are possible. All of these options are also applicable to other commodities associated with negative environmental and human rights impacts such as deforestation or forced or child labour. One of the issues that needs to be agreed is the standards that EU regulation is aiming to achieve. Four broad sets of aims cover the main negative impacts of cocoa production: 1. Respect for the laws of the producer country, including laws relating to: human rights; forced and child labour; employment conditions, such as minimum wages and health and safety conditions; rights of ownership and access to land; and environmental protection, including the protection of forests. 2. Respect for international human rights and labour law, including prohibitions against forced and child labour; the protection of decent working conditions; the right to a living wage for workers and a living income for smallholders; and the right to free association. 3. Improvements in governance and law enforcement. 4. The promotion of higher standards in cocoa production than provided in national laws, including: improved protection for rights of tenure, ownership and access to land; higher labour and living standards, such as improved incomes facilitated by guaranteed minimum prices and other interventions; and higher environmental standards, particularly relating to the protection of forests, such as a prohibition on deforestation, the maintenance of existing shade trees or requirements to plant new trees.

  6. 6 The options are grouped under three broad headings: 1. Producer-country-focused options These two options focus on actions directly aimed at influencing conditions in the cocoa-producing countries. Option A envisages the EU negotiating bilateral agreements with the main cocoa-producing countries, and providing financial and capacity-building assistance, to achieve the agreed standards for cocoa production in the producer country and improve standards of governance and law enforcement. This is complementary to almost all of the options discussed below and could form a valuable component of a broad package of measures addressing cocoa production and consumption. This option is modelled on the Voluntary Partnership Agreements (VPAs) between the EU and timber-exporting countries under the FLEGT Action Plan. VPAs are designed to ensure that all timber products exported to the EU have been legally produced – once products are FLEGT-licensed, they gain easier access to the EU market. Although it has proved difficult and time-consuming to establish the licensing systems, the process of negotiating and implementing VPAs has in some cases significantly improved governance and law enforcement, making the forest sector more transparent and accountable, and reducing illegal logging. This model could be adopted to cocoa, either based on the legality of production or on wider objectives. Option B is a ‘carding’ system through which the EU enters into dialogues with countries that export cocoa to the EU and issues yellow cards (warning) or red cards (import ban or notification of high risk) to those countries not combatting illegal behaviour in the supply chain. This is based on the EU Illegal, Unreported and Unregulated (IUU) Fishing Regulation, which has proved effective in encouraging several countries to improve standards. Cocoa, however, is produced in a much smaller number of countries, so the option of import bans is not directly applicable. However, there may be scope to examine whether the early warning element could be used in conjunction with bilateral agreements similar to VPAs. This could be further combined with a due diligence regulation discussed below, which would serve as an indicator of the level of risk associated with the producer country. 2. EU-market-focused options These two options focus on measures applied to the EU market, or to a specific segment of it, products bought by government purchasers. Under Option C, governments would use their public procurement policies to require that cocoa, or any products containing cocoa, purchased by government buyers meet minimum criteria for responsible sourcing in terms of legality and high social and environmental standards. All EU Member States are significant purchasers of food and catering services, and most already possess frameworks for sustainable procurement; many have adopted timber procurement policies to restrict buyers to legal and sustainable timber products. In practice, applying this approach to cocoa would probably mean that public purchasers would need to rely on cocoa certification schemes and company programmes; while this would be a step in the right direction (as it would help to grow the market for responsibly sourced cocoa and send a signal to the market of the government’s aims), it is unlikely to achieve major change in the cocoa sector. Option D sees this approach extended to the entire EU market, through a requirement that all cocoa and cocoa products placed on sale on the market meet minimum criteria for responsible sourcing – or, possibly, be produced legally. As with public procurement policies, this should have

  7. 7 an impact, mainly through increasing the uptake of cocoa certification schemes and company programmes for responsible or sustainable sourcing. It may, however, trigger a challenge under World Trade Organisation (WTO) trade disciplines, and is a less flexible and probably slower approach than the due diligence regulations included in Options F and G. It seems unlikely to trigger radical change in the sector. 3. Business-focused options Under Option E, EU competition law is reviewed or clarified or amended to allow businesses greater freedom to collaborate for sustainability purposes, factor in externality costs and, in particular, discuss and address low prices paid to farmers. In practice, competition law aiming to protect consumers against anti-competitive conduct by businesses can have a chilling effect on collaboration in pursuit of sustainability outcomes, particularly on issues such as low prices and farmer incomes which lie at the root of many of the cocoa sector’s problems. Competition law could be given a broader interpretation – either through clarifications issued by the competition authorities or by rewriting the legislation – to allow businesses to collaborate for long-term sustainability purposes. Option F envisages placing a requirement on enterprises placing cocoa or cocoa products on the EU market to scrutinise their supply chains for the risk of handling illegally produced cocoa, or cocoa not produced to high social and environmental standards, and to publish regular reports on the extent of the risk. There is no requirement to take action other than to publish the report, but it is assumed that this in itself would create an incentive for action because of increased transparency. This builds on several reporting initiatives, including CDP’s voluntary system for reporting on forest risk commodities, the UK Modern Slavery Act and the EU Non-Financial Reporting Directive. However, most of the large cocoa and chocolate companies already report significant amounts of information, and this has not succeeded in resolving the problems. Under Option G a requirement would be placed on all enterprises placing cocoa or cocoa products on the EU market to have in place a system of due diligence designed to minimise the risk of their handling illegally produced cocoa, or cocoa not produced to high social and environmental standards. In practice this would mean they would need to have information about the original source, an analysis of the risk of non-compliance with the agreed aims, and a strategy to mitigate the risk of non-compliance. This is modelled on the EU Timber Regulation, which has helped to exclude illegal timber from the EU market, and the Conflict Minerals Regulation. Both require companies to have due diligence systems in place. Cocoa could be included in a wider due diligence regulation covering all forest risk commodities, with its application phased by commodity. It could play a valuable role in providing an incentive to cocoa-producing countries to agree a bilateral VPA-type agreement. Option H proposes a broader regulation under which companies operating in the EU would be required to have in place a system of due diligence aimed at minimising the risk of breaches of human rights, fundamental freedoms and health and safety rights and of environmental damage in their operations and supply chains. Companies would be required to implement a due diligence plan to avoid such impacts, and to publish reports on progress. This is modelled on the French Devoir de Vigilance law of 2017, and similar proposals being put forward in other countries. It is not specific to any commodity or product, but covers operations and supply chains in general, though specific guidance could be issued for individual supply chains, like cocoa. This has advantages over the commodity-focused due diligence approach of Option G, in that it avoids the need to legislate commodity by commodity, or sector by sector, but it does represent a major change in business practices, and in practice could be slower to implement and is more likely to be restricted only to the largest companies.

  8. 8 Conclusions and recommendations As the major global consumer of cocoa from Ghana and Côte d’Ivoire, the EU has a critical role to play in improving the problems of the cocoa sector. There is, however, no simple or single solution. Rather, action is required at many levels and by many actors. None of these options in isolation is likely to succeed in encouraging the development of a sustainable cocoa supply chain which meets the aims of ensuring respect for human rights and labour standards, payment of fair prices, protection of the environment and improvements in governance and law enforcement. A package of options to be implemented by the EU and its Member States, however, could have much greater impact. Option A (bilateral agreements) would help trigger essential action by producer-country governments. Either Options G or H (due diligence regulation) would directly affect the behaviour of companies in the cocoa supply chain that operate in the EU, and could help to provide an incentive to producer-country governments to sign such an agreement. Option E (review of competition law) should help companies address the problem of low prices for producers and may also be necessary to enable the inclusion of criteria relating to prices and remuneration in a due diligence regulation. None of the options described in this paper are easy to implement; all will face challenges and barriers. But the need for action is clear. It is to be hoped that discussion on the options outlined here may contribute to efforts to find a solution, create sustainable cocoa farms and supply chains for the future, and encourage further and more detailed exploration of the many questions these options raise, and of the likely impacts on all stakeholders in the cocoa supply chain. Photo: jbdodane / Flickr

  9. 9 1 Introduction Recent years have seen a growing focus on the negative impacts of cocoa production. Although cocoa is an important source of income and employment for rural populations, most cocoa growers live in poverty, and the cocoa poverty trap has led to the widespread use of child labour. Cocoa is also a major driver of deforestation, particularly in Côte d’Ivoire and Ghana, which between them account for about two-thirds of global production. While there have been many voluntary – often company-led or company-based – initiatives to tackle these impacts, they have not been able to resolve the problems. This is a position taken by many NGOs and also, in recent months, by a number of chocolate companies and those trading and processing cocoa. Since the EU is the world’s largest importer of cocoa, any standards it imposes on imports or sales, has the potential to affect the conditions of production. There is, accordingly, a strong case that EU-wide government regulation should be considered – though as yet there is no consensus on what form such regulation should take. This paper is designed to stimulate the debate around EU regulation. It describes a range of options for possible action – mostly based on experiences with existing initiatives in other sectors – and discusses their advantages and disadvantages. Section 2 provides a brief summary of the impacts of cocoa production, while Section 3 reviews the major existing initiatives designed to tackle them, and other developments of relevance. Section 4 summarises the case for regulation at the EU level, and Section 5 discusses the appropriate standards of cocoa production this regulation should aim to meet. Sections 6, 7 and 8 describe eight possible options grouped under three broad headings: two focused on producer countries (Section 6), two on consumer markets in the EU (Section 7) and four on regulation of businesses (Section 8). Most of these are not mutually exclusive, and various combinations are possible. Section 9 draws conclusions for the optimum combination of options for the EU to adopt, and discusses briefly the possible impacts. Measures designed to influence the private and public financial institutions that invest in the cocoa sector provide another route to address the challenges; these raise many different issues, so are not covered in this report. It should also be noted that the paper refers primarily to the situation in the main West African producer countries, Côte d’Ivoire and Ghana, as both the major global producers and the countries where the main negative impacts are evident. All the options analysed here are of course applicable to other producer countries (for those which affect trade, World Trade Organisation (WTO) disciplines would require them to be applied in a non-discriminatory manner in any case). And although this paper focuses on cocoa, all the options it discusses could be applied to other commodities associated with negative impacts such as deforestation, poverty and forced or child labour.

  10. 10 2 Background: cocoa and its impacts The cocoa tree (Theobroma cacao) grows in hot, rainy climates in the tropics, mostly in a narrow band 10 degrees north and south of the equator. The cocoa tree’s fruit pods contain seeds which are extracted, fermented and dried to become cocoa beans. The beans are roasted and shelled; the resulting nibs are ground into a paste known as cocoa liquor, which is then pressed to extract cocoa butter, leaving a mass known as cocoa cake, which is ground into fine cocoa powder. While some cocoa butter is used for cosmetics, the vast bulk of cocoa liquor, butter and powder is destined for the baking and confectionery industries, particularly for the manufacture of chocolate. West Africa dominates cocoa production, accounting for four of the six biggest producer countries: Côte d’Ivoire (43 per cent of the global total in 2016/17), Ghana (20 per cent), Cameroon (5 per cent) and Nigeria (5 per cent); Ecuador (6 per cent) and Indonesia (6 per cent) were the other two.1 Global cocoa production has increased by an average of 3 per cent a year for the past two decades (though with substantial annual variations), reaching an estimated 4.7 million tonnes in 2016/17. The bulk of production is destined for export, particularly from West Africa, where domestic consumption is very small. The main export markets lie in Europe, which accounts for about 40 per cent of global consumption; next is the United States (US), with about 20 per cent.2 In 2017, two- thirds of the EU’s imports of cocoa beans, by value, originated from Côte d’Ivoire (52 per cent) and Ghana (15 per cent) together.3 Côte d’Ivoire is also now home to the world’s largest cocoa grinding (processing) industry, having recently overtaken the Netherlands as processing and manufacturing companies increasingly invest in grinding operations in producer countries. Globally, the value of the cocoa market amounted to about US$10 billion in 2016; some projections see it growing by about 3.5 per cent a year, reaching US$14.5 billion by 2026 (see Section 2.5).4 The global chocolate market, valued at US$103 billion in 2017, is currently projected to grow at about seven per cent a year, reaching US$162 billion by 2024.5 While Europe is expected to remain the major consuming region, strong growth is expected in the Asia-Pacific region, particularly in China and India (accompanying rising incomes and the spread of Western dietary habits) and in North America. Mergers and takeovers have led to a high degree of industry concentration; between them Barry Callebaut (Switzerland), Cargill (US) and Olam (Singapore) account for about 60 per cent of world cocoa processing, and Mars (US), Nestlé (Switzerland), Mondelez (US), Hershey’s (US), Ferrero (Italy) and Lindt (Switzerland) account for 40 per cent of the global consumer chocolate market.6 Cocoa is an important source of income and employment for rural populations in producer countries, particularly for the five to six million small-scale farmers who grow over 90 per cent of the world’s cocoa.7 A further 14 million rural workers directly depend on cocoa, and close to 50 million people in total are dependent on cocoa production for their livelihoods. In Ghana, cocoa 1 2 3 4 5 International Cocoa Organisation, ICCO Quarterly Bulletin of Cocoa Statistics estimates for 2016/17 (August 2018). Antonie Fountain and Friedel Huetz-Adams, Cocoa Barometer 2018. UN Comtrade, statistics for HS180100 (cocoa beans). Cocoa Market: Global Industry Analysis 2012–2016 and Opportunity Assessment 2017–2026 (Future Market Insights, February 2017). Chocolate Market by Type of Chocolate (Dark, Milk, and White) and by Sales Category (Everyday Chocolate, Premium Chocolate, and Seasonal Chocolate): Global Industry Perspective, Comprehensive Analysis, and Forecast, 2017 – 2024 (Zion Market Research, October 2018). Fountain and Huetz-Adams, Cocoa Barometer 2018. All figures in this paragraph: Fairtrade and Cocoa: Commodity Briefing (Fairtrade Foundation, April 2016). 6 7

  11. 11 provides a livelihood for around six million people, 30 per cent of the population. Cocoa is also an important source of foreign exchange: it is Côte d’Ivoire’s largest export, worth US$2 billion in 2013, accounting for 17 per cent of all exports, and Ghana’s third most important export, worth US$1.4 billion in 2013 and accounting for 11 per cent of all Ghana’s exports. 2.1 Poverty, incomes and productivity Many cocoa growers live in poverty. In 2018 Fairtrade International calculated that on average, cocoa farmer households earned only 37 per cent of the minimum living income in rural Côte d’Ivôire: US$0.78 per day compared to an estimated minimum living income level of $2.51 per day.8 Despite the size of the global chocolate market, the value of the final product that reaches cocoa farmers is small and shrinking: on average, farmers receive only between 3 and 7 per cent of the retail price of a chocolate bar, compared to up to 50 per cent in the 1970s and 16 per cent in the 1980s.9 By comparison, the chocolate brand manufacturers capture about 40 per cent of the price and retailers about 35 per cent. Most cocoa farms are very small: farmers in West Africa typically work plots of a few hectares, often growing cocoa alongside food crops; cocoa provides the family’s main cash income. In general, however, this income is too low to allow farmers to generate enough capital to invest in improvements in productivity or more sustainable practices. A farm of 2–4 hectares (ha) typically produces 300–400 kg of cocoa beans per hectare per year, perhaps half of its potential output. Photo: Tropenbos International 8 9 Ivorian Center for Socio Economic Research (CIRES), Living Income Report: Rural Côte d’Ivoire Cocoa-growing areas (ISEAL Alliance, Sustainable Food Lab and GIZ, 2018). Fairtrade and Cocoa; Oliver Nieburg, ‘Paying the Price of Chocolate: Breaking Cocoa Farming’s Cycle of Poverty’, ConfectioneryNews.com, 2014.

