Introduction
The proliferation of ‘sustainability’ standard initiatives constitutes a new form of regulation where private actors such as NGOs and industry associations are involved in negotiating standards for labour and the environment, and for monitoring compliance with these standards. Companies are increasingly signing up to these new systems of regulation which have expanded rapidly across industries of interest to critical Western consumers such as garments, toys, forest products, oil and gas, agricultural products, chemicals and electronics (Gereffi et al. 2001, O'Rourke 2006, Utting 2008, Egels-Zandén 2009).
These new and private systems of regulation, however, remain highly disputed, particularly since the intended positive impact on producers, workers and the environment is by no means guaranteed. Standard initiatives have been criticised for implementing a Northern agenda on southern-hemisphere producers and workers, for not being sensitive to local specific conditions, and for providing consumers with a false sense of problem solving because they fail to reach marginalised workers, such as temporary or migrant workers, and fail to address more embedded issues like discrimination or the lack of freedom of association (Barrientos et al. 2003, Utting 2005, 2008, O'Rourke 2006, Barrientos and Smith 2007, Blowfield and Dolan 2008). Others argue that such private initiatives are more efficient than traditional labour or environmental regulation and moreover suit the current global production system (see for instance Bernstein 2001 or Carasco and Singh 2003).
The cut flower industry is a particularly interesting example of the emergence and proliferation of standards that seek to regulate the social and environmental conditions of production. The character of cut flower production and trade has set the frame for some highly criticised working conditions in the industry. The largest developing country exporters (Kenya and Colombia) in particular have been favourite targets for campaigns demanding better environmental and social conditions both locally and in Europe and North America. In 2006, developing countries supplied 22% of EU imports of cut flowers and foliage and 60% of US imports, and their share is increasing.1 The seasonal nature of the cut flower trade, with demand peaking at northern-hemisphere festivals such as Valentine's, Mother's day and Easter, and lowest demand during the northern summer, makes labour demand in production highly uneven. Seasonality is a major force behind the employment of large numbers of temporary workers at times of peak demand. The increase in sales to large retailers has further intensified the need for a flexible workforce to meet the ever-changing requirements of retailers, whose orders are often adjusted on the day of delivery (Hale and Opondo 2005, Riisgaard 2009a). The perishability of the product means that workers often have to work long hours to complete critical tasks such as harvesting and spraying, but it is the heavy use of chemicals that constitutes the main health hazard to workers and the surrounding environment and communities.
The industry has reacted to criticisms of the working conditions by adopting a range of private social and environmental standard initiatives since the mid 1990s. Initially, the standards mostly covered technical issues such as chemical usage and environmental management. The social components of standards relating to workers' welfare are a more recent addition (cf. Barrientos et al. 2003). The majority of standard initiatives have been conceived and formulated in Europe, but in recent years a variety of standard initiatives have also been developed in producer countries. In all, at least 20 different social and/or environmental standards (international and national) exist for cut flower export (CBI 2007, Dolan and Opondo 2005, Riisgaard 2007, 2009a).
In the flower industry, we thus have a standard landscape with many schemes existing in parallel. These standards are not neutral mechanisms but institutional mechanisms that differ in the way they seek to implement their objectives and in the impact they have on intended beneficiaries. A growing pool of critical literature on sustainability standards argues that consistent forms of improvements for intended beneficiaries are only achieved by marginal standards applied in niche supply chains (Barrientos and Smith 2007, Macdonald 2007, Nelson et al. 2007, Raynolds et al. 2007). Some also point to a tendency for standard harmonisation attempts to spur convergence around a lowest common denominator (e.g. Mutersbaugh 2005). This paper examines overall trends in cut flower standards aimed at identifying whether (in terms of labour issues) there is a move towards more stringent standards or the reverse. In this regard, the paper examines how more stringent standards have converged around a specific flower base code, the International Code of Conduct for Cut Flowers (ICC). Although the ICC has been interpreted differently by different standard initiatives it has introduced the practice of active inclusion of local NGOs and trade unions in monitoring standard compliance. However, within the more stringent standards there are different interpretations of how to raise the bar. There are also differences in what kind of bar should be raised and who should reap the benefits. One attempt to harmonise existing flower standards, while showing real potential to ‘scale up’ more stringent standards, has so far mainly benefited Dutch growers and workers.
In this paper, I first present theoretical discussions of standard ‘parallelism’ and of the focus and function of standards (section two). In section three, I relate these discussions to flower standards and in section four, I discuss attempts at convergence around the International Code of Conduct for Cut Flowers. This is followed by a conclusion in section five.
The analysis presented in this paper relies on primary material gathered from 21 semi-structured interviews with representatives from flower standard schemes and other industry actors, including follow-up emails and phone communication (see Appendix 1 for details of the interviews). I also draw selectively on fieldwork carried out in the flower sectors of Kenya and Tanzania in 2006 and 2010. Cut flower exports from sub-Saharan Africa (led by Kenya) have increased from approximately US$13 million in 1980 to almost US$300 million in 2007, representing one of Africa's most significant cases of non-traditional export development during the past two decades (UN Comtrade). Kenya is one of the top players in the world cut flower industry being the largest supplier to Europe, and here private social standards have been widespread since the mid 1990s. Tanzania is a newcomer with a much smaller industry that has only recently begun to adopt private social standards.
The function and focus of standards
Risk management and product differentiation
The key role of standards in agricultural value chains2 is to facilitate coordination between firms by transmitting credible information on the nature of products and the conditions under which they are produced, processed and transported (Humphrey and Schmitz 2001, Ponte and Gibbon 2005, Henson and Humphrey 2008). Apart from reducing transaction costs, the literature on global value chains identifies two distinct motives for explicit coordination of activities between enterprises in value chains. The two reasons are risk management and product differentiation. Standards codify arrangements for handling risk and product differentiation (Ponte and Gibbon 2005, Hughes 2006, Henson and Humphrey 2008, Nadvi 2008).
Following the categorisations of Aragrande et al. (2005), Henson and Humphrey thus distinguish two categories of standard in agrofood chains. Risk management standards where the predominant role is to provide a level of assurance that a product is in compliance with defined minimum product and/or process requirements. And product differentiation standards where standards are mainly aimed at differentiating the firm and/or its products in the ‘eyes of the consumer’, most often through a consumer label.
According to Henson and Humphrey, it is possible to categorise most private standards into one of these two categories, although they recognise that some private standards have dual functions. Furthermore, the ratio between risk management and product differentiation is far from static. Henson and Humphrey suggest that there is an upward trend in the minimum level as consumer expectations and/or regulatory requirements are enhanced. Thus, the territory currently governed by product differentiation standards might cede ground to risk management standards, while product differentiation standards will shift their focus to new attributes (Henson and Humphrey 2008).
‘Parallelism’ and ‘forum shopping’
The diverging interests of different actors and the role that standards play in how lead firms are governing value chains make standards highly contested arenas. Consequently, we often see a wide range of standard schemes working in parallel. Abbott and Snidal (2006) term this ‘parallelism’, describing the sometimes supportive, sometimes competitive relations among independent standard schemes working in parallel within roughly the same issue area. As they observe, parallelism opens up opportunities for forum shopping in two senses. First, actors (both NGOs and firms) can shop around for the standard scheme that they deem most favourable for them. Second, standard schemes themselves can ‘shop’ for business participation for example by adapting standard content or rules to attract specific businesses (Abbott and Snidal 2006).
This argument is picked up by Macdonald (2007) in her discussion of sustainability standards in the coffee industry, where she argues that the landscape of parallel standard schemes opens up a large discretionary space for buyers in which they can define the substance and scope of how they respond to activist demands for sustainability in production. According to Macdonald, this has consequences for the ability to realise consistent forms of empowerment for workers and producers in developing countries because the discretionary space allows multiple systems to coexist in parallel, with consistent forms of empowerment only being achieved by those participating in ‘niche’ supply chains (Macdonald 2007).
This position seems to support the notion that parallelism is counterproductive in terms of empowerment and thus leads to a race towards the lowest standards (at least within mainstream markets). Another study, also on coffee standards, by Raynolds et al. (2007) reaches a similar conclusion. Looking at the five major consumer labels in coffee, they ask if these initiatives largely hold the bar (i.e. halting the decline in social and environmental conditions caused by receding state regulations), or if they actually raise the bar (bringing about improvements in social and environmental conditions). While they do find that some standard initiatives raise the bar, they conclude that such standards are likely to be increasingly challenged by those that seek to simply uphold current standards (Raynolds et al. 2007). A similar conclusion is reached by Mutersbaugh in his study on harmonisation and mainstreaming in fair-trade and organic standards. He finds that there is a drive towards global standards and that this spurs a tendency towards ‘a “lowest common denominator”, minimising protections in national standards and displacing more comprehensive network-based standards’ (Mutersbaugh 2005, p. 2039).
Ingenbleek and Meulenberg (2006) examine the strategies that standard organisations pursue to put their sustainability objectives into practice. In their comparison of 10 sustainability standard schemes, they find that many strategic differences between schemes can be traced back to two types of standard organisations: those weighting principles over size, and those weighting size over principles. The most essential difference is thus found in the trade-off between the principles of sustainable production and the size of the programme, i.e. the number of farmers that adopt the standard. This is consistent with Macdonald's (2007) claim that high principles, or in her words, ‘consistent forms of empowerment’ (2007, p. 808), will only be achieved in marginal standard schemes. It also points to the difficulties inherent in trying to scale up certification to ‘high principles’ standards.
According to Ingenbleek and Meulenberg, standard schemes follow either a differentiation or a lowest cost strategy when targeting markets. In a differentiation strategy (pursued by schemes that weight principles over size), a standard scheme enables primary producers to differentiate themselves from mainstream producers on the basis of sustainability, and communicate this to the end consumer. Standard organisations that weight size over principles apply lower sustainability requirements in their standards, and these standards are not communicated directly to the end consumers. This difference in strategies of standard schemes can be related to the difference between product differentiation and risk management standards pointed out by Henson and Humphrey (2008). Thus, standard organisations weighting principles over size use product differentiation standards, and those weighting size over principles use risk management standards.