  12. 12 This is due mainly to poor farming methods and a lack of access to inputs, technology, finance and credit and extension services. Farmers generally cannot afford to improve depleted soil fertility or replace ageing or diseased cocoa trees; only about 10 per cent of West African cocoa farmers can afford to use fertiliser.10 Higher production in recent years has derived mainly from an increase in the total planted area rather than any rise in productivity; between 1990 and 2017, the area planted with cocoa rose from 0.7 million ha to 1.7 million ha in Ghana and from 1.6 million ha to 4.1 million ha in Côte d’Ivoire – a 150 per cent increase across the two countries.11 Historically, cocoa is a highly volatile commodity in terms of markets and prices, with many factors contributing to unstable prices and boom-bust cycles. Overproduction in 2016–17 saw the world market price crash by a third between September 2016 and February 2017, with the price falling from above US$3,000 per tonne to below US$1,900 in a matter of months; smallholder farmers saw their incomes decline by over a third in the following year. Cocoa-growers’ cooperatives have sometimes helped to raise farmers’ incomes and have contributed to reduced production costs by distributing inputs and services, but in most cases do not have the scale or levels of support to negotiate better prices; and in any case most cocoa farmers are not members. Government support has also helped but has been insufficient to resolve the problems. Weather patterns (now becoming more variable due to climate change), pests, disease and political disruption (Côte d’Ivoire experienced civil wars in 2002–07 and 2011) load additional challenges on to cocoa farmers. 2.2 Child labour The cocoa poverty trap, coupled with failures of law enforcement, has led to the widespread use of child labour. In 2015 a Tulane University report estimated that 1.2 million children in Côte d’Ivoire and 0.9 million children in Ghana worked in child labour in cocoa, with 80–90 per cent of them engaged in hazardous work, including carrying heavy loads, using hazardous chemicals and working with dangerous implements such as machetes.12 Subsequent surveys found that 26 per cent of children in Côte d’Ivoire and 46 per cent of children in Ghana in medium and high cocoa- growing areas worked more than the allowable hours for a child of their age in the week before reporting.13 After media and NGO reports in 2000 and 2001 on the widespread use of child labour in cocoa production, including its worst forms such as child trafficking, in 2001 US Senator Tom Harkin and Representative Eliot Engel negotiated the ‘Protocol for the Growing and Processing of Cocoa Beans and Their Derivative Products In a Manner that Complies with [International Labour Organization] ILO Convention 182 Concerning the Prohibition and Immediate Action for the Elimination of the Worst Forms of Child Labour’ (commonly known as the Harkin-Engel Protocol) with the major global cocoa companies. The protocol is a non-binding agreement for the cocoa industry to regulate itself without any legal implications, though Engel threatened to introduce legislation if the deadlines set in the protocol – including an end to child labour by 2005 – were not met. This threat has not been carried out. In 2002 the International Cocoa Initiative was established to work with the cocoa and chocolate industry, civil society, farming communities and national governments in cocoa-producing countries to advance the elimination of child labour.14 It carries out awareness-raising activities 10 11 12 13 14 Fairtrade and Cocoa. FAOSTAT data at http://www.fao.org/faostat/en/#data. Tulane University School of Public Health and Tropical Medicine, 2013/14 Survey Research on Child Labor in West African Cocoa Growing Areas (2015). Bitter Sweets: Prevalence of forced labour and child labour in the cocoa sectors of Côte d’Ivoire and Ghana (WalkFree Foundation, September 2018). https://cocoainitiative.org

  13. 13 and informs communities of the causes and consequences of child labour, of the importance of schooling and vocational training to improve children’s future prospects, and of the negative impacts of child labour on children’s health and education. It also helps cocoa-farming communities to develop and implement their own Community Action Plans; and embeds Child Labour Monitoring and Remediation Systems in the supply chain of chocolate and cocoa companies to identify and remediate cases of child labour. After some, though not all, of the initial targets in the Harkin-Engel Protocol had been missed by the 2005 deadline, it was extended to 2008; successive agreements extended it again to 2010, 2015, and, most recently, 2020. In 2010, the target of reducing the worst forms of child labour in Côte d’Ivoire and Ghana by 70 per cent (in aggregate) by 2020 was agreed. Despite these sectoral initiatives, and despite individual company commitments and programmes, in 2015 Tulane University concluded that the absolute number of child labourers in cocoa production in Côte d’Ivoire had increased by 46 per cent between 2008–09 and 2013–14, mostly because of the growth in total production and the corresponding increase in the number of cocoa-growing households.15 In Ghana, however, the number had fallen by six per cent, and across both countries there had also been a reduction in the number of hazardous tasks in which child labourers were involved, and an improvement in school enrolment, including for child labourers. Nevertheless, as the 2018 Cocoa Barometer report concluded, ‘Not a single company or government is anywhere near reaching the sector-wide objective of the elimination of child labour, and not even near their commitments of a 70 per cent reduction of child labour by 2020’.16 2.3 Deforestation Cocoa trees grow well under shade, and shade is indispensable in their early years; cocoa trees’ original environment was the Mesoamerican tropical forest, which provides natural shade trees. Nevertheless, some trees in existing forests must be removed to allow for cocoa, and often the forest is cleared completely to encourage faster growth of the trees and to increase total output; full-sun production provides higher yields in the short and medium term and is generally the model promoted by authorities.17 Logging the forest can also provide income to support the farmer, though the trees are often simply burnt. As cocoa trees age, however, their productivity declines and they become more vulnerable to disease. The ecological impacts of mass clearance, and the removal of shade trees – including lower soil moisture content, lower soil fertility, erosion and disturbed rainfall patterns – reduce yields yet further. Replanting in these circumstances is less attractive than opening up new forest, contributing to a pattern in which cocoa production migrates from one forest region to another, clearing forest as it goes.18 Although on a global scale cocoa is a less significant contributor to deforestation than palm oil, soy or beef, it has been a primary driver in West Africa. As noted above, the combined production area in Ghana and Côte d’Ivoire rose from 2.3 million ha to 5.8 million ha between 1990 and 2017. Only small remnants of primary forest remain in the two countries; Côte d’Ivoire lost 80 per cent of its forest area 15 16 17 18 2013/14 Survey Research on Child Labor in West African Cocoa Growing Areas. Fountain and Huetz-Adams, Cocoa Barometer 2018. Alan Kroeger et al, Eliminating Deforestation from the Cocoa Supply Chain (Climate Focus and World Bank Group, March 2017). Ibid.

  14. 14 between 1960 and 2010. Cocoa accounted for about eight per cent of the deforestation embodied in EU imports between 1990 and 2008, equivalent to an estimated 0.6 million ha of forest lost.19 The impacts of this deforestation include loss of biodiversity, destruction of livelihoods of those communities dependent on forest products other than cocoa, and local and regional environmental change. Northern Ghana and Côte d’Ivoire are hot, dry zones near the edge of the Sahel, and the frontier of this zone is creeping southward every year. One of the areas most vulnerable to desertification – now made worse by climate change – the destruction of West African forests threatens to accelerate the advance of the desert. 2.4 Governance and illegality Poor standards of governance and weak law enforcement also underlie many of the problems of cocoa production in West Africa. In Ghana, forest clearance is prohibited by law in national parks and forest reserves, but in practice the law is not well enforced. Although some cocoa farms in protected areas have been destroyed, the Ghana Cocoa Board (COCOBOD) has provided services to cocoa farmers within forest reserves that have been illegally cleared, and there are no sanctions on the purchase of cocoa from such illegally cleared areas.20 While all cocoa beans could be traced to communities, it is difficult to trace them to their farm of origin, as the traceability system is based on yield estimates per farm by individual sourcing companies; farms often sell to several companies, and it is not difficult to hide cocoa produced from forest reserves in the figures reported. Standards of governance are even worse in Côte d’Ivoire, not least because parts of the country are still in a post-conflict situation, with the state exercising only weak levels of authority. Illegal clearance of forests for cocoa is widespread, including, as in Ghana, in designated forest reserves and national parks.21 Although in recent years some illegal cocoa farmers have been evicted from forest reserves, little effort has been made to provide alternative livelihoods, and there have been reports of extortion and physical abuse by authorities. It is estimated that as much as 30–40 per cent of the Ivorian cocoa harvest currently comes from inside classified or protected areas, which technically makes it illegal.22 Weak ownership rights over land and forests also cause problems in both countries. In practice, trees are owned by the state, not the cocoa farmer, so cocoa trees may be cut down if the government allocates the area to a timber concessionaire; this undermines any incentive for cocoa farmers to allow natural tree regrowth (alongside the cocoa trees) on their farms. Farmers may also cut their trees to avoid illegal or informal extraction by loggers. In addition, many cocoa farmers are migrant workers, with only temporary rights of access to the land, and no incentive to replace ageing and unproductive cocoa trees. New tree tenure and registration systems being introduced in Ghana, and recent tree tenure reforms in Côte d’Ivoire, may help address some of these problems, though this will depend on their implementation.23 However, there is no effective national land-use planning process in either 19 20 Dieter Cuypers et al, The Impact of EU Consumption on Deforestation (European Commission, 2013), Vol. 1. Samuel Kwabene Nketiah et al, Scoping Study for the Relevance of FLEGT-VPA for Sustainable Agro-Commodity (Cocoa) Initiatives in Ghana (Tropenbos International, August 2018); Learning Lessons from FLEGT-VPA to Promote Governance Reform in Ghana’s Cocoa Sector (Tropenbos International, 2018). Etelle Higonnet, Marisa Bellantonio and Glenn Hurowitz, Chocolate’s Dark Secret: How the Cocoa Industry Destroys National Parks (Mighty Earth, September 2017); Cécile Barbière, ‘Forests decimated by cocoa farming in Côte d’Ivoire’, EURACTIV.fr, January 2018. Fountain and Huetz-Adams, Cocoa Barometer 2018. See ‘Unleashing the Potential of Tree Ownership in Ghana’, World Cocoa Foundation, 28 February 2019. 21 22 23

  15. 15 Photo: Tropenbos International country, and coordination between government agencies, such as the national cocoa agencies, land ministries and forest ministries, is weak. Both countries also possess legal frameworks that prohibit the worst forms of child labour. In recent years Côte d’Ivoire has passed legislation to make trafficking illegal and expanded its definition of hazardous work; the US Department of Labour (which publishes regular reports on international child labour and forced labour) concluded that the country had made ‘significant advancement’ in 2017.24 Nevertheless, inadequate resources, personnel, and training for law enforcement staff and labour inspectors meant that the laws were applied inconsistently; for example, during 2017 no inspections were conducted in the informal sector, which is where the majority of child labour occurs. Ghana was rated as having achieved ‘moderate advancement’, though the situation was not as bad as in Côte d’Ivoire. Similar problems existed, however, including a lack of resources and personnel for inspection and law enforcement, as well as some gaps in the law. In recent years, illegal mining operations have also emerged as a threat to cocoa farmers.25 The sharp increase in world gold prices since the beginning of the century (more than a six-fold increase by 2011, greater than four-fold today) led to both a rise in numbers of indigenous artisanal miners and to an influx of foreign miners, mainly Chinese, to both countries. Although some of this activity is legal, much is not, including the destruction of cocoa farms in search of gold deposits. 24 25 US Department of Labor’s 2017 Findings on the Worst Forms of Child Labor, Required by the Trade and Development Act of 2000 (US Department of Labor, 2018). ‘Ivory Coast: Illegal gold digging destroying cocoa plantations’, AfricaNews.com, 4 September 2016; Stacey Knott, ‘Swapping Cocoa for Karats’, USNews.com, 21 July 2017.

  16. 16 2.5 Future challenges All the factors described above pose a potential threat to the long-term viability of the cocoa industry. Ageing and diseased cocoa trees, sustained under-investment at farm level, deforestation and soil degradation are undermining attempts to increase productivity. Poverty, exacerbated by low and volatile prices, appears to be dissuading younger generations to enter cocoa farming; in 2008 it was reported that the average age of a West African cocoa farmer was 51.26 In recent years the attractions of gold mining have also lured cocoa farmers away. In 2012, Mars warned of future shortages of cocoa beans due to ‘unsustainable economic and environmental pressures on cocoa farms’, a warning echoed by Barry Callebaut in 2014, and reflected in the Berlin Declaration adopted by stakeholders at the Fourth World Cocoa Conference in April 2018.27 The companies’ fears of a one million tonne shortage by 2020, however, proved unfounded, as further investments in cocoa production, among other factors, led to a large surplus (and stockpiling) in 2016/17. In turn, this contributed to the crash in the market price mentioned in Section 2.1, with accompanying impacts on farmers’ incomes and levels of poverty. In 2018 the government of Côte d’Ivoire temporarily suspended the distribution of high-yield planting material to cocoa farmers in an effort to limit production volumes – though since much of the problem was due to the expansion of cocoa growing areas rather than to farm productivity, this was probably counterproductive. The short-term fears of cocoa shortages may not have been realised, but it seems highly unlikely that the problems outlined above – poverty, child labour, deforestation, illegality – can be resolved without more radical solutions.28 Increases in production, whether through increased productivity or an expansion in the cocoa area, may simply depress world prices further – and there are now some signs that global demand for cocoa may be levelling-off, or at least increasing more slowly than in recent years, adding urgency to the need to avoid driving cocoa farmers’ incomes even lower.29 A combination of small cocoa farm areas, low yields and low prices make cocoa production currently unsustainable; holistic solutions, including mechanisms to deliver higher prices, are needed to help farmer households earn a living income. To achieve sustainable production, farmers also need to be supported to produce more and better on less land, practising agricultural techniques that protect forests and the local environment and allowing diversification of their farms and the development of alternative sources of revenue. Further development of alternative livelihoods outside the cocoa sector is also likely to be essential. The next section examines current initiatives aiming to produce just these outcomes, and discusses why they have so far been insufficient. 26 27 Fairtrade and Cocoa. Rick Pendrous, ‘Mars urges action on cocoa to beat shortage, Food Manufacture, 12 June 2012; Tamsin Ford et al, ‘The cocoa crisis: why the world’s stash of chocolate is melting away’, Guardian, 21 November 2014; Berlin Declaration, April 2018, https://www.icco.org/about-us/icco-news/387-berlin-declaration-of-the-fourth-world-cocoa-conference.html. For a longer discussion, see, e.g., Laurent Pipitone, ‘The state of today’s cocoa market, as ICCO revises its forecast for production and grindings’, ConfectioneryNews.com, 11 June 2018. Fountain and Huetz-Adams, Cocoa Barometer 2018. 28 29