In terms of the social content and outcome of standards, a distinction has been drawn between standards that focus on ‘enabling rights’ versus standards that focus on ‘protective rights’ (Rodriguez-Garavito 2005). A similar division is developed by Barrientos and Smith (2007), who distinguish between standards which focus on ‘outcome standards’ versus standards that focus on ‘process rights’. Process rights (or enabling rights), for example the principles of freedom of association and non discrimination, describe intrinsic principles of social justice that enable workers to claim their rights. These process rights provide a route to the negotiation of and access to other entitlements and specified conditions of employment, such as a health and safety policy, minimum wages, working hours and deductions for employment benefits such as health insurance and pensions. These entitlements and specified conditions of employment are labelled outcome standards (or protective rights) (Barrientos and Smith 2007).
In a comprehensive study of the effects of social standards amongst suppliers to members of the Ethical Trading Initiative (ETI)3, it was found that while standards were having an effect on outcome standards, they were having little or no effect on process rights, and furthermore failed to reach the most marginalised (often female) workers such as casual, migrant or subcontracted workers (Barrientos and Smith 2007, see also Nelson et al. 2007 for similar findings). The fact that most standards have more impact on outcome standards (protective rights) than process rights (enabling rights) is related to the way social standards are audited and reflects the dominance of a technical compliance perspective characterised by checklist auditing and self-assessments which is less capable of identifying problems related to enabling rights (Barrientos and Smith 2007). Conversely, the use of Participatory Social Auditing methodologies is argued to facilitate such identification and advance enabling rights (Auret and Barrientos 2004, Dolan and Opondo 2005). This methodology is based on thorough participation in the audit process by all groups of workers as well as local unions and NGOs and it includes the use of participatory interview techniques (such as focus groups or the use of drama or storytelling). Thus, a serious challenge for standard initiatives that seek to promote enabling rights is that they demand more resource-intensive and costly monitoring procedures. Furthermore, not all value chain actors are interested in promoting enabling rights, thus the scaling-up of initiatives focusing on enabling rights is a big challenge. See Table 1 for an illustration of the concepts linked in the discussion so far (although obviously some standards fit less well in these ideal typical categories).
| Type 1 standard schemes | Type 2 standard schemes |
|---|---|
| Schemes weighing size over principles | Schemes weighing principles over size |
| Focus mainly on risk management | Focus mainly on product differentiation |
| Focus on protective rights (outcome standards) | Focus on enabling rights (process rights) |
The critical literature reviewed here argues that consistent forms of improvements for intended beneficiaries is only achieved by marginal standards and point at a tendency for standard harmonisation and mainstreaming attempts to spur convergence around a lowest common denominator, although there are more optimistic views (e.g. Bernstein and Cashore 2007). Utting (2008) is cautiously optimistic about the use of multi-stakeholder initiatives and inclusive auditing processes as a means to continually ratchet up standards and procedures while challenging less stringent and less inclusive standards.
In the critical literature on sustainability standards there seems to be general consensus on the fact that multi-stakeholder standards using independent third-party auditing4 and participatory auditing methods (e.g. involving local actors and participatory interview techniques) are likely to lead to better social results compared to unilateral standards which employ first- or second-party auditing and a checklist approach to auditing (see for example Auret and Barrientos 2004, Utting 2005, O'Rourke 2006, Barrientos and Smith 2007, Blowfield and Dolan 2008, Tallontire et al. 2009). In this paper, I define stringent or ‘high bar’ standards as multi-stakeholder standards using third-party auditing and focusing on enabling rights. A focus on enabling rights here refers to standards which contain specific procedures aimed at enabling workers, such as demanding elected worker committees, using participatory interview techniques and inclusion of local NGOs and trade unions in implementing and monitoring. Discussions about the stringency of standards, however, should not divert attention from considerations about the limits of private standards as a means of improving working conditions in global production and particularly as a means of altering the power relations between labour and capital. Recently, the adverse effects of corporate buying strategies (particularly price cuts, short lead times and rapid turn-around) on labour conditions have been highlighted, and consequently the limits of standards which focus on production sites as sealed from the structural dynamics of the value chains they feed into (Barrientos et al. 2003, Utting 2005, Barrientos and Smith 2007, Riisgaard 2007, 2009a, Blowfield and Dolan 2008).5 This underlines the significance of unequal power relations in value chains such as the ones for cut flowers where buyers set the agenda and where producers are standard-takers.