  17. 17 3 Background: existing initiatives Concerns over cocoa production and its impacts have been expressed for many years. As a consequence, a wide range of voluntary, private-sector and multi-stakeholder initiatives have been launched, with varying impacts. More recently, both producer-country and consumer-country governments have launched or participated in their own initiatives, sometimes jointly with the private sector. This section includes a brief summary. 3.1 Certification and standards Certification schemes exist to promote the production and consumption of products produced to higher social and/or environmental standards than the market norm. They normally feature a set of principles, criteria and indicators, sometimes drawn up through multi-stakeholder consultative processes, and generally subject to regular revision. They establish procedures for accredited certification bodies to award certificates to companies and products that meet agreed criteria, and procedures for tracing the movement of the products through the supply chain. Four main certification schemes have been in use in the cocoa sector. (Two of them, UTZ and Rainforest Alliance, have now merged; a new standard is expected to be launched by early 2020.) Combined, they certified between 1.7 million and 3.1 million ha in 2015 (the total figure is difficult to calculate because some production is certified under more than one scheme); the median, of 2.4 million hectares, represents 23 per cent of the global cocoa-growing area.30 In 2015 the coverage of the four schemes was as follows: Over 1.5 million hectares were UTZ certified. This standard covers both social and environmental issues, including specific criteria relating to deforestation and forest degradation. There is no guaranteed price premium for certified products, though certification does tend to increase yields and, therefore, incomes. • Rainforest Alliance cocoa, which covered more than 0.74 million hectares, is certified to Sustainable Agriculture Network (SAN) standards, a comprehensive set of environmental, social, and economic criteria, including specific criteria related to forests. Like UTZ, there is no guaranteed price premium for certified products. • Fairtrade International certified over 0.57 million hectares. Fairtrade standards are aimed mainly at tackling poverty and empowering producers; trading criteria for buying companies include a guaranteed minimum price level (which was activated in the 2016–17 price crash) and a fixed premium for producer organisations. Negative environmental impacts on protected areas and areas of high conservation value must be avoided, and certified producer organisations must have procedures in place to ensure their members do not cause deforestation in defined areas. • Organic cocoa covered more than 0.27 million hectares. Organic standards vary from country to country and region to region but include a general commitment to protect the environment and avoid the use of pesticides and certain fertilisers; there are generally no specific references to forests, and no criteria relating to human rights or labour standards. • 30 Julia Lernoud, Jason Potts, Gregory Sampson, Salvador Garibay, Matthew Lynch, Vivek Voora, Helga Willer and Joseph Wozniak, The State of Sustainable Markets – Statistics and Emerging Trends 2017 (ITC, 2017).

  18. 18 More certified cocoa tends to be produced than sold; in 2017 an estimated 22 per cent of globally traded cocoa was certified.31 This represents a ten-fold increase from about 2 per cent ten years before, a higher rate of growth of certification of any tropical commodity other than coffee.32 Some companies in the cocoa and chocolate industries have committed to sourcing 100 per cent certified cocoa by a target date; these include Ferrero and Hershey’s (both by 2020). Other companies, however, recognising the limitations of certification (see below), are now tending to combine investing in certification or buying certified cocoa with their own programmes, or to move away from using certification altogether. Several companies have set targets for sourcing 100 per cent responsibly, or sustainably, sourced cocoa – including Barry Callebaut, by 2025, and Cargill, by 2030 – including buying certified cocoa alongside their own programmes for sourcing and traceability. Others, including Lindt and Mondelez, have stopped using certification in favour of developing their own systems; Mondelez’s Cocoa Life programme has taken over their previous Fairtrade certification in Ghana. Mars is currently working with certification schemes but intends to go beyond the schemes’ current standards and practices in order to ensure that 100 per cent of the company’s cocoa is responsibly sourced and traceable by 2025. In addition to certification schemes and company initiatives, in 2019 the International Organisation for Standardisation (ISO) finished developing ISO 34101, a new standard for the process of producing sustainable and traceable cocoa. The standard covers four elements: requirements for cocoa sustainability management systems, for performance related to economic, social and environmental aspects, for traceability, and for certification schemes. The European Committee for Standardisation (CEN) is also in the process of approving the standard. Many elements of ISO 34101 are already incorporated in existing certification standards, but it does not include any requirements for minimum prices or price premiums; as potential barriers to trade and competition, these elements are not permitted for inclusion because of ISO’s relationship with the WTO (the WTO recognises measures based on ISO standards as in compliance with WTO rules). In March 2019, it was reported that the governments of both Ghana and Côte d’Ivoire had decided to develop their own standard in preference to ISO 34101, mainly because of its failure to deal with cocoa farmers’ incomes.33 3.2 Multi-stakeholder and producer-country initiatives As well as the certification schemes and company programmes discussed above, a range of private- sector and multi-stakeholder initiatives have been launched, some aiming to improve labour standards, some aimed at wider sustainability issues. As noted in Section 2.2, the International Cocoa Initiative was established in 2002 to work with the cocoa and chocolate industry to ensure that child and forced labour is not used in production. The World Cocoa Foundation, established in 2000, is an international membership organisation representing more than 100 member companies across the cocoa value chain. It aims to catalyse public-private action to accelerate cocoa sustainability, including through multi-stakeholder partnerships, public and private investment, policy dialogue, and joint learning and knowledge sharing.34 Among other initiatives it convenes Cocoa Action, launched in 2014 as a voluntary industry-wide strategy to align leading cocoa and chocolate companies, producer-country 31 32 33 34 Oliver Nieburg, ‘What does “going beyond certification” in cocoa sustainability really mean?’, ConfectioneryNews.com, 20 February 2018. Laurent Pipitone, ‘Cocoa markets still “unethical, unfair” despite 10 years of sustainability in the industry’, ConfectioneryNews.com, 13 November 2018. ‘Ghana joins forces with Ivory Coast to kick against inimical ISO standard to cocoa farmers’, mynewsgh.com, 13 March 2019. https://www.worldcocoafoundation.org

  19. 19 governments and key stakeholders on regional priority issues in cocoa sustainability.35 Most of the main cocoa and chocolate companies are participants. Following on from the 2007 and 2009 conferences of the Roundtable for a Sustainable Cocoa Economy, a multi-stakeholder process set up to foster dialogue about sustainability in the cocoa economy, the International Cocoa Organisation (ICCO) has organised a series of biennial World Cocoa Conferences to bring together actors from all stakeholder groups.36 The most recent was held in Berlin in April 2018. The most recent development is the Cocoa and Forests Initiative, established in 2017 by the World Cocoa Foundation, the IDH Sustainable Trade Initiative, and the Prince of Wales’ International Sustainability Unit.37 Originally including Côte d’Ivoire and Ghana, joined in 2018 by Colombia, each government has agreed Frameworks for Action with chocolate and cocoa companies, aiming to end deforestation and restore forest areas. Commitments include the introduction of farm mapping and traceability systems, no further conversion of any forest land for cocoa production, the elimination of illegal cocoa production in national parks, stronger enforcement of national forest policies and the development of alternative livelihoods for affected farmers. Updated action plans aimed at fulfilling these commitments were published in March 2019 by the governments of Côte d’Ivoire and Ghana and the 33 companies involved in the Initiative.38 Key priorities for Côte d’Ivoire included passage of the new Forest Code, the creation of a National Forest Preservation and Rehabilitation Fund, the development and implementation of the national cocoa traceability system and implementation of pilot projects in five priority regions. The National Plan for Ghana aimed to use the Ghana Cocoa Forest Reducing Emissions from Deforestation and Degradation (REDD+) Programme as a lever to significantly reduce greenhouse gas emissions from deforestation and enhance carbon stocks through sustainable forest management. Activities would include scaling up landscape approaches to end forest degradation in six hotspot intervention areas, improving cocoa yields through the adoption of environmentally sound practices and strengthening supply chain mapping. 3.3 Consumer-country commitments A number of consumer-country governments are now beginning to adopt commitments aimed at restricting sales or imports to sustainable or deforestation-free cocoa; this follows similar initiatives for other forest risk commodities (i.e. commodities associated with a high risk of deforestation), particularly palm oil and, more recently, soy. Belgium is the third largest importer of cocoa beans in Europe; about 500 companies are active in the cocoa processing industry and chocolate sector, varying from large multinationals to small and medium-sized enterprises (SMEs) and artisanal chocolatiers. In December 2018, the government announced its ‘Beyond Chocolate’ commitment to tackle deforestation and child labour and ensure a living income for cocoa producers.39 The aim of the commitment is for all Belgian chocolate produced or traded in Belgium to meet certification standards or to be produced with cocoa products from company-specific sustainability programmes by the end of 2025 at the latest. 35 36 37 38 https://www.worldcocoafoundation.org/about-wcf/cocoaaction/ http://www.worldcocoaconference.org https://www.worldcocoafoundation.org/initiative/cocoa-forests-initiative/ See https://www.worldcocoafoundation.org/press-release/action-plans-to-end-deforestation-released-by-governments-of-cote-divoire-and-ghana-and-leading-chocolate- cocoa-companies/ https://diplomatie.belgium.be/en/newsroom/news/2018/beyond_chocolate 39

  20. 20 Similarly, the agreements between governments and private partners under the Cocoa and Forests Initiative must be fully respected by the end of 2025. By 2030, deforestation as a result of cocoa production for the Belgian chocolate sector should end, and all cocoa producers should earn at least a living income. Signatories to the commitment included international and Belgian cocoa and chocolate companies, retailers, trade unions, civil society, social impact investors, universities and certification schemes. Germany is the second largest importer of cocoa beans in Europe. In 2012 the German Initiative on Sustainable Cocoa (GISCO) was launched jointly by the government, confectionery and retail companies and civil society, with the aim of ensuring that at least 70 per cent of the cocoa in the end products sold in Germany should originate from certified or independently verified cultivation by 2020. In January 2019, the German Federal Ministries for Economic Cooperation and Development and for Food and Agriculture launched a Ten-Point Action Plan for a Sustainable Cocoa Sector. This included the long-term aim of sourcing 100 per cent certified cocoa and the objective of establishing: ‘binding regulations setting a uniform standard for sustainably produced cocoa. Standards may be set on a voluntary or legislative basis. We are committed to establishing a uniform standard for the sustainable cultivation of cocoa at European level.’40 The German government has also committed to establishing a Living Income Task Force during 2019 to act on prices and incomes in the cocoa sector.41 Looking more broadly than cocoa, in November 2018, the French government published a national strategy to deal with imported deforestation, including proposals to stop importing products linked to deforestation and unsustainable agriculture by 2030, to help companies meet their own deforestation goals and to encourage financial institutions to take environmental and social issues into account in investment decisions.42 It also proposed the establishment of a due diligence requirement on companies at European and international levels, based on the French Devoir de Vigilance law of 2017, which requires companies to identify and mitigate the social and environmental risks related to their operations and supply chains (see Section 8.4). The strategy included the use of development aid, public procurement policy and minimum criteria for biofuel feedstocks, and the creation of a national platform for combating deforestation, bringing together businesses, NGOs and public authorities, to support the implementation and monitoring of the zero-deforestation commitments made by private-sector stakeholders, in particular by facilitating their work on traceability and risk analysis in supply chains. The government recognised that any regulations affecting imports would have to be applied at the EU level and noted that cocoa seemed ‘ripe for a rapid adoption of this type of regulation’. The UK government is beginning to explore options for a national approach to deforestation, following the commitment in its 2018 paper A Green Future: Our 25 Year Plan to Improve the Environment to ‘leave a lighter footprint on the global environment by enhancing sustainability and supporting zero deforestation supply chains’.43 The paper identified cocoa and palm oil as key commodities on which to focus (though it also looked forward to expanding this approach to other commodities) and included a proposal to establish a ‘cross-government global resource initiative … to work with businesses, NGOs, producer countries and intermediary countries. This will bring together key actors to identify actions across supply chains that will improve the sustainability of products and reduce deforestation.’ 40 41 42 See summary in English at https://www.worldcocoafoundation.org/blog/germany-presents-10-points-action-plan-for-a-sustainable-cocoa-sector/. Craving a Change in chocolate: How to secure a living income for cocoa farmers (Fairtrade Foundation, 2019). Stratégie Nationale de Lutte Contre la Déforestation Importée 2018–2030 (Ministère de la Transition Écologique et Solidaire, 2018); summary available in English at https://www.gouvernement.fr/en/ending-deforestation-caused-by-importing-unsustainable-products HM Government, A Green Future: Our 25 Year Plan to Improve the Environment (2018), p. 110. 43

  21. 21 The European Commission is currently considering the scope for EU action on deforestation, following several years of debate on the issue, partly stimulated by the model of the EU’s FLEGT Action Plan designed to tackle illegal logging and the trade in illegal timber (see Sections 6.1 and 8.3). In March 2018 the Commission published a feasibility study on options to step up EU action against deforestation, outlining twenty possible policy instruments which could be adopted by the EU and/or its Member States.44 Proposals included greater support for deforestation-free agriculture in producer countries, the wider use of public procurement policy in EU Member States, the adoption of a due diligence regulation for forest risk commodities, and greater scrutiny of agriculture investments in producer countries. In early 2019, the Commission launched a consultation on a roadmap designed to lay out options for the new Commission which will take office in late 2019. EU institutions have also discussed potential action against child labour. In 2012 the European Parliament adopted a resolution calling for action against the use of child labour in the cocoa sector, including the possible use of trade preferences.45 Two years earlier, the Parliament had called on the EU to establish a traceability mechanism for goods produced by forced or child labour, in order to ban their import, a call that was reiterated in 2016 after a report by the European Commission had concluded that trade restrictions would be of limited or possibly negative effect.46 No such legislative proposal has yet been put forward. 3.4 Are existing initiatives enough? There is increasing acknowledgement that the existing initiatives on cocoa reviewed above, while helpful, are not achieving, and are not likely to achieve, the aims of a fully sustainable supply chain for cocoa and chocolate, in either social or environmental respects. In April 2018, for example, the Berlin Declaration adopted by the Fourth World Cocoa Conference recognised that: ‘sector-wide efforts have proliferated to improve the lives of farmers, communities and the environment. However, these have not been enough to achieve significant impact at scale’ (para 1). The Declaration concluded that: ‘A new vision is needed in order to achieve true sector-wide sustainability’ (para 12) and set out a series of recommendations related to production, industry, consumption and management, including one recommendation calling for ‘all stakeholders … to strengthen human rights due diligence across the supply chain, including through potential regulatory measures by governments’.47 In October 2018, Joseph Boahen Aidoo, the chief executive of the Ghana Cocoa Board, stated that the cocoa market is still ‘unethical, unfair’.48 In February 2019 Michel Arrion, the newly appointed director of the International Cocoa Organisation (ICCO), accepted that there was a valid argument to ‘triple the price of cocoa for farmers … We have been talking about child labour, deforestation and too low farmgate prices for at least 15 years. It is time for action, and time to scale up.’49 Despite the relatively high penetration of certification in the cocoa sector compared to many other agricultural commodities, it is generally accepted that certification alone cannot be a sufficient solution to the problems of cocoa production (and indeed, the certification schemes themselves do not claim that it can).50 Although certification certainly has had positive impacts, it has not so far contributed significantly to achieving a living income for farmers, reducing child labour or halting deforestation. Although the average income of certified farmers does tend to be 44 45 46 Available at http://ec.europa.eu/environment/forests/studies_EUaction_deforestation_palm_oil.htm. European Parliament resolution of 14 March 2012 on child labour in cocoa sector. ‘Ban on import of goods produced using modern forms of slavery and forced labour, including that of children’, European Parliament website, 20 January 2018 (http://www. europarl.europa.eu/legislative-train/theme-europe-as-a-stronger-global-actor/file-ban-on-import-of-goods-produced-using-modern-forms-of-slavery). See https://www.icco.org/about-us/icco-news/387-berlin-declaration-of-the-fourth-world-cocoa-conference.html Pipitone, ‘Cocoa markets still “unethical, unfair” despite 10 years of sustainability in the industry’. Anthony Myers, ‘Chocoa 2019: the economics and political of cocoa’, ConfectioneryNews.com, 22 February 2019. Including, for example, Fountain and Huetz-Adams, Cocoa Barometer 2018. 47 48 49 50