Social and environmental standards in cut flowers
The landscape of flower standards
In Table 2, I I have provided the basic features of standards aiming to advance social or social and environmental conditions in the production of cut flowers. The table, which lists a total of 13 standards, only includes examples of standards that have some element of social issues (e.g. excluding organic standards). The first standards that emerged in the industry were mainly set by buyers or producer groups and tended to be weak on social issues and rely mainly on internal monitoring. During the 1990s, there was a general development towards convergence around International Labour Organisation (ILO) core labour rights, the use of third-party monitoring and the emergence of new multi-stakeholder initiatives. Standards furthermore have tended to broaden from social or environmental issues to include both, and from only covering cut flowers to also include pot plants and foliage. Although most standards are beginning to cover foliage and pot plants, most sustainability certification still takes place in cut flowers, wherefore in this paper I limit myself to looking at cut flowers. Furthermore, I focus on standards that are present in the EU market, since this is where we see the largest proliferation of standards.
| Standard | Origin and structure | Release | Characteristics and cooperation |
|---|---|---|---|
| Northern sectoral and corporate (examples) | |||
| GlobalGAP (flowers and ornamentals) | Coalition of European supermarkets | 2003 | Good agricultural practices with a small section on worker health, safety and welfare |
| Milieu Programma Sierteelt (MPS) | Dutch growers and auctions | 1995 | Environmental management (MPS A, B, C) with optional social qualification (MPS-SQ based on the ICC code) and MPS-GAP (benchmarked to GlobalGAP). MPS-A together with MPS-SQ has been benchmarked to FFP |
| Sainsbury's Code of Conduct for Socially Responsible Sourcing | UK supermarket chain Sainsburys | 2000 | Code on social responsibilities of suppliers based on the Ethical Trading Initiative's Base Code. |
| Morrison's Ethical Trading Code | UK supermarket chain Morrisons | ? | Code of practice based on ILO core labour rights |
| Southern sectoral (examples) | |||
| Kenya Flower Council (KFC) Code of Practice | (KFC) Association of Kenyan flower exporters | 1998 | Environmental and social certification system benchmarked to GlobalGAP and with a recognition agreement with Tesco's ‘Nature's choice’ from 2006. KFC is certified to ISO 65 guidelines |
| FlorVerde | Asocoflores (association of Colombian flower growers) | 1998 | Environmental and social certification system. Benchmarked to GlobalGAP |
| FlorEcuador | Expoflores (association of Ecuadorian flower growers) | 2005 | Environmental and social certification system |
| Multi-stakeholder or NGO consumer labels | |||
| The Fairtrade Labelling Organisation (FLO) (flowers and plants) | Fairtrade labelling organisations and Fairtrade producer networks | 2006 | Principles of Fairtrade, including a minimum price and the Fairtrade premium. Until 2006, Fairtrade flowers were certified by Max Havelaar cooperating with FLP and MPS |
| Flower Label Program (FLP) | German importers and wholesalers, NGOs and trade unions | 1996 | Environmental and social certification system benchmarked to the ICC base code |
| Fair Flowers Fair Plants (FFP) | Union Fleurs (the International Floricultural Trade Association), NGOs and unions | 2005 | Environmental and social labelling scheme based on both the ICC and MPS-A criteria |
| Rainforest Alliance (flowers and ferns) | Environmental NGOs | 2001 | Certification system for conservation of biodiversity and sustainable livelihoods |
| Multi-stakeholder base codes (no certification system) | |||
| International Code of Conduct for Cut Flowers (ICC) | Developed by a coalition of European NGOs and the International Union of Food and Agricultural Workers (IUF) | 1998 | Base code with criteria on human rights, labour conditions and basic environmental criteria |
| The Ethical Trading Initiative (ETI) | An alliance of UK companies, NGOs and trade unions | 1998 | Social base code (not restricted to cut flowers) |
| The Horticultural Ethical Business Initiative (HEBI) | Kenyan flower growers and labour NGOs | 2003 | Base Code on Social Accountability for The Flower Industry based on the ETI. It is unclear whether HEBI is still functioning (see Tallontire et al. 2009) |
Flower standards can be argued to diverge on five key dimensions (see Table 3). These dimensions include the characteristics listed in Table 1 as well as monitoring procedures, actors involved in standard setting and how the standard is communicated: 1) private business standards (elaborated by buyers or producer groups) versus collective private standards (elaborated by multiple stakeholders including NGOs and/or trade unions); 2) third-party monitoring versus first- or second-party monitoring; 3) standards focusing mainly on risk management (and size) versus standards focusing mainly on product differentiation (and high principles); 4) standards that are communicated to consumers (consumer labels) versus business-to-business standards; 5) standards focusing on protection rights versus standards focusing on enabling rights. Furthermore, some standards are national in scope (mainly those created by developing country producer associations) while other standards are international.6
| Standard | Number of certified growers1 | Business standard | Collective standard | Third-party auditing | First/second-party auditing | Mainly risk management | Mainly product differentiation | Protective rights/ Outcome standards | Enabling rights/ Process rights | Business to business | Consumer label |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Retailer codes2 | ? | x | x 3 | x | x | x | |||||
| GlobalGAP | ? | x | x | x | x | x | |||||
| MPS-(A,B,C,GAP) | 3800 | x | x | x | x | x | |||||
| MPS-SQ | 115 | x | x | x | x | x | |||||
| Kenya Flower Council (KFC) | 50 | x | x 4 | x | x | x | |||||
| FlorVerde | 108 | x | x | x | x | x | |||||
| Flor-Ecuador | 92 | x | x | x | x | x | |||||
| Fairtrade Labelling Organisation (FLO) | 42 | x | x | x | x | x | |||||
| Flower Label Program | 58 | x | x | x | x | x | |||||