  22. 22 higher than of their non-certified counterparts, the overall impact has been relatively low. Price premiums delivered through Fairtrade certification make only a slight difference to a farmer’s income – though increases in Fairtrade’s mandatory premium and mandatory minimum price scheduled for October 2019 will help significantly, and price premiums can also be an important source of revenue for cooperatives.51 Certification is also difficult to deliver to individual farmers not organised through cooperatives or other groups. The use of child labour may be missed by, or hidden from, certification auditors making occasional visits; no certification programme claims that it can guarantee that its products are free from child labour (though action will be taken if child labour is detected during audits). Neither certification schemes nor companies’ own systems have direct control over a range of variables that are essential to farmers’ livelihoods, including access to infrastructure (schools, health care, roads and access to markets, etc.) and the availability of inputs enabling them to increase productivity. In addition, while companies may be able to achieve improved standards where they buy cocoa beans directly from farms, this is much more difficult in indirect supply chains featuring traders and middlemen, which deliver the majority of some companies’ supplies. Although a national traceability system exists in Ghana, it does not provide full traceability back to the forest of origin (see Section 2.4), and no such system exists in Côte d’Ivoire, despite the country’s commitment under the Cocoa and Forests Initiative to introduce one. More broadly, land and forest governance and law enforcement, land use and cocoa sector planning, the determination of national cocoa prices and annual cocoa production levels, and education policy, all of which have significant impacts on conditions in the cocoa sector, are the responsibilities of government, which may be difficult for external stakeholders to influence, and cannot be addressed by certification or company schemes applying at the farm or supply chain level. Perhaps most importantly, systemic weaknesses in governance and law enforcement, including the prevalence of corruption, undermine many efforts to achieve sustainable production at the national level. Photo: Maartje de Graaf 51 ‘Cocoa farmers to earn more through a higher Fairtrade Minimum Price’, Fairtrade International, 3 December 2018.

  23. 23 4 The need for EU regulation As explained, the existing initiatives discussed in Section 3 have struggled to make progress in addressing problems of poverty, child labour, deforestation and illegality that persist across the cocoa sector as a whole. The impacts and coverage of certification schemes and company programmes are too limited and too many factors lie outside their ability to influence, including activities by producer-country governments on broader issues affecting the cocoa sector, and actions to tackle weak governance and poor law enforcement. At the same time, however, pressure is growing on the cocoa and chocolate sector to improve its standards. This could be through action by companies, or governments, or both. Governments and companies are focussing attention on forest risk commodities, including cocoa, for a number of reasons-- including the negative impacts examined above, wider awareness of the importance of forests and land use in combating climate change, and growing investor interest in sustainability. On top of this, recent years have seen a growing international focus on business behaviour and its impacts on human rights and the environment more broadly, and an increasing readiness amongst consumer-country governments to regulate business behaviour and increase scrutiny of supply chains. 4.1 Business, human rights and the environment In 2008, Professor John Ruggie, the UN Secretary-General’s Special Representative for Business and Human Rights, presented his ‘protect, respect and remedy’ framework, outlining states’ duty to protect against human rights abuses (including by business); the responsibility of companies to respect human rights (both to avoid infringing the rights of others and to address adverse impacts that occur); and the need for both states and businesses to strengthen access to appropriate and effective remedies for victims of business-related human rights abuses. The framework did not establish any new legal obligations on companies or states; the core rights it supports are those found in a range of international instruments, such as the International Bill of Human Rights and the International Labour Organisation’s core labour standards. The UN Guiding Principles on Business and Human Rights, agreed in 2011, provide guidance on how Parties should operationalise this ‘protect, respect and remedy’ framework.52 The publication of the Guiding Principles had an impact on other processes, including revisions of the Organisation for Economic Cooperation and Development (OECD) Guidelines for Multinational Enterprises and the International Finance Corporation’s (part of the World Bank Group) Environmental and Social Performance Standards. It also led to the formulation of the ISO 26000 social responsibility standard, which closely incorporates the ‘protect, respect and remedy’ framework and its implementing steps. It helped to prepare the ground for a series of national initiatives aimed at placing responsibility on businesses to scrutinise their supply chains for exposure to human rights abuses such as modern slavery, and to exercise due diligence to avoid human rights abuses and environmental harm in their operations and supply chains. Legislation dealing with modern slavery in supply chains now exists in California, the UK and Australia; with child labour in supply chains in the Netherlands; and with human rights abuses and environmental 52 Office of the UN High Commissioner for Human Rights, Guiding Principles on Business and Human Rights: Implementing the United Nations “Protect, Respect and Remedy” Frame- work (UNOHCR, 2011).

  24. 24 harm in general in France and (in preparation) in Switzerland (see further in Section 8). Similar legislation is under discussion in many other countries. In 2016, the German government set out its aim of seeing at least 50 per cent of all enterprises in Germany with more than 500 employees adopting human rights due diligence approaches by 2020; in 2018, the new coalition government stated that if that objective was not achieved, it would introduce appropriate legislation at the national level and also advocate EU-wide regulation.53 In March 2019, the European Commission’s Directorate General (DG) Justice commissioned a detailed examination of existing regulations and proposals for due diligence in the supply chain in the area of human rights, environment and governance in order to ‘help the Commission in assessing a range of options regarding due diligence through the supply chain and their likely impacts’.54 On 19 March the European Parliament’s informal working group on Responsible Business Conduct launched a Shadow EU Action Plan on Responsible Business Conduct; the Plan includes the adoption of legislation requiring corporations to carry out human rights due diligence regarding their operations, investments, business relationships and supply chains.55 4.2 The case for EU action As the world’s largest importer and processor of cocoa beans and cocoa products, and the largest consumer of cocoa and chocolate, the EU has the potential to affect the conditions under which cocoa is produced, and therefore the opportunity to address the negative impacts discussed above. It can do this through passing EU legislation to specify standards for the cocoa products placed on the EU market or to change business behaviour in respect of companies’ supply chains for cocoa and cocoa products, or through engaging with producer-country governments to encourage and support them to take action – or through a combination of all three approaches. These steps would provide incentives for producer-country governments, and requirements for all companies involved in the cocoa supply chain and operating in the EU, to provide a more enabling environment for the production of cocoa, and also act to close down EU markets for products not meeting the standards specified in the new measures (see Section 5). EU-wide regulation would also avoid the development of an array of different approaches by different EU Member States. In recent years the EU has adopted these kind of consumer-country, or demand-side, approaches with regard to illegally logged timber, illegal fish and conflict minerals. As discussed above, individual Member States are beginning to regulate aspects of business behaviour, and the European Commission has embarked on processes that could lead to regulation for forest risk commodities, and for business behaviour in general. Against this background of national and EU-level initiatives, a number of cocoa and chocolate companies have called for EU-level regulation, recognising that voluntary measures have not achieved the desired results. In March 2019 Virginie Mahin, global social sustainability and human rights lead for Mondelez International, argued that: ‘many companies like ours are already implementing due diligence measures and we want to avoid a patchwork of legislation at national level … While there are a lot of good voluntary initiatives we still think it would be beneficial to have a binding law at EU level to provide a level playing field, and bring along companies upstream 53 Saskia Wilks and Johannes Blankenbach, ‘Will Germany become a leader in the drive for corporate due diligence on human rights?’ Business and Human Rights Resource Centre, 20 February 2019. ‘Call for input: EU Commission study on regulatory options for mandatory human rights due diligence’, Business and Human Rights Resource Centre, 22 March 2019. The cited text is from the linked DG Justice announcement. Shadow EU Action Plan on the Implementation of the UN Guiding Principles on Business and Human Rights within the EU, available at https://responsiblebusinessconduct.eu/wp/ wp-content/uploads/2019/03/SHADOW-EU-Action-Plan-on-Business-and-Human-Rights.pdf. 54 55

  25. 25 in the supply chain, which may not be under the same consumer-facing pressure … We need to make sure that at EU level there is policy coherence across policy sectors from trade to foreign aid, with a smart mix of measures including supply chain due diligence.’56 The Barry Callebaut Group tweeted that: ‘An EU due diligence policy adds value by creating a level playing field among companies in driving the demand for sustainably sourced raw materials. For it to be meaningful, it has to be embedded in a strategy led by the EU agreeing action plans with origin governments.’57 Earlier that month John Ament, Global Vice President of Cocoa at Mars, had accepted that: ‘it seems inevitable that there will be some form of regulation of the cocoa supply chain in the future … Legislation could create the enabling environment needed to overcome the challenges facing the cocoa sector’.58 What form should such regulation take? Sections 6, 7 and 8 of this paper outline eight possible options for action by the EU (and, in principle, by other consumer countries): Options focused on actions by and with producer countries (Section 6): Option A: Bilateral agreements with cocoa-producing countries Option B: ‘Carding’ system Options focused on regulation of the EU market, or segments of it (Section 7): Option C: Public procurement policies Option D: Regulation of sales in the EU market Options to regulate business behaviour (Section 8): Option E: Corporate reporting for cocoa supply chains Option F: Competition law reform Option G: A due diligence regulation for cocoa placed on the EU market Option H: A broader corporate due diligence requirement Almost all of these measures have been applied to other commodities or products in some jurisdictions, some by the EU itself. Lessons from these other sectors are drawn in the sections below. Most of these options are not mutually exclusive, and various combinations are possible. All options are also applicable to other commodities associated with negative environmental and human rights impacts such as deforestation or forced or child labour. Before analysing the options for regulation, however, a discussion needs to be had on the standards for cocoa production it is aiming to achieve. 56 57 58 Benjamin Fox, ‘Companies will support EU law on due diligence, but need assurances on liability’, Euractiv.com, 19 March 2019. See https://twitter.com/BCgroupnews/status/1111153494537445377. See https://www.linkedin.com/pulse/threats-opportunities-facing-cocoa-what-role-can-industry-john-ament/.

  26. 26 5 Standards for EU regulation In order to tackle poverty, child labour, deforestation and poor governance in the cocoa sector, what social and environmental standards should any new EU approach promote? This is a matter for discussion in the context of the development of the policy, but four objectives cover the impacts discussed above: 1. Respect for the laws of the producer country, including in particular laws relating to: human rights; forced and child labour; employment conditions, such as minimum wages, and health and safety conditions; rights of ownership and access to land; and environmental protection, including in particular the protection of forests. 2. Respect for international human rights and labour law, including prohibitions against forced and child labour; the right to decent working conditions; the right to a living wage for workers and a living income for smallholders; and the right to free association. 3. Improvements in governance and law enforcement. 4. The promotion of higher standards in cocoa production than provided in national laws, including: Ū Improved protection for rights of tenure, ownership and access to land. Ū Higher labour and living standards, such as improved incomes facilitated by guaranteed minimum prices and other interventions. Ū Higher environmental standards, particularly relating to the protection of forests, such as a prohibition on deforestation, the maintenance of existing shade trees or requirements to plant new trees. Many of these issues are reflected in the criteria set in some of the certification schemes, and some in the emerging ISO 34101 standard, and in the aims of many of the voluntary initiatives examined above in Section 3. Many of them are also included in the OECD’s Due Diligence Guidance for Responsible Business Conduct, published in 2018 and based on the OECD Guidelines for Multinational Enterprises.59 The Due Diligence Guidance sets out a due diligence framework to be used by enterprises to avoid and address adverse impacts in their operations, supply chains and business relationships, covering human rights, employment and industrial relations, environment, bribery and extortion, consumer interests, and disclosure of information. This framework is also reflected in the OECD – FAO Guidance for Responsible Agricultural Supply Chains published in 2016, which also includes a model enterprise policy outlining the standards enterprises should observe in building responsible agricultural supply chains, a description of the major risks faced by enterprises and the measures to mitigate these risks, and guidance for engaging with indigenous peoples.60 59 60 Available at http://www.oecd.org/investment/due-diligence-guidance-for-responsible-business-conduct.htm. Available at http://www.oecd.org/daf/inv/investment-policy/rbc-agriculture-supply-chains.htm.

  27. 27 Other international frameworks provide appropriate guidance – including, for example, the Voluntary Guidelines on the Responsible Governance of Tenure of Land, Fisheries and Forests in the Context of National Food Security agreed by the Committee on World Food Security in 2012 to promote secure tenure rights and equitable access to land, fisheries and forests as a means of eradicating hunger and poverty, supporting sustainable development and enhancing the environment.61 In setting standards for the sustainable (or ‘responsible’) production of cocoa (or any other commodity), there is always a potential tension between resting the criteria on the laws and regulations of the producer country (i.e. groups 1, 2 and 3 above), thus respecting their national sovereignty and right to determine their own development path, and imposing, or being perceived to be imposing, standards from the outside (group 4 above). Clearly there are many advantages to the first approach, which is, by and large, that taken by the EU in its strategy to deal with illegal logging and the trade in illegal timber (see further below, Sections 6.1 and 8.3). It has the disadvantage, however, that the laws in force, and their enforcement, may not be adequate to achieve the desired social and environmental aims. The certification systems discussed in Section 3 go beyond national laws, but they are voluntary and not mandatory standards. Some of the objectives, including, in particular, improvements in law enforcement and legal reform in producer countries, will be difficult to achieve through EU-level regulation in isolation – unless the regulation is effective enough to promote action by the producer-country governments in order to support their cocoa exports. For these objectives, engagement by the EU and its member states with producer-country governments, including financial and capacity-building support, is likely to be necessary, possibly within the framework of a bilateral agreement (which is one of the options explored below in Section 6). This is also true of the broader objectives discussed in Section 2.5, such as support for improved farming techniques, the development of alternative livelihoods and investment in education. And all else being equal, initiatives agreed and implemented jointly between consumer and producer countries are always preferable to unilateral action by consumer countries. For this reason, a combination of some of the options discussed below is likely to result in the best outcomes. Photo: Tropenbos International 61 See http://www.fao.org/cfs/home/activities/vggt/en/.