| Fair Flowers Fair Plants (FFP) | 165 | x | x | x | x | x | |||||
| Rainforest Alliance | 35 | x | x | x | x | x | |||||
| International Code of Conduct for Cut Flowers (ICC) | NA | x | NA | NA | x | x | x | ||||
| Ethical Trading Initiative (ETI) | NA | x | NA | NA | x | x | x | ||||
| Horticulture Ethical Business Initiative (HEBI) | NA | x | NA | NA | x | x | x | ||||
| Note: 1 As appear on standard websites as of Janurary 2009. Unfortunately the number of certified growers does not say anything about the volume or value of certified flowers. In developing countries the largest farms are most often certified and often to multiple standards. 2 The analysis of retailer codes is based on the top five UK supermarkets: Sainsbury's, Waitrose, Morrison's, Tesco and Asda. 3 Several retailers do use third-party auditing of high risk suppliers but risk assessments are based on supplier self-assessments. 4 KFC also has a gold standard which requires third-party auditing. Nine farms are certified to the gold standard. |
In terms of market coverage, social and environmental standards have become mainstream for flowers imported into the EU. A rough estimate assesses that between 50% and 75% of imported flowers adhere to one or more of the standards mentioned in Table 3. However, the vast majority of these standards are business-to-business standards (governed by business and not communicated to the consumer). Of the standards aimed at the EU, Milieu Programma Sierteelt (MPS), GlobalGAP and retailer codes are by far the biggest in terms of number of producers certified. It is not known how many producers are certified under GlobalGAP or retailer codes, but any producer exporting to British retailers will have to prove compliance.7 MPS has almost 4000 certified growers – although most under their environmental or GAP schemes only. These standards are business standards (MPS is owned by the Dutch auctions and flower growers which account for half of total production value in the EU,8 while GlobalGAP was initiated by a group of large European retailers).
In terms of the earlier discussion on whether standards seek to raise the bar on social issues or merely hold the bar (i.e. Type 1 versus Type 2 standards as listed in Table 1), Table 3 suggests that large standards (in terms of number of producers certified) are not aimed at raising the bar. While there have been developments towards convergence around ILO core labour rights and increased use of third-party monitoring, these standards still focus mainly on risk management and protective rights, not enabling rights (cf. Tallontire et al. 2009). The standards that aim at raising the bar are mainly consumer labels operating in niche markets. Thus the current state of affairs seems to confirm the findings from other agricultural commodities set out above in the section on the function and focus of standards (Mutersbaugh 2005, Macdonald 2007, Raynolds et al. 2007).
The standards, that are communicated through a consumer label characterise a much smaller portion of the market (no exact figures exist, but an estimate puts their market share between 5% and 10%, depending on the country).9 These standards tend to be collective standards focusing mainly on product differentiation and enabling rights. However, the share of consumer-labelled flowers has been rising quite rapidly over the last years (sales of Fairtrade flowers, for example, increased by 66% from 2006 to 2007).10 Furthermore, the ambitious harmonisation initiative carried out by Fair Flowers Fair Plants (FFP) (as will be discussed below) has potential to significantly increase the market share of labelled flowers. Thus it can be argued that standards seeking to raise the bar on social issues are becoming more important.
The introduction of sustainability-labelled flowers at the Dutch auctions
Simply put, the flower value chain for import into the EU entails two distinctive strands (the direct strand and the auction strand). The Dutch flower auctions11 (owned by the Dutch flower growers) have historically been the most important channels through which flowers are distributed to European wholesalers and retailers. But lately the proportion of flowers imported into the EU that is directly sourced by large retailers is increasing (in 2006, the auctions had an estimated 40% market share of flowers, while large retailers had an estimated market share of 25–30%) (CBI 2007).12 For Tanzania and Kenya, a rough estimate places around two-thirds of exports still going to the Dutch auction halls while approximately one-third of total exports is supplied directly to individual wholesalers and/or retailers, particularly in Germany and the United Kingdom (Riisgaard 2007).
Social and environmental standards are one of the governance tools through which retailers seek to reduce risks, minimise costs and differentiate their products. Certification to social and environmental standards is most often a requirement for producers exporting through this channel. Social standards so far have not played a significant role for flowers sold through the auction. Until 2007, the only standard differentiated at the clock auctions (i.e. with descending price) were MPS-A, B and C (the environmental standards owned by Dutch growers, who also own the auctions).13
The FFP standard initiative, however, has managed to negotiate an agreement with the FloraHolland auctions (FloraHolland represents 98% of the turnover at the Dutch auctions). From January 2007, the auctions started indicating FFP certification at the clock front and in their supply systems so that auction buyers were for the first time able to differentiate products certified to a social standard (Interview 9, 2008). Given that approximately 40% of all flowers go through the auctions and the fact that FFP is backed by very influential industry actors (uniting producers and traders covering an estimated 80–90% of the EU flower market), this development could increase the overall market share of labelled flowers and at the same time flowers certified to a standard focusing on enabling rights (Interview 19, 2008).14 In the remainder of this paper, I take a closer look at developments within the ‘high bar’ category of flower standards and illustrate the importance of asking for whom is the standard seeking to raise the bar.
How to raise the bar
The International Code of Conduct for Cut Flowers (ICC) is a base code (it has no standard organisation behind it) that can be adopted by any standard scheme. The ICC aims to guarantee that flowers have been produced under socially and environmentally sustainable conditions, with implementation mechanisms designed to include the meaningful participation of workers, local organisations and unions (ICC 1998, Interview 12, 2008).
The ICC code has formed the basis for the social content of most of the ‘high bar’ standards (FLP, FFP, MPS-SQ [Socially Qualified], FLO and HEBI. However, within this category of standards there are different interpretations of how to raise the bar. There are also differences in what kind of bar should be raised and who should reap the benefits. As will be illustrated in the following two sections, this has quite different implications for different stakeholders.
Trade union empowerment and the International Code of Conduct for the Production of Cut Flowers (ICC)
The ICC was developed by a coalition of European NGOs and the International Union of Food and Agricultural Workers (IUF) in 1998 and represented the beginning of trade union involvement in private standard setting in flowers.15 Preceding its involvement with the ICC, the IUF had tried to run traditional organising campaigns in flower-producing countries. These were ultimately found to be based on insufficient union strength at national level. At the same time, a plethora of standards (mostly unilateral) were emerging that favoured employers rather than unions. In this context, the IUF decided to challenge unilateral business codes of conduct (weak in content, scope and monitoring), constructively engage in private standard setting and build its own multi-stakeholder model standard, the ICC (Interview 12, 2008 and Interview 13, 2006).
The ICC formed the centrepiece of a campaign to regulate work and employment in flower production. Initially, importers in Germany were targeted, with success, to accept the ICC. In 1999 the Flower Label Program16 standard was benchmarked against the ICC. Shortly after the German campaign, the Dutch MPS17 initiative was targeted, and this led after several years of discussion to the development of the MPS Socially Qualified standard (MPS-SQ) also based on the ICC (Interviews 1 and 12, 2008). The ICC has also indirectly formed the basis for the FLO18 Fairtrade standard for flowers19 and the social criteria for certification to the FFP label are those of the ICC.
Even though several standard schemes are based on the ICC code, the interpretation of how to actively involve workers, NGOs and trade unions differs significantly between standard schemes. A section on implementation in the ICC code, on the involvement of NGOs and unions states: ‘1. To overview the implementation of the Code of Conduct an independent body, accepted by all parties involved (for example trade unions, NGOs, employers), shall be formed. 2. This body will set the terms for an independent process of verification of compliance with the Code of Conduct’ (ICC 1998). Adjoining the ICC code are the ICC guidelines which interpret how the code should be read. The first version of the guidelines does not elaborate on the participation of NGOs and unions. In 2003, FIAN (FoodFirst Information and Action Network) Germany, Switzerland and Holland updated the guidelines of the ICC20 which now state that ‘during the audit, a trade union and/or NGO representatives are present as observers’ (Both ENDS 2005).
Thus, NGOs and trade unions where involved in setting the ICC standard, which stipulates that a body accepted by all parties shall overview standard implementation and that local trade unions and NGOs shall participate in all audits. In particular, the practice of securing local trade union and NGO participation in audits constitutes a unique strategy. In section two, I related findings which strongly suggest that social auditing – while having an effect on the more visible aspects, such as health and safety and working hours (protective rights) – is having little or no effect on more embedded issues such as discrimination and freedom of association (enabling rights). Seen in this light, audit shadowing by local NGOs and trade unions (although not without problem)21 might help to remedy this serious shortcoming of social auditing in several ways: First, it secures the inclusion of local knowledge about the sector and the particular flower farm which potentially greatly increases the possibility for uncovering more hidden problems; second, it gives local organisations access to farms and workers in the flower industry and potentially aid organising efforts and the sustainability of any improvements that might come about as a result of standard implementation; third, it gives local organisations the possibility of contesting audit results and voicing their recommendations.
But even though this base code has been adopted by several standard schemes, the way they adopted it differs. MPS-SQ and later FFP have institutionalised audit shadowing as recommended in the ICC guidelines version II. In the Flower Label Program, participation of unions and NGOs is guaranteed in the institutional structure of the organisation and is implemented in the auditing procedures as follows: ‘Workers, trade unions and NGOs have the right to join the inspections’ (FLP 2007). However, the Flower Label Program does not automatically inform unions and NGOs about the inspections (nor do they pay allowances for observers), but unions and NGOs are free to ask if they can join the inspections (Interviews 6 and 7, 2008).22
The FLO standard for Flowers and Plants for Hired Labour23 (although indirectly based on the ICC), does not mention the inclusion of local trade unions or NGOs (Interview 10, 2008). In FLO, the only issues explicitly concerning trade unions are the provisions in the FLO generic standard for hired labour relating to the right to organise and to collective bargaining. There is no institutionalised connection to trade unions or local NGOs – either in the standard organisation or in implementation. As commented by a FLO representative, ‘FLO standards don't really have anything to do with unions, the weight in FLO standards is put on other issues such as the trade contracts and the premium committee’ (Interview 10, 2008). Thus, while in the FLO standard emphasis is not placed on active inclusion of local NGOs and unions, it does actively include workers through the Fairtrade premium committees, thereby potentially empowering these actors.