  28. 28 6 Producer-country-focused options The two options below focus on actions to influence conditions in cocoa-producing countries: one through reaching bilateral agreements with producer countries (Option A), and the other through a ‘carding’ system applied to producer countries’ exports (option B). 6.1 Option A: Bilateral agreements with cocoa-producing countries This option envisages the EU negotiating bilateral agreements (or possibly, though less likely, a multilateral agreement) with the main cocoa-producing countries, to raise standards of production and improve the supporting environment, including governance and law enforcement. This is complementary to almost all the other options discussed below and could form a valuable component of a broad package of measures to address cocoa production and consumption. Existing model: timber VPAs This option is modelled mainly on the EU’s FLEGT VPAs with timber-producing countries. By May 2019, VPAs had been concluded with nine countries and negotiations were under way in a further six, with several more expressing interest.62 Under the terms of the VPAs, each partner country must establish a timber legality assurance system (a national traceability system) and an export licensing system to ensure that only timber products that have been produced legally can be licensed for export to the EU (in fact all VPA countries so far have announced their intention to license all exports regardless of destination). The EU has legislated to require the presence of the FLEGT license before it permits the entry of timber products from the partner country into the EU market. In November 2016, Indonesia started to issue FLEGT licenses to accompany its timber exports, its timber legality assurance system having been judged adequate by both parties to the VPA. To date it is the only VPA country to establish a successful licensing system, though Ghana is expected to finalise its own scheme soon; in general, legality assurance systems have been more complex and difficult to establish than originally anticipated. Despite this slow progress in finalising the licensing systems, in many cases the process of negotiating the VPAs has itself driven significant improvements in forest governance. In most cases the VPA negotiations have seen the adoption of multi-stakeholder processes to agree operational definitions of ‘legal timber’ – i.e. which of the producer country’s laws are relevant – and all the agreements contain commitments to regulatory and policy reform to make forest laws and regulations clearer and more comprehensive, together with improvements in transparency and stakeholder involvement. They also contain provision for independent auditors to check the integrity of the legality assurance and licensing systems; in some cases, civil society independent monitors also scrutinise the forest sector and illegal activities more broadly. The EU and Member 62 The nine countries with which VPAs have been concluded are Cameroon, Central African Republic, Ghana, Guyana, Honduras, Indonesia, Liberia, Republic of Congo and Vietnam. The six with which negotiations are ongoing or anticipated shortly to start are Côte D’Ivoire, Democratic Republic of the Congo, Gabon, Lao People’s Democratic Republic, Malaysia and Thailand.

  29. 29 States have provided capacity-building assistance to partner countries to help them set up the licensing scheme, improve enforcement and, where necessary, reform their laws. Studies of the VPAs have concluded that in some countries their main achievement, even before the implementation of the licensing system, has been a significant improvement in governance. A 2016 study of the VPA processes in Ghana and Indonesia, for example, concluded that they had: ‘resulted in significant improvements in forest governance in both countries, including measurable declines in illegal logging … the VPA implementation process has led in both countries to substantially increased participation by civil society and other stakeholders in forest governance, greater transparency and accountability of forestry administration, and heightened recognition of community rights. In both countries, too, the VPA process has focused attention on protecting the needs and livelihoods of small producers in the transition to the new timber legality regime … the VPA process has contributed to reducing arbitrary administrative discretion in forest governance, including the award of concessions and harvesting permits, while creating new mechanisms for exposing corruption across the supply chain, whose effectiveness can be expected to grow as the monitoring, reporting, and review provisions of their timber legality assurance schemes kick into full gear with the onset of FLEGT licensing.’63 Associated benefits in Ghana have included the adoption of the 2012 Forest and Wildlife Policy, which was heralded as providing the most ambitious steps to date towards addressing tree tenure and community participation in forest management, and a 640 per cent increase in logging taxes collected by the government.64 The linkage of trade to governance and law enforcement has been important, with access to the EU market providing an incentive to partner countries to negotiate and implement VPAs. The adoption of the EU Timber Regulation in 2010 was a necessary complementary step (see further below in Section 8.3). The Regulation obliges any timber operator first placing timber products on the EU market to have in place a system of due diligence to minimise the risk of them handling illegal timber, but FLEGT-licensed products automatically meet the requirements of the Regulation; evidence suggests that it is now becoming easier to sell Indonesian timber products on the EU market as a consequence.65 Some countries with very little trade with the EU have nevertheless expressed an interest in agreeing VPAs, suggesting that market access is not the only incentive; participation in a system which aims to improve standards of forest governance and reduce levels of illegal logging seems itself be a reason to join a VPA, as well as recognition of the country’s efforts by the EU. Application to cocoa In principle, it should be relatively straightforward to apply at least some elements of the VPA model to cocoa and cocoa products. Indeed, the two main cocoa-producing countries already have a VPA for timber agreed (Ghana) or are negotiating one (Côte d’Ivoire). In Ghana, the Civil Society Cocoa Platform, which includes NGOs and community organisations working on cocoa and forests, called for a system similar to the VPAs and the EU Timber Regulation to be applied to cocoa, in their submission to the European Commission’s consultation on a deforestation action plan. The details of what such a system could look like for cocoa were described by Ghanaian and European NGOs in a briefing paper published in 2018.66 63 Christine Overdevest and Jonathan Zeitlin, Experimentalism in Transnational Forest Governance: Implementing EU Forest Law Enforcement Governance and Trade (FLEGT) Voluntary Partnership Agreements in Indonesia and Ghana (Amsterdam Centre for Contemporary European Studies, SSRN Research Paper 2016/02), pp. 60–61. ‘Civic Response and the EU-Ghana VPA: protecting forests, communities and the climate’ (Civic Response Ghana, February 2017). Rupert Oliver, ‘The changing tropical timber trade’, presentation to Forest Governance Markets and Climate forum, London, March 2019. Tropenbos, EcoCare Ghana, Forest Watch Ghana and Fern, Transferring Lessons from FLEGT-VPA to Promote Governance Reform in Ghana’s Cocoa Sector (2018). 64 65 66

  30. 30 Furthermore, the size of the cocoa trade between the two regions is significant: in 2017, 67 per cent of the EU’s imports of cocoa beans, by value, originated from Côte d’Ivoire (52 per cent) and Ghana (15 per cent) together, while 57 per cent of Côte d’Ivoire’s exports of cocoa beans and 43 per cent of Ghana’s were destined for the EU.67 The West Africa – EU trade is a far greater proportion of world cocoa trade than the trade in timber between the EU and VPA partner countries is as a proportion of world timber trade. Many of the negative impacts of cocoa production stem from failures of governance, including weak enforcement of forest and child labour laws, lack of legal protection for some types of trees, weak tenure rights for farmers, lack of transparency and accountability in the government agencies responsible for cocoa, and poor government coordination and land-use planning. As discussed in Section 2.4, illegal activity is common in the cocoa sector in both countries. Bilateral agreements on cocoa could help to address these issues, including through the establishment of similar multi-stakeholder decision-making frameworks to those that have emerged in the timber VPAs, and through similar processes of defining the problems and reaching agreed solutions for policy and legal reform. Improvements in transparency around government decision-making would also have the potential to improve outcomes. A major advantage of this approach is that it leaves the core of decision-making in the hands of the producer country; it focuses on improving, applying and enforcing the producer country’s laws (though these may be reformed), not at imposing any external standard. Effective enforcement of the producer countries’ laws on cocoa production, including those on child labour and the protection of forest reserves, would be a major step forward. By themselves, however, this may not be adequate to achieve all the aims set out in Section 5. Other potential elements of a cocoa agreement could therefore include strategies to create enabling national policy environments to make faster progress on child labour, fair prices, land and tree tenure, land use planning, deforestation and transparency over taxation and revenues. Going beyond existing national frameworks, these would require commitments to policy and legislative reforms, with appropriate technical, financial and diplomatic support from the EU. Some existing initiatives could help pave the way to such bilateral agreements. As well as the timber VPA with Ghana, and the agreement under negotiation with Côte d’Ivoire, the Cocoa and Forests Initiative may provide some elements of a cocoa agreement, though these would need to be supplemented extensively; the Initiative does not put in place mechanisms to deal with governance and law enforcement, for example, nor to better involve cocoa farmers or civil society in national policy issues. It does, however, include the aim of establishing national traceability systems, which a VPA-type agreement could help support. The work of the International Cocoa Initiative on child labour could also be assisted through the agreements. As noted above, preferential market access has been an important incentive to persuade countries to take VPAs seriously and would probably be necessary for equivalent cocoa agreements. In the VPAs, this has been provided via the licensing component. Once the EU Timber Regulation entered into force, FLEGT-licensed products from VPA partner countries gained automatic entry to the EU market; companies first placing FLEGT-licensed timber on the EU market do not have to meet any of the other requirements of the Regulation. A bilateral agreement for cocoa could thus be complemented by the type of regulation discussed below in Section 8.3. As the VPAs have shown, however, it has proven difficult and resource-intensive to develop national legality assurance and licensing schemes for timber. After more than a decade of implementation of the FLEGT Action Plan, and several years after VPAs have entered into force, only Indonesia has 67 UN Comtrade, statistics for HS180100 (cocoa beans).

  31. 31 started to fully implement a licensing scheme, though Ghana is reportedly close to doing so; other national systems are in varying stages of construction. Long-term capacity-building support would be needed for the producer countries to set up the legality or sustainability assurance and licensing schemes, as well as to implement the broader reforms listed above. Could there be alternatives to licensing systems that would provide trade incentives for governance improvements? One idea could be to offer lower import tariffs to cocoa imports from countries with agreements. However, like unprocessed timber, imports of cocoa beans to the EU already face zero import duties, so there is no real scope to apply different levels of duty. An alternative option would be to borrow elements of the approach discussed in Option B below, and use a ‘carding’ system to indicate to importing businesses the producer country’s progress in meeting its commitments, with a green card (or rating) signalling good progress, a yellow rating concern and a red rating a high risk of non-compliance. If this was accompanied by a due diligence regulation like that described in Section 8.3, importers would then need to adopt stricter standards of due diligence depending on the colour of the rating and level of risk. If the national traceability system develops adequately, a carding system could be applied to regions or forests of origin rather than to entire countries. Photo: CIFOR / Flickr

  32. 32 6.2 Option B: ‘Carding’ system This option envisages the EU entering into dialogue with countries exporting cocoa to the EU, and operating a ‘carding’ system, issuing yellow or red cards to those countries not acting effectively to combat illegal behaviour in the supply chain. Existing model: the EU IUU Fishing Regulation Illegal, Unreported and Unregulated (IUU) fishing is one of the most significant impediments to the achievement of sustainable fisheries at a time of mounting threats to marine biodiversity and food security. While many individual countries have adopted measures to combat IUU fishing, the most extensive regulation has been introduced in the EU, which collectively is the largest global importer of seafood products, accounting for about a quarter of world trade by value. The EU Regulation to Prevent, Deter and Eliminate IUU fishing ((EC)1005/2008), which entered into force in 2010, aims to exclude IUU fish from the European market.68 It allows for substantial penalties to be applied to EU nationals who engage in or support IUU fishing anywhere in the world, and establishes a catch certification scheme under which only marine fishery products validated as legal by the competent flag state can be imported to or exported from the EU (similar in outline to the FLEGT licensing system for timber imports discussed above). It also enables the European Commission to enter into dialogue with non-EU countries that are not combating IUU fishing effectively. Countries may be issued a ‘yellow card’ (officially warned) and if reforms are not carried out, or not carried out quickly enough, a ‘red card’, resulting in a ban on imports to the EU of fish products caught by vessels flying the flag of the red-carded country. Both yellow and red cards can be lifted when there is clear evidence that the situation has been rectified. A 2015 summary of activities to implement the Regulation showed that since 2010 a total of 91 third countries had notified the Commission that they had the necessary instruments, procedures and administrative structures for the certification of catches by vessels flying their flag. The Commission had entered into dialogue with more than 50 third countries to assess the adequacy of their systems to combat IUU fishing in line with international law, and since 2012, the EU had ‘yellow carded’ 23 countries, ‘red carded’ four and delisted (i.e. removed the cards from) 11.69 Both yellow and red cards appear to have prompted significant reforms by flag states.70 For example, after being yellow-carded in 2012 and red-carded in 2014, Belize renationalised its vessel registry (after evidence that the privatised version had been used to issue flags of convenience), removed vessels with a record of IUU fishing, and instituted more rigorous policing of vessels fishing under its flag. Similarly, Korea was issued a yellow card after failing to curb IUU fishing activity off the coast of West Africa by vessels in its distant water fleet. It was delisted in April 2015 after it revised the legal framework governing its fleet in line with international requirements, established a fisheries monitoring centre to control all its fishing fleets, installed vessel monitoring systems on board all Korean-flagged distant water fishing vessels (approximately 300), and improved its on-board observer programme. 68 See summary and links at https://ec.europa.eu/fisheries/cfp/illegal_fishing_en; The EU IUU Regulation – Building on Success: EU progress in the global fight against illegal fishing (Environmental Justice Foundation, Oceana, Pew Charitable Trusts and WWF, February 2016). European Commission, ‘Question and Answers on the EU’s fight against IUU fishing’, 21 April 2015 (http://europa.eu/rapid/press-release_MEMO-15-4807_en.htm). The EU IUU Regulation – Building on Success, pp. 8–9. 69 70

  33. 33 Application to cocoa Of all the options examined here, this is probably the one of least relevance to cocoa, at least in the form described above, a carding system linked to import bans. Although this approach does appear to have had a positive impact in combating IUU fishing, it depends in part on the fact that a large number of countries export fish to the EU; as noted above, by 2015 no less than 91 had notified the European Commission of their arrangements, and 50 had entered into dialogue over the adequacy of their systems. In contrast, cocoa imports to the EU are dominated by just two countries, Ghana and Côte d’Ivoire which accounted between them for 67 per cent of the EU’s imports of cocoa beans, by value, in 2017. Any ban on imports of cocoa from either of those countries, and particularly from Côte d’Ivoire (52 per cent of EU imports), would be hugely disruptive to EU- based companies’ supply chains. The system could perhaps be operated against smaller supplying countries, but this would be obviously discriminatory, and would fail to tackle the major problems of cocoa production in the two main West African countries. As a unilateral trade restriction, the EU IUU Regulation also raises some unresolved questions about its compatibility with WTO trade rules, though as yet there have been no challenges to it through the WTO. However, there may be scope to examine whether the early warning system embodied in the yellow card could be adapted to the kind of bilateral agreements discussed above in Section 6.1, when partnered with a due diligence regulation such as described in section 8.3. Under that approach, the carding system would not lead to import bans, but rather serve as an indicator of the level of risk of the producer country to importers into the EU. Officially raising the risk profile of the country via a yellow or red card would offer a means for either party to raise concerns over its implementation, and lead to a process of reform and resolution. A yellow card could require a process of dialogue and improvements in transparency, perhaps with independent verification systems being put in place, and a red card could lead to a mutually agreed corrective implementation plan, to be implemented over time with support from the EU. Photo: jbdodane / Flickr