The discussion of how different standard schemes (all seemingly based on the ICC code) interpret the requirement in the ICC code on worker, NGO and trade union involvement illustrates that interpretation is important and affects which stakeholders are empowered. FFP, MPS-SQ and the Flower Label Program institutionalised active inclusion of local NGOs and unions in standard setting and monitoring (although the Flower Label Program in a weaker form). For the FLO Fairtrade standard, emphasis is not placed on active inclusion of local NGOs and unions but on active inclusion of workers in implementing the standard (through the Fairtrade premium committees) and on addressing terms of trade through contractual requirements.
The only flower standard initiative to incorporate local civil society organisations in standard setting and implementation as well as demanding the use of participatory interview techniques is the Kenyan Horticultural Ethical Business Initiative (HEBI) established in 2003. The UK-based Ethical Trading Initiative (of which most UK retailers are members) was particularly active in the establishment of HEBI, and the HEBI code was soon endorsed by both the Fresh Produce Exporters' Association of Kenya (FPEAK) and KFC (the two Kenyan business associations with flower grower members). However, in practice HEBI's participatory social auditing methodology has been effectively sidelined by the local flower industry. Indeed, the auditing procedures of the FPEAK and KFC standards remain modelled on international technical audit procedures, which are not participatory in nature. This way FPEAK, KFC and ETI retail members have been able to be seen as acting to redress civil society calls for more participatory governance structures while in practice not changing anything substantial. Recently HEBI seems to have ceased to function altogether, and UK retailers are increasingly demanding that suppliers submit audit data to SEDEX, the Supplier Ethical Trade Data Exchange – a system for the exchange of audit data which dictates an audit format that is quantitative and far from participatory in nature (Tallontire et al. 2009, Riisgaard 2010).
The fate of HEBI reminds us that in practice the outcome for different local stakeholders varies not just according to standard content and monitoring methodology but also according to local interpretations of standard implementation and the relative strength and positioning of local stakeholders. Research conducted in Tanzania and Kenya investigated labour responses to (mainly) multi-stakeholder standards that attempt to place greater emphasis on enabling rights in the cut flower industry. In 2006, approximately half of around 150 flower farms in Kenya were certified to one or more sustainability standards. Approximately a quarter of certified farms were certified to a standard focusing on enabling rights. In Tanzania 3 out of the 10 existing farms were certified, all to standards focusing on enabling rights (as larger export farms tend to be certified this correlates to a larger percentage of total flower export in both Kenya and Tanzania) (Riisgaard 2007).
Several difficulties have been experienced in implementing these standards, particularly where they focus on enabling workers to claim their rights. Apart from exporters resisting participatory monitoring (as happened with HEBI), in some countries, national labour organisations are not interested in engaging with standards, as they consider them ‘Northern driven’. In Kenya, trade unions have perceived standards as a threat to their position. Thus, where adoption has occurred, there has been little labour union participation, no significant increase in unionisation, and actually worse relations between national unions and national NGOs promoting labour rights (Riisgaard 2009a). However, in other circumstances where national trade unions have engaged with standard initiatives, standard implementation has enhanced union organisation and facilitated the adoption of collective bargaining agreements. In Tanzania, this occurred through constructive interaction between the Tanzania Plantation and Agricultural Workers Union (TPAWU) and the FLP standard, whereby FLP requested union assurance that freedom of association and collective bargaining rights were in place before certifying farms. Ensuing activity resulted in collective bargaining agreements on the two largest farms. This facilitated access by TPAWU to the rest of the sector and now the ICC base code is routinely integrated in the text of farm-level collective bargaining agreements (Interview 21, 2009). Below, I discuss a particular attempt to scale up and harmonise flower standards around the ICC base code, to illustrate that it is important to ask for whom the standard is actually raising the bar.
Fair Flowers Fair Plants (FFP)
FFP was initiated in 2002 by Union Fleurs (the International Flower Trade Association uniting national producer, importer and traders’ organisations) but early in the process NGOs, unions and other standards were involved, particularly MPS and the Flower Label Program (Interview 5, 2008). FFP is a consumer label launched in 2005 and so far present in 10 European countries in supermarkets, florists and flower chains. FFP consists of a social and an environmental part and demands that growers are certified to a standard equivalent to the International Code of Conduct for Cut Flowers (ICC) and to the Dutch MPS-A standard. Thus FFP is a label based on existing standards benchmarked to the criteria of FFP. This way, the idea is to keep all existing standard schemes but to harmonise them through benchmarking to the FFP criteria and unite them under one consumer label (Interviews 1–5, 2008).24
This modular approach is interesting in relation to the distinctions between standard functions discussed in section two. FFP is combining a product differentiation standard (the ICC) with a risk management standard (MPS-A) and at the same time combining a standard that weights principle over size, focusing on enabling rights (the ICC), and a standard that weights size over principle (MPS-A). FFP is growing rapidly, and if growth continues, this modular strategy will possibly lead to existing risk management standards (like the producer association standards) being transformed into product differentiation standards. It also means that FFP might have found an answer to the dilemma of how to scale up ‘high bar’ standards which focus on enabling rights. However, although the number of FFP participants has increased rapidly (with 4593 members in January 2010),25 of the 159 members which are flower growers, 134 are situated in the Netherlands and only 27 in developing countries. Consequently, scaling up in this particular initiative has so far mainly reached northern (particularly Dutch) workers and workplaces.