  34. 34 7 EU-market-focused options The following two options focus on measures applied to the EU market (Option D), or to a specific segment of it, that supplying government purchasers (Option C). 7.1 Option C: Public procurement policy Under this option, governments would use their public procurement policies to specify that cocoa, or any products containing cocoa, purchased by government buyers would be required to meet minimum criteria for responsible sourcing – e.g. to meet the aims discussed in Section 5 in terms of legality and high social and environmental standards. Existing models Public procurement is the acquisition of goods and services from a third party on behalf of a public agency, such as a government department or local authority. In all EU Member States, purchasing by public authorities – central, regional and local as well as their agencies – is significant, accounting for about 12 per cent of Gross Domestic Product (GDP) on average, though this varies substantially by sector and product.71 Many governments use their public procurement policy to encourage the purchase of sustainably or responsibly sourced products for use in the public sector; for example more than 30 countries, mostly in the EU, now require or encourage public-sector purchasers to buy or specify timber products that are legally or sustainably produced (the details differ by country).72 The public sector is a major purchaser of food and catering services, for schools, nurseries, hospitals, care homes, canteens, prisons and the military, and many governments already include criteria relating to food and catering in their public procurement policy; these are probably more common in local and regional governments than at the central level, though no systematic survey has been conducted.73 The criteria used include preferences for organic, healthy, seasonal and sometimes fairly traded products; a few governments, including the UK, Sweden and the Netherlands, now include criteria related directly to deforestation for products containing palm oil or soy. (EU procurement rules do not allow procurement policies to specify particular certification schemes, like Fairtrade; they need to set out criteria that any product can satisfy whether or not it is certified. The rules do, however, allow procurement policies to identify which certification schemes meet their criteria.) At the EU level, the European Commission publishes voluntary green public procurement (GPP) criteria for a range of products and services and encourages Member States to use them. The criteria for food and catering are currently under development, but the preparatory technical document includes criteria relating to certified palm oil and soy, and ‘fair and ethical’ coffee, tea, cocoa, sugar and bananas.74 71 72 73 74 For a longer discussion, see Duncan Brack, Reducing Deforestation in Agricultural Commodity Supply Chains: Using Public Procurement Policy (Chatham House, 2015). Duncan Brack, Promoting Legal and Sustainable Timber: Using Public Procurement Policy (Chatham House, 2014). Brack, Reducing Deforestation in Agricultural Commodity Supply Chains: Using Public Procurement Policy. See current draft proposed criteria at http://susproc.jrc.ec.europa.eu/Food_Catering/docs/170127_EU%20GPP%20Food%20catering%20criteria_TR2.0.pdf.

  35. 35 Application to cocoa Criteria for cocoa and cocoa products are already included in at least some EU Member States’ public procurement policies. The UK central government, for example, encourages the purchase of organic and fairly-traded cocoa. It should be relatively straightforward to extend this to include other certification systems and for other Member States to apply the same or similar policies. Given that much food and catering purchasing is probably carried out at regional and local government level, the adoption of procurement policies by these authorities would have more impact. In practice, however, public purchasers would almost inevitably need to rely on the certification schemes and company programmes discussed in Section 3 to meet the criteria set in their procurement policies. Dealing with hundreds of products, often purchased through wide-ranging procurement services or supplier contracts, public buyers cannot be expected to research the background of all the cocoa or chocolate products they may end up purchasing. They therefore need readily accessible signals or guidance. Governments could establish advisory services to assist government purchasers in choosing the right products, as the Dutch and British governments did for timber procurement; though in practice the main outcome was still for government buyers to purchase certified timber. Whether the outcomes of this approach would meet the standards set out in Section 5 would then depend on the adequacy of the criteria contained in the certification schemes and company programmes. As discussed in Section 3, these cannot meet all the objectives, so while this option would be a step in the right direction, helping to grow the market for responsibly sourced cocoa and sending a signal to the market of the government’s aims, it is unlikely to achieve major change in the cocoa sector. 7.2 Option D: Regulation of sales in the EU market Under this option, any cocoa or product containing cocoa placed on the EU market would be required to meet minimum criteria for responsible sourcing – e.g. to meet the kind of standards discussed in Section 5 in terms of legality and high social and environmental standards. This is the extension of the approach discussed above for public purchasing to cover all sales in the EU market. Existing models Virtually all products offered for sale on markets in the EU must meet a variety of minimum standards, most obviously concerning consumer health and safety. This is particularly true of the food and catering sector, where foodstuffs must be produced and sold hygienically, and must meet rules governing, for example, the presence of genetically modified ingredients. In addition, various regulations govern labels, designations and claims; for example, food produced, manufactured, imported, sold or traded into the EU as ‘organic’ must meet the minimum standards set out in the EU Organic Regulation, relating to sustainable crop rotations, recycling of nutrients, cultivation techniques, avoiding the use of fertilisers and pesticides, etc., and must be certified by recognised certification bodies. The EU Timber Regulation (see Section 8.3) prohibits the placing of illegally harvested timber and timber products on the EU market, whether sourced from domestic production or from imports. ‘Legality’ is defined in relation to the laws of the producer country on the right to harvest; payment of fees and taxes; forest management and biodiversity conservation; land use and tenure rights; and trade and customs laws.

  36. 36 Application to cocoa Applying legality or broader sustainability criteria across the entire EU economy is, clearly, a much bigger challenge than including criteria in public procurement policies. It should in principle be possible to mirror the approach of the EU Timber Regulation and require all cocoa products to have been legally produced; this would probably work most effectively accompanying a due diligence regulation (see Section 8). It could be possible to apply a set of criteria extending beyond producer-country legality, but this would depend primarily on the rate of expansion of the certification schemes and the company programmes discussed in Section 3, and perhaps the new ISO 34101 standard (depending on the extent of its uptake); in practice, these are probably the only means besides implementing producer-country legality by which criteria relating to human rights, child labour, deforestation, etc., could be required across the entire EU market. In principle, this could be achieved given sufficient time; and adopting such a policy, even if the target date is set many years in the future, would provide a powerful incentive for the uptake of the schemes and programmes. In effect, this is what the Belgian ‘Beyond Chocolate’ commitment discussed in Section 3 is aiming at, though how this aim is to be achieved, other than through voluntary action, is not clear. It is not clear whether this kind of sales restriction could be challenged as incompatible with WTO trade rules. Although in one sense it is non-discriminatory, applying equally to domestic production and imports (one of the WTO’s core principles), there is no cocoa produced inside the EU, so in reality the measure applies only to imports. It could nevertheless be found to be justified under one of the allowed exceptions to WTO disciplines, but this is a broad debate that goes beyond the remit of this paper.75 Whether the outcomes of this approach would meet the objectives set out in Section 5 would then depend on the adequacy of the criteria contained in the certification schemes and company programmes. As discussed in Section 3, these cannot meet all the objectives. While this option could help to grow the market for responsibly sourced cocoa, this is a less flexible option than the due diligence approaches discussed in Section 8, depending more on existing certification schemes and company programmes and taking longer to implement. It seems unlikely to trigger radical change in the sector. 75 For discussion on the interaction of WTO trade rules with restrictions on sales and imports of illegal or deforestation-associated products, see Duncan Brack, WTO Compatibility with EU Action on Deforestation (Fern, 2015).

  37. 37 8 Business-focused options These four options are focused on regulating the behaviour of businesses based in or operating in the EU or placing cocoa or cocoa products on the EU market. 8.1 Option E: Competition law reform Under this option, EU competition law is reviewed, clarified or amended to allow businesses greater freedom to collaborate for sustainability purposes, factor in externality costs and, in particular, discuss and address low prices paid to farmers. Existing law European competition law today derives mostly from articles 101 to 109 of the Treaty on the Functioning of the European Union, as well as a series of regulations and directives, and is enforced mainly by DG Competition in the European Commission, in cases with an EU-wide impact, and by Member State competition authorities for national cases. It aims to prevent or sanction anti- competitive conduct on the part of business, such as agreements between companies to fix prices or restrict choice, which negatively affect consumers. However, the implementation of sustainability objectives often benefits from or requires collaboration, such as industry-wide initiatives aimed at improving human rights or environmental outcomes. Such arrangements are likely to result in higher prices, since responsibly sourced products usually come at a cost. However, they are not aimed at increasing companies’ revenues or shares of the market; rather, they seek to internalise environmental or social externalities, and deliver public goods. Competition law, however, forbids agreements between competitors where they produce economic harm to consumers by increasing prices. This can cause real problems for sustainability outcomes. For example, in the ‘Chicken of Tomorrow’ case in the Netherlands in 2015, government brokered an industry-wide agreement to improve living standards for chickens; parties to the agreement agreed to ban chicken meat produced in a certain manner from supermarkets and to impose minimum standards providing chickens with a better life. The Dutch competition authority, however, considered the agreement to be anti-competitive, arguing that the benefits in terms of animal welfare did not outweigh the adverse effect on competition which had led to price increases.76 Research conducted by the Fairtrade Foundation on the grocery sector in the UK suggested that while companies did not perceive any regulatory barriers to collaborative discussions on issues such as child labour, deforestation or low productivity, they strongly felt that competition law placed severe constraints on discussions about low farmer incomes and wages.77 Due to the close relationship between incomes and prices, discussions on low farm-gate prices and farmer incomes were a highly sensitive issue that often led businesses to preclude any discussion on the subject whatsoever. Interviewees for the Fairtrade study described situations in which they had felt obliged to cite strict disclaimers at the beginning of meetings forbidding any reference to prices in the supply 76 See discussion in Giancarlo Piscitelli and Anna Gerbrandy, ‘The sustainability dilemma in competition law’, Great Insights, 8:1, Winter 2018/19 (European Centre for Development Policy Management). Sophie Long, Competition Law and Sustainability: A study of industry attitudes towards multi-stakeholder collaboration in the UK grocery sector (Fairtrade Foundation, 2017). 77

  38. 38 chain, or where other companies had refused to attend meetings on the topic, citing competition law risks. Overall, they felt that there was very limited space to discuss even pre-competitive efforts to address low farm-gate prices. Interviewees called for clearer guidance on current rules, and a more supportive policy environment, to help unblock the barriers to progress they were experiencing. The study concluded that while competition law was rightly designed to protect consumers from price- fixing and other harmful practices, ‘unless farmers and workers receive higher incomes and wages, the medium to long-term supply of commodities such as cocoa and bananas will be put at risk by loss of labour supply and the worsening impacts of climate change. These impacts will ultimately harm UK consumers, raising retail prices and even potentially compromising the nation’s food supply chains.’ Application to cocoa Companies can of course take unilateral action to raise the prices they pay for commodities; indeed, some of the smaller ethically branded cocoa companies do this already. Understandably, however, most companies, and particularly the larger ones, are not likely to do so, fearing the competitive disadvantage that could result from an increase in their cost base, leading to a loss of market share to competitors not paying higher prices – and thereby not achieving the objective. It seems likely that a significant number of retailers, brands and traders will need to work together to take action on the issues of low farm-gate prices, incomes and wages across cocoa supply chains. Existing competition law clearly has a chilling effect on such collaborative initiatives. There is accordingly a strong case for it to be given a broader interpretation – perhaps through clarifications issued by the competition authorities, or perhaps by rewriting the relevant legislation – to allow businesses to collaborate for long-term sustainability purposes. While some of this could possibly be implemented by national competition authorities, it would be carried out more effectively at the EU level, which would involve providing guidance or policies that would clarify the application of the prohibition and the exemption criteria under Article 101 of the Treaty on Functioning of the European Union.78 Clearly, this has the potential to affect any supply chain where low prices are a problem, which includes cocoa. The implementation of this option may also be necessary for the due diligence approaches discussed below to be able to address fair prices to be paid to farmers. Photo: Ecosystem Services for Poverty Alleviation / Flickr 78 For a longer discussion, see Tomaso Ferrando and Claudio Lombardi, EU Competition Law and Sustainability in Food Systems: Addressing the Broken Links (Fair Trade Advocacy Office, March 2019).

  39. 39 8.2 Option F: Corporate reporting for cocoa supply chains Under this option, companies would be required to scrutinise their supply chains for the risk of their handling illegally produced cocoa, or cocoa not produced to high social and environmental standards, and to publish regular reports on the extent of the risk. Existing models: reporting requirements Many companies already have in place various reporting requirements for forest risk commodities whose production may be linked to deforestation. The main example is probably the voluntary reports compiled by CDP (formerly the Carbon Disclosure Project), a not-for-profit organisation which works on behalf of investors to encourage companies to investigate the risks and opportunities in their supply chains from four commodities responsible for the majority of deforestation and forest degradation: cattle products, palm oil, timber products and soy.79 Many larger companies, especially those with voluntary commitments to eliminate deforestation from their supply chains, already publish relevant information, including, for example, the mills from which they source palm oil (and sometime the individual suppliers to the mills) or the factories in which their products are manufactured. Certification schemes such as those for timber, palm oil or cocoa often include requirements for chains of custody enabling the product to be traced back through its supply chain. The EU Timber Regulation (see Section 8.3) requires all companies first placing timber products on the EU market to have information on their source, including not just the country of origin, but in some circumstances the sub-national region and concession of origin. Other reporting requirements are not linked specifically to commodities but also provide possible models. In the last seven years, California, the UK and Australia have all adopted legislation requiring companies to scrutinise their supply chains for the presence of modern slavery or human trafficking. In the UK, Section 54 of the Modern Slavery Act 2015 requires all companies operating in the UK above a specified annual turnover (currently £36 million) to produce an annual slavery and human trafficking statement, including the steps the organisation has taken to ensure that slavery and human trafficking is not taking place in any of its supply chains, and in any part of its own business, or a statement that the organisation has taken no such steps. This is a requirement for transparency rather than action; the company is not required to guarantee that its entire supply chain is slavery-free, but it is assumed that the requirement to publish the statements will raise the profile of the issue and facilitate pressure on the company to improve its performance, for example from campaigning groups or shareholders. By May 2019, 9,500 statements had been published (the government estimates that about 9,000–11,000 companies are required to report under the Act), though many of them had not met all the minimum requirements set out in the legislation.80 An independent review of the supply chain transparency provisions published in January 2019 concluded that the lack of enforcement and penalties, as well as some confusion surrounding the reporting obligations, had helped to render the Act less effective than had been hoped.81 It recommended several amendments to the provisions; the Australian Modern Slavery Act 2018 had already learnt from the shortcomings of the UK experience and may prove more effective in practice. 79 80 81 See https://www.cdp.net/en/forests. See https://www.modernslaveryregistry.org. Frank Field MP, Maria Miller MP and Baroness Butler-Sloss, Independent Review of the Modern Slavery Act – Second interim report: Transparency in supply chains (January 2019). Also see Full Disclosure: Towards Better Modern Slavery Reporting (International Corporate Accountability Roundtable and Focus on Labour Exploitation, March 2019).