Interestingly, FFP has also changed the competitive strategies available to the Dutch flower growers (who account for half of total production value in the EU) (CBI 2007). The Dutch flower growers have been threatened by increasing imports of flowers from developing countries. Moreover, most Dutch growers are co-owners of the auctions and are obliged to sell all their produce through the auctions.26 Before FFP, the only social consumer label available to developed country growers was the Flower Label Program aimed almost solely at the German florist market (the label only has one certified producer from a developed country, while the Fairtrade label is only available to developing country producers). Thus, before FFP came into existence there was a market (small but growing rapidly) for socially certified flowers, which the Dutch growers were not able to enter: a) because of restrictions inherent in existing standards and b) because they are obliged to sell through the auctions where previously product differentiation by a label was not possible. For the Dutch growers, FFP offers an opportunity to enter the market for flowers differentiated by social certification, which earlier was restricted to southern producers such as the ones from Kenya and Tanzania.
Conclusion
In this paper, I have explored whether parallelism has spurred a race to the bottom in flower standards. The answer is not an ‘either/or’. For standards aimed at the EU market, risk management standards (focusing on protective rights), such as retailer codes and GlobalGAP, still dominate in number, but ‘higher bar’ standards (focusing on product differentiation and enabling rights) are gaining importance. ‘Higher bar’ standards are converging around the ICC base code and adopting innovative auditing practices and are expanding and entering new value chain strands (the ‘traditional’ Dutch auction strand). The FFP harmonisation attempt, if it gains currency with consumers and other standards schemes (particularly with some of the not very stringent developing country producer association standards), has potential to scale up ‘higher bar’ standards by benchmarking risk management standards to the FFP criteria, thereby turning them into product differentiation standards and multiplying the practice of active inclusion of local NGOs and trade unions in monitoring standard compliance. On this basis, the analysis carried out in this paper suggests cautious optimism about developments in flower standards.
While it is indeed promising and inspiring that the innovative practice of audit-shadowing by local NGOs and trade unions is gaining ground among flower standards, there are still serious concerns to be addressed. For example, such practices are not yet adopted by mainstream standards, and none of the functioning flower standards have yet extended inclusive auditing to encompass participatory worker interview techniques. Growers from Kenya along with other southern-hemisphere growers tend to chose the less stringent national producer association standards, if any standard at all, and only adopt ‘higher bar’ standards when pressured to do so by their buyers. Furthermore, most flower standards still do not address the terms of trade (e.g. low prices, increasing quality demands, and retailer practices such as just-in-time ordering) which constrain suppliers' ability to comply with social standards.
The analysis carried out in this paper shows that even though ‘higher bar’ standards are converging around the ICC code, within this category of standards there are different interpretations of how to raise the bar. There are also differences in what kind of bar should be raised and who should reap the benefits. For production-end stakeholders, it matters whether a standard focuses on active inclusion of local NGOs and unions (empowering these stakeholders and potentially addressing more hidden problems like for example discrimination or exploitation of subcontracted, casual or migrant workers) or whether a standard focuses on addressing the terms of trade (aiming at empowering producers via contractual requirements) and worker inclusion (via involvement in spending a Fairtrade premium).
Furthermore, while the FFP harmonisation attempt has potential to scale up enabling standards, it also illustrates some of the difficulties in navigating between the objectives of ‘size’ and ‘principles’. ‘Fair labelled flowers’ in the minds of many consumers will spur images of ‘a fair deal’ for southern workers and southern producers. However, in practice only few southern workers and producers are yet involved in this particular initiative and thus the deal for southern producers is perhaps less clear since producers for example from Kenya or Tanzania now have to compete with developed country producers for whom certification is perhaps less of a challenge. The analysis thus draws attention to a relationship that deserves further research, namely that changes in the market for standards can shape competition in the market for goods. In this paper, this has been illustrated via the FFP harmonisation initiative which effectively has altered the terms of participation in the growing market segment for socially labelled flowers, where northern growers are now able to compete in a market which before was restricted to Southern producers.
Note on contributor
Lone Riisgaard holds a PhD in International Development Studies from Roskilde University, Denmark. She is currently a Project Researcher at the Danish Institute for International Studies (DIIS). She has researched and published on sustainability standards, labour rights issues and Global Value Chain analysis, with focus on the cut flower value chain originating in Kenya and Tanzania.