  40. 40 At the EU level, the EU Non-Financial Reporting (NFR) Directive (2014/95/EU) already applies to companies which have more than 500 employees and are of significant public relevance because of the nature of their business, size or corporate status, including listed companies, banks, insurance companies and other companies designated by national authorities as public-interest entities; there are an estimated 6,000 in total in the EU. They are required to report on how the company’s performance, position and activities affect environmental, social, employee, human rights, anti- corruption and bribery issues. Information should cover the company’s policies on each issue and their outcomes, its due diligence processes, principal risks, the business relationships, products and services which are likely to cause adverse impacts in those areas of risk, and a description of how the company manages the principal risks. Since 2018, national legislation implementing the Directive has required companies to start producing reports. Two analyses of the initial sets of reports suggested that they still had a long way to go in achieving the aims of the Directive. One concluded that while the vast majority of companies acknowledged in their reports the importance of environmental and social issues for their business, ‘in only 50 per cent of cases for environmental matters and less than 40 per cent for social and anti-corruption matters, this information is clear in terms of concrete issues, targets and principal risks. The general information that most companies provide does not allow readers to understand their impacts and by extension their development, performance and position, as required by the NFR Directive.’82 The analysis concluded by recommending that guidance was needed to ensure reports focused on clear indicators and specific risks and greater transparency around high-risk supply chains. The other analysis, focusing on human rights, was similar, concluding that companies were ‘severely struggling with their human rights reporting obligations’.83 It recommended using the UN Guiding Principles reporting framework to ensure human rights due diligence. Application to cocoa The application of a reporting requirement to cocoa should be relatively straightforward in principle, though the extent of implementation would depend on the topics and the level of detail of risks which would be required. Many of the major cocoa and chocolate companies already report some of this information through their use of certification schemes or their own initiatives, and some is covered by the EU Non-Financial Reporting Directive. As noted in Section 3, this information is more difficult to acquire in cases where companies purchase through traders or middlemen in the countries of origin instead of direct from the farmers. Such a reporting obligation could create an incentive to increase the proportion of direct purchasing, a development which appears to be gradually under way in any case. The requirement would be more effective if a compliance mechanism was in place; in the UK Modern Slavery Act, for example, the government puts almost no effort into ensuring compliance with the supply chain reporting obligation. The impact of the reports would depend, in the final analysis, on whether their publication led to any change in behaviour amongst companies in the supply chain, whether triggered by their own improved scrutiny, or by the perception of a threat – or actual damage – to their brand from external scrutiny. However, as just noted, most of the large cocoa and chocolate companies already report significant amounts of information, and this has not succeeded in resolving the problems. Although a legislative reporting requirement could help to standardise the information collected, and create a level playing field for company reporting, it seems unlikely that it would result in radical change in the sector. 82 83 2018 Research Report: The state of corporate sustainability disclosure under the EU Non-Financial Reporting Directive (Alliance for Corporate Transparency Project, 2019). A Human Rights Review of the EU Non-Financial Reporting Directive (European Coalition for Corporate Justice, 2019).

  41. 41 8.3 Option G: A due diligence regulation for cocoa placed on the EU market This option envisages a regulation in which enterprises importing cocoa or cocoa products, including chocolate, to the EU would be required to have in place a system of due diligence designed to minimise the risk of their handling illegally produced cocoa, or cocoa not produced to the social and environmental standards discussed in Section 5. Existing models: EU Timber Regulation and Conflict Minerals Regulation Like the VPAs discussed in Section 5.1, this option is modelled partly on one of the elements of the FLEGT Action Plan: the EU Timber Regulation ((EU)995/2010).84 Entering fully into operation in 2013, the Timber Regulation has two elements: it prohibits the placing of illegally harvested timber and timber products on the EU market, and it requires operators who first place such products on the market to implement a system of due diligence in order to minimise the risk of them doing so. Operators trading in such products further down the supply chain, after they have been first placed on the market, are required to keep records of who they buy from and sell to. Due diligence systems must include means of ensuring access to information on the products and a process of analysing and mitigating against the risk of placing illegal products on the market. This includes obtaining full information on the products, including their legal status and the countries, regions and sometimes forests of origin. The higher the risk of illegal behaviour in the place of origin, the greater the degree of knowledge the operator must have on the product and its chain of custody. The presence of a document verifying legality – such as a document issued by a timber certification scheme – is helpful, but not necessarily conclusive, so reliance on certification by itself will not satisfy the Timber Regulation. There is no minimum company size for the application of the Timber Regulation: all timber operators first placing timber products on the EU market are covered, however small. Member States nominate ‘competent authorities’ to check compliance by operators, and to receive ‘substantiated concerns’ from third parties such as NGOs, where illegal behaviour or non-compliance is suspected. Any timber products accompanied by a permit issued under the Convention on International Trade in Endangered Species (CITES) or a FLEGT licence issued by a VPA partner country are considered to have been legally harvested: this provides a ‘green lane’ for access to the EU market (see Section 6.1). The EU Timber Regulation has proved an important complement to the VPAs, helping to reassure VPA partner countries that imports to the EU from non-VPA countries will be subject to scrutiny and that there is a direct trade benefit from agreeing a VPA and implementing the licensing scheme; this accordingly provides an incentive to negotiate and implement a VPA. To date, the prohibition element of the EU Timber Regulation has not been used in a prosecution; proving the illegal origin of products is difficult and would require corroborative information from the country of harvest. Some observers have argued that this element may therefore act more as a signal of intent than as a practical enforcement tool. The due diligence element is likely to have a more lasting impact, requiring companies to scrutinise their supply chains closely, helping to make information about both timber supply chains and company practices more visible and requiring companies to address the risks. The framers of the Timber Regulation accepted that companies would not always be able to guarantee that every single piece of timber they handled had not been illegally produced; their obligation was to make the best efforts they could to put in place due diligence systems designed to minimise the risk. 84 See http://ec.europa.eu/environment/forests/timber_regulation.htm.

  42. 42 To date there have been no studies analysing exactly how companies are responding to the EU Timber Regulation. Anecdotal evidence suggests that while importers may be tending to avoid high-risk sources, there appears to have been no overall impact on volumes of tropical timber imported into the EU.85 An evaluation published in 2016 observed (unsurprisingly) that large companies, particularly those with experience of responsible sourcing policies, were finding it easier to implement the Timber Regulation than smaller ones, many of whom were still unaware of their obligations; at the time the review was carried out, the Timber Regulation was still quite new.86 The other existing model is the EU Conflict Minerals Regulation ((EU)2017/821).87 Passed in 2017, this will enter fully into force in 2021. It aims to target the trade in gold, tin, tantalum and tungsten from areas affected by or at high risk of conflict. Any enterprise importing these minerals to the EU, whether as ores, concentrates or processed metals, will be required to exercise due diligence in their supply chains, with the aim of ensuring that the minerals and metals they buy and sell are not funding armed groups or security forces in areas of conflict. The specific guidance for the process of due diligence is drawn from the OECD’s Due Diligence Guidance for Responsible Supply Chains from Conflict-Affected and High-Risk Areas, originally agreed in 2011 and revised to cover these minerals in 2013.88 in August 2018 the European Commission published guidelines to help enterprises identify high-risk areas, and is in the process of producing a global list of responsible smelters and refiners. The procedure is similar to that set out in the EU Timber Regulation, requiring companies to establish strong management systems, identify and assess risks in the supply chain and design and implement a strategy to mitigate the risks. It is different from the EU Timber Regulation in also requiring an independent third-party audit of the due diligence system and an annual report from each company. The Regulation will not apply to companies below a set threshold of volume of imports, specified in the Regulation and different for each mineral and metal – the aim is to place requirements on roughly the top 80 per cent of imports. Downstream companies processing the metals into finished products have no obligations under the Regulation, but are encouraged to use reporting and other tools to make their due diligence practices more transparent. Unlike the EU Timber Regulation, there is no prohibition on placing any of the minerals sourced from conflict areas on the EU market. The Conflict Minerals Regulation was based on experience in the US, where Section 1502 of the Dodd-Frank Wall Street Reform and Consumer Protection Act, passed in 2010, requires US listed companies that manufacture products containing conflict minerals (tin, tantalum, tungsten and gold) in their supply chain to disclose annually whether any of them originated in the Democratic Republic of the Congo or an adjacent country, and if so, to report on the steps they have taken to exercise due diligence. Analysis suggests that while it may have had an impact on sourcing of tin, tantalum and tungsten, it has not affected gold (which is less bulky and easier to smuggle), and has not helped to reduce conflict over the control of gold mines.89 Some observers have argued that more effort needs to be put into creating conditions for the responsible sourcing of minerals as well as trying to exclude irresponsibly sourced minerals, for example by addressing governance problems in the mining sector, setting up reliable mineral-tracing systems, strengthening public services assisting artisanal miners, providing assistance to mining cooperatives, and addressing the impunity of national military involvement in illegal mining.90 85 See, e.g., Holger Weimar and Georg Becher, ‘Timber Trade and the EUTR’ (Thünen Institute of International Forestry and Forest Economics), presentation at conference on ‘Timber trade and the conservation of forests, Hamburg, October 2018. Evaluation of Regulation (EU) No 995/2010 of the European Parliament and of the Council of 20 October 2010 laying down the obligations of operators who place timber and timber products on the market (the EU Timber Regulation), Accompanying the document {COM(2016) 74 final} {SWD(2016) 33 final} (European Commission, 18 February 2016). See http://ec.europa.eu/trade/policy/in-focus/conflict-minerals-regulation/ Available at https://www.oecd-ilibrary.org/governance/oecd-due-diligence-guidance-for-responsible-supply-chains-of-minerals-from-conflict-affected-and-high-risk-ar- eas_9789264185050-en See, for example, Nik Stoop ,Marijke Verpoorten and Peter van der Windt, ‘More legislation, more violence? The impact of Dodd-Frank in the DRC’ PLoS ONE 13 (8), 2018. Accompanying Measures to the EU Regulation on Responsible Mineral Sourcing: Towards an improved governance of the artisanal mining sector in the DRC (European Network for Central Africa, March 2017). 86 87 88 89 90

  43. 43 Application to cocoa It should in principle be possible to adapt the due diligence approach to cocoa. This would have the effect of ensuring that all companies registered in, operating in, or selling into the EU market control their supply chains and ensure that cocoa and cocoa products within them are produced to higher standards. It would level the playing field, ensuring that all companies faced the higher costs following from, for example, establishing full traceability, avoiding deforestation or paying higher prices. A proposal for a due diligence regulation for forest risk commodities was one of the main options included in the European Commission’s 2018 feasibility study for policy options to address the EU’s impact on deforestation (see Section 3.3). Several campaigning groups have called for a human rights due diligence regulation for cocoa specifically, or for forest risk commodities more generally.91 A number of choices would need to be made in drawing up such a regulation, including: • Should it be based on the legality of production of cocoa (like the EU Timber Regulation) or on a wider set of criteria, such as those listed in Section 5, covering human rights, labour and environmental issues, and fair prices, as well as legality? • If it is based, partly or wholly, on legality, what are the relevant laws and regulations in force in the producer countries? • What information will companies need to have access to? For example, the details of the source of the cocoa, the relevant laws and regulations, other information on the conditions of production. • What role could the cocoa certification systems, and company programmes for responsible or sustainable cocoa, play in minimising the risk of non-compliance? • Would companies need to publish regular reports on their due diligence measures (as in the Conflict Minerals Regulation)? • Should the regulation apply to all companies placing cocoa or chocolate on the EU market or just on those first placing the products on the EU market (as in the EU Timber Regulation)? • Should there be a minimum company size (whether measured by employment, turnover or volumes of imports or sales) below which the regulation would not apply? Or could there be less onerous requirements for smaller enterprises? • Should the regulation include a prohibition on sales of cocoa products not meeting the criteria (as in the EU Timber Regulation), or only the due diligence requirement (as in the Conflict Minerals Regulation)? The extent to which the application of such a regulation would affect cocoa production on the ground depends on a range of factors: to what extent can importers changing their behaviour improve social and environmental standards; would additional action by producer-country governments, for example to improve law enforcement, be needed; would the companies affected be able to persuade those governments to make improvements; and would additional assistance from consumer-country governments be necessary? The introduction of full traceability throughout the supply chain seems 91 See, e.g., Fountain and Huetz-Adams, Cocoa Barometer 2018 (for cocoa); joint submission to the European Commission’s Roadmap on Deforestation and Forest Degradation from eight NGOs (six based in developing countries), January 2019 (for forest risk commodities); https://ec.europa.eu/info/law/better-regulation/initiatives/ares-2018-6516782/ feedback_en?p_id=343654.

  44. 44 likely to be necessary. The extent to which importers of cocoa to the EU market are able to move their suppliers, either between or within countries, in response to the regulation is a key question. Another question is the compatibility of this kind of due diligence regulation with WTO trade rules. This seems unlikely to be a problem in practice; neither the EU Timber Regulation nor the Conflict Minerals Regulation have ever been subjected to a WTO challenge. Like these two regulations, a due diligence regulation for cocoa would be non-discriminatory, applying to products wherever they are sourced from and to operators regardless of nationality (as long as they place products on the EU market); would not create unreasonable barriers to international trade; and the objectives to which it would apply – combating child labour and deforestation – are the subject of many internationally agreed declarations and initiatives.92 Although a due diligence regulation could be implemented independently of any bilateral cocoa agreements (see Section 6.1), it would work well together with them, in the same way as the EU Timber Regulation complements the VPAs. The VPAs provide means for consumer and producer country governments to work together to improve governance and law enforcement (which may be difficult for company-level action to achieve by itself), while the EU Timber Regulation provides a mechanism to ensure that timber from non-VPA countries cannot enter the EU market without regulation, and in turn creates an incentive for countries to sign VPAs. The same combination of measures and incentives could be created for cocoa. Should the regulation apply only to cocoa, or to forest risk commodities more widely? Clearly, it is undesirable for the EU and its Member States to have to legislate for commodities one by one, but equally, the situations of the various supply chains are very different. Compared to several other forest risk commodities, the cocoa sector has a longer history of private-sector initiatives, a higher penetration of certification and probably a greater recognition amongst companies of the need for action. In addition, the EU is the world’s single largest importer of cocoa. It could be possible for a general forest-risk commodity due diligence regulation to be legislated for, but for its application to be phased by commodity, starting with those where it is most likely to be implemented with success – including cocoa – and moving on later to those where more preparatory work would be needed. Photo: MinIsterio de Agricultura y Ganadería de Ecuador / Flickr 92 For a longer discussion, see Brack, WTO Compatibility with EU Action on Deforestation and Dylan Geraets and Bregt Natens, The WTO Consistency of the European Union Timber Regulation (Leuven Centre for Global Governance Studies, 2013).

  45. 45 8.4 Option H: A general corporate due diligence requirement This option also rests on a due diligence approach, but is much broader than the previous option, placing a general requirement on companies to exercise due diligence in respect of environmental harm and human rights and labour abuses, in all their operations and supply chains. Unlike option G, it is not specific to cocoa, or any other commodity, but applies to all products and services in the company’s operations and supply chains. Existing model: French Devoir de Vigilance law This option is based on recent developments in corporate responsibility and transparency in company supply chains, as summarised in Section 4. National legislation that is now beginning to appear includes the Dutch Child Labour Due Diligence Act adopted in May 2019, which requires companies to identify, prevent and if necessary, address the issue of child labour in their supply chains. Legislation on mandatory human rights due diligence is currently under discussion in Switzerland. In Finland the new government formed after the April 2019 election has committed to introduce domestic legislation on human rights due diligence and pursue it at the EU level. And in Germany the government has promised to legislate if less than 50 per cent of German enterprises with more than 500 employees adopt human rights due diligence approaches by 2020. In February 2017 France adopted a corporate Devoir de vigilance des sociétés mères et des entreprises donneuses d’ordre (Due diligence of corporations and main contractors) law applying to companies incorporated under French law with more than 5,000 employees in France or 10,000 world- wide.93 Companies subject to the legislation (an estimated 150–200) must exercise due diligence in seeking to identify and avoid human rights violations, breaches of fundamental freedoms, violations of health and safety rights and environmental damage. Building on the UN Guiding Principles on Business and Human Rights, this includes the identification of risks, procedures for regular assessments of subsidiaries, sub-contractors and suppliers, actions to mitigate risks or prevent serious harm, and mechanisms for alerts and monitoring. Furthermore, the companies must implement a diligence plan setting out these risks and procedures and publish annual reports on progress. In March 2017 the proposed penalties for failing to prepare such a plan – fines of up to €10 million, or up to €30 million if the failure to develop a plan led to injuries that could otherwise have been prevented – were struck down as unconstitutional.94 The general due diligence obligation and the requirement to implement a diligence plan remain, however, as do civil liability mechanisms in case of failure to implement the plan or if there are weaknesses in it. The state plays no role in compliance; the civil liability mechanisms must be pursued by third parties such as NGOs. Initial implementation of this law is still under way, and companies are beginning to publish their reports. Various surveys and analyses of the reports suggest that while the majority of the companies affected have made commitments to due diligence, they have not provided much detail on the way in which they have identified and addressed human rights risks.95 Risks were often assessed only in regard to the company rather than to the environment or human rights, and without regard to specific high-risk countries or regions; many appeared to be one-off analyses with no information available about how they would be reviewed or updated. Risk mitigation strategies were not always included, and risks relating to sub-contractors were often omitted entirely. On 93 94 95 http://www.csrandthelaw.com/2017/02/28/french-national-assembly-adopts-corporate-duty-of-vigilance-law/ https://www.jdsupra.com/legalnews/french-companies-must-show-duty-of-care-56981/. See Strategies for Responsible Business Conduct (PwC for Netherlands Ministry of Foreign Affairs, December 2018); Loi sur le devoir de vigilance des sociétés mères et entreprises donneuses d’ordre – Année 1: Les entreprises doivent mieux faire (Sherpa, February 2019).

  46. 46 a more positive note, some of the companies interviewed felt that the legislation had helped to highlight the role of the UN Guiding Principles and due diligence approaches more broadly. Having a better understanding of the approaches would help companies integrate responsibilities that were often dispersed widely throughout the company, meaning that the quality of reports can be expected to improve. Several companies requested detailed guidance on how to map and address all the human rights and environmental issues arising in their supply chains. What exactly is meant by companies’ due diligence obligations, and their liability if human rights abuses or environmental harms arise from their operations and supply chains, is of course a central issue. It is not clear whether this will only be clarified if and when court cases take place, or whether the government may decide to issue more specific guidance. Application to cocoa The Devoir de Vigilance law echoes the logic of the OECD’s Due Diligence Guidance for Responsible Business Conduct. As noted in Section 4, the OECD Guidance sets out a due diligence framework to be used by enterprises to avoid and address adverse impacts in their operations, supply chains and business relationships. It covers human rights, employment and industrial relations, environment, bribery and extortion, consumer interests, and disclosure of information. The French law makes a due diligence approach compulsory and adds the requirement to publish a plan to mitigate risks or prevent serious harm, including mechanisms for alerts and monitoring. Extending this approach to all EU Member States would be a major undertaking, but one that was called for in the French deforestation strategy published in 2018; the Finnish and German governments have also indicated interest in pursuing EU-level legislation of this kind. Applying the legislation at EU level would have significant potential benefits to supply chains for all kinds of products and commodities associated with negative social and environmental impacts. It has obvious application to companies handling cocoa and cocoa products, as it covers human rights violations, such as child labour, and environmental damage such as deforestation. It would level the playing field, ensuring that all companies faced the higher costs following from, for example, establishing full traceability, avoiding deforestation or paying higher prices. Introducing it at the EU level would avoid the emergence of a patchwork of different approaches in different EU Member States. The number of companies to which the Devoir de Vigilance law applies is limited. This is because the law applies an extensive set of requirements which smaller companies could struggle to meet (though smaller companies also generally have much simpler supply chains, with a smaller range of suppliers and sub-contractors). The extent to which any new EU law would need to be limited by size of company, whether measured by employment, turnover or volumes or imports or sales, would be a matter for discussion. The French law also applies only to companies registered in France, but several of the main cocoa and chocolate companies are not based in the EU. An EU law could apply more broadly, to companies registered or operating in the EU (the UK Modern Slavery Act applies to companies registered or operating in the UK), and perhaps to any companies placing products on the EU market (as in the EU Timber Regulation). Clearly, the issues on which the companies will be required to conduct due diligence must be carefully specified. While there are well-recognised international agreements on human rights and labour rights, this may prove more problematic for environmental issues, where there is no single international instrument but rather a wide range of agreements and goals dealing with specific environmental matters. However, the OECD’s Due Diligence Guidance for Responsible Business Conduct contains some detail, and the OECD – Food and Agricultural Organisation (FAO) Guidance for Responsible Agricultural Supply Chains contains much more. The latter recommends enterprises adopt standards that cover human rights, labour rights, health and safety, food security

  47. 47 and nutrition, tenure rights over and access to natural resources, animal welfare, environmental protection and sustainable use of natural resources, governance and technology and innovation. It also has a number of recommendations for cross-cutting issues such as disclosure, consultation and resolution of grievances. Forests are mentioned specifically in the standards on tenure rights over and access to natural resources, and on environmental protection and sustainable use of natural resources. Neither set of guidelines, however, include criteria guaranteeing fair prices, incomes or remuneration, possibly because of the fears of anti-competitive behaviour discussed in Section 8.1. If such a law was adopted at EU level, it may be possible for the European Commission to develop even more detailed guidance relating to specific commodities and supply chains, such as cocoa. This could also deal with the aspect of compensation for damage suffered, which is a key element of the UN Guiding Principles, perhaps building on existing systems, e.g. for remediation of child labour. The extent to which the application of such a law would affect cocoa production on the ground depends on much the same factors discussed in Section 8.3: to what extent can companies changing behaviour improve social and environmental standards amongst producers; would additional action by producer-country governments, for example to improve law enforcement, be needed; would the companies affected be able to persuade governments to make improvements; and would additional assistance be necessary from consumer-country governments? As with Option G, combining such a law with bilateral agreements with producer countries (see Section 6.1) could help address some of these issues. Photo: Francisco Sierra

  48. 48 9 Conclusions and recommendations Option A: Bilateral agreements with cocoa-producing countries Producer Country Focussed Options Option B: ‘Carding’ system Option C: Public procurement policy EU-market-focused options Option D: Regulation of sales in the EU market Option E: Competition law reform Business-focused options Option F: Corporate reporting for cocoa supply chains Option G: A due diligence regulation for cocoa placed on the EU market Option H: A general corporate due diligence requirement As the major global consumer of cocoa from Ghana and Côte d’Ivoire, the EU has a critical role to play in improving the problems of the cocoa sector outlined in Section 2. There is, however, no simple or single solution. Rather, action is required at many levels and by many actors. This paper is designed to stimulate debate on the way forward, through outlining potential options and discussing their advantages and disadvantages; it is hoped that it will encourage further and more detailed exploration of the many questions these options raise. This section suggests a possible combination of options the EU could take to encourage the development of a sustainable cocoa supply chain which meets the aims set out in Section 5, including respect for human rights and labour standards, payment of fair prices, protection of the environment and improvements in governance and law enforcement. It includes Option A (bilateral agreements) plus either of Options G or H (due diligence requirement) as the core elements.Option F (review of competition law) would add a valuable additional component. A possible package Although the bulk of this paper considers options for the EU to regulate its market, or businesses operating within it, in many ways the most important measures are those that must be implemented within the countries of production: improving governance, enforcing laws, investing in education and alternative livelihoods, clarifying land and tree tenure, improving transparency and establishing full traceability systems, among many other steps. Clearly, action by producer- country governments is essential – and many of these measures are referenced in commitments those governments have already made, for example through the Cocoa and Forests Initiative. The EU can assist by providing financial and capacity-building assistance; but many of them require commitment to deep-seated reforms of policy and legislation. The experience of the Voluntary Partnership Agreements for timber products have shown how bilateral agreements aimed at these problems, accompanied by assistance from the EU, can help to trigger lasting change, improving the transparency and accountability of forest administration, increasing participation by civil society and other stakeholders in forest and land governance, improving the recognition

  49. 49 of community rights, exposing corruption and stemming illegal behaviour. The Cocoa and Forests Initiative, which does not include consumer-country governments and does not contain processes to improve governance or open the cocoa sector to stakeholder participation, is not set up to achieve these aims. Adapting these bilateral agreements to the cocoa sector – as proposed in Option A – should be possible, not least because the two key countries, Ghana and Côte d’Ivoire, already have or are negotiating a timber VPA. This option should therefore form part of the optimum package of measures to be pursued by the EU. A modification of the carding system used in the IUU Fishing Regulation (as described in Section 6.2), could form a component of these agreements. Bilateral agreements themselves, however, will not directly affect the behaviour of companies in the supply chain, so there is a strong argument for regulation aimed at cocoa placed on the EU market or at companies operating within it. Of the options examined here, the two requiring companies to exercise due diligence – either Option G, requiring due diligence in the cocoa supply chain (or, more likely, a regulation covering a range of forest risk commodities, potentially with a phased introduction by commodity) or Option H, requiring companies to exercise due diligence to avoid human rights abuses and environmental harm across all their operations and supply chains, are likely to have the greatest impact. Both are designed to affect company behaviour, ensuring that all companies registered in, operating in, or selling into the EU market control their supply chains and ensure that cocoa and cocoa products within them are produced to higher standards. Both would level the playing field, ensuring that all companies faced the higher costs following from, for example, establishing full traceability, avoiding deforestation or paying higher prices to increase farmer household incomes. Introducing either of them at the EU level would avoid the emergence of a patchwork of different approaches in different EU Member States. Each has its advantages and disadvantages. With regards to option G, the EU already has experience of commodity-specific due diligence regulation through the EU Timber and Conflict Minerals Regulations. Because such legislation would apply to a much narrower range of behaviour and a smaller number of companies than option H, it would be less difficult to legislate for and probably faster to show results, since the behaviour required of companies would be less difficult to specify in detail. The company-based due diligence approach described in Option H is only now in operation in France, and only for a relatively small number of companies – though legislation like this is being discussed much more widely. It has the advantage of covering a very broad range of activities, not just one commodity supply chain (or several supply chains), with correspondingly broader impacts. It avoids the need to legislate topic by topic. On the other hand, since it is much more broad- ranging – and more innovative – it is likely to be harder to legislate for and slower to implement in full, and its application would probably be restricted to just the largest companies, at least to start with. As the experience of the French law has shown, detailed guidance is likely to be necessary; this could be developed specifically for the cocoa supply chain, and perhaps other commodities. As well as these considerations, the experience of the VPAs has shown that countries need incentives to negotiate and implement them, and the best incentive is provided by enhanced market access (as in the EU Timber Regulation, which affords easier access for timber licensed under the VPAs). Of the two due diligence options, Option G, the commodity-specific regulation, fits this model better; cocoa licensed under the agreement (or originating in countries or regions with a ‘green card’, if using a variation of the carding system discussed in Section 6.2) could be deemed to have met the requirements of the regulation and thereby given an easier route into the EU market. It is difficult to see how the broad company-based due diligence approach of Option H could provide this incentive.

  50. 50 Other options could add value to the overall package. In particular Option E, the review of competition law, should help companies address the problem of low prices for producers, and could be implemented more quickly than most of the other options; it may also be necessary to enable the inclusion of criteria relating to prices and remuneration in a due diligence regulation. Option C, including responsibly sourced cocoa in public procurement policies, could also be introduced quickly, as in general it requires no new legislation, but is likely only to have a limited impact. The remaining options, D (sales ban) and E (corporate reporting) are of less value. Impacts Clearly, many details of whatever package is settled upon would need to be resolved; this paper has only touched on the headline features of the different options. It would be critical to assess the likely impacts on all stakeholders in the supply chain, including the cocoa farmers (mostly smallholders) and traders in the countries of production, and their families and communities; producer-country governments; large and small cocoa trading and processing companies and chocolate companies operating within the EU; EU governments; and EU consumers. A full exploration of this topic is beyond the remit of this paper, but some pointers can be derived from the existing models reviewed above: • Based on the experience of the VPAs, a VPA-type bilateral agreement for cocoa could lead to significant improvements in governance, law enforcement and transparency, with associated benefits including better protection of tree and land tenure, a reduction in corruption and improved transparency around government decision-making and revenue administration. Also based on the VPA experience, it would require a commitment of diplomatic resources and financial and capacity-building assistance from the EU and its Member States. • If the commodity-level due diligence approach outlined in Option G is adopted, companies trading in cocoa will need to more closely scrutinise their operations and supply chains to ensure the cocoa is produced in compliance with the criteria listed in Section 5. If the broader company-level due diligence approach outlined in Option H is followed, a much wider range of activities will need to be investigated. Either approach will involve costs, which SMEs are likely to find more difficult to meet – although they may be excluded from the coverage of the regulation or given lighter requirements. • Among other things, either type of regulation will almost inevitably require the extension of full traceability mechanisms for the cocoa supply chain (this may in any case be a component of a bilateral agreement). • Smallholder farmers in the producer countries would clearly benefit from an increase in prices and, more broadly, from the improvements in governance and law enforcement outlined above, which could help to create a stable environment in which they have the confidence to invest in their farms, for example by replanting cocoa trees or developing other crops. A reduction in poverty would be likely to lead to a fall in the incidence of child labour. • However, there is the real danger that some farmers or some regions may be abandoned by companies seeking to source less risky products. It will be important to develop mechanisms to incentivise continued company engagement, and investment in improvements, together with means of monitoring them.