C OMPETITION AND C OOPERATION IN THE M ARKET OF V OLUNTARY S USTAINABILITY S TANDARDS

Voluntary sustainability standards (VSS) have proliferated over the last two decades. Currently there are at least more than 400 schemes operating internationally. The paper starts with a description of the emergence, proliferation and global diffusion of VSS. It subsequently focuses on the diversity between systems in the wider group of VSS. What emerges is a diverse and growing (both on the supply and demand side) market of VSS. This proliferation and diversification of VSS is creating problems such as a credibility gap, increased costs to gain certification by multiple VSS systems and a potential race to the bottom of systems. These problems point to the need for increased cooperation between VSS systems. Next, the paper introduces and discusses two mechanisms which can facilitate more cooperation, mutual recognition and meta-regulation. Mutual recognition is especially suited to address cost issues, while meta-regulation might contribute to leveling the playing field. The paper analyzes these mechanisms and argues that cooperation in the market of VSS is currently more developed with regard to addressing the credibility gap (meta-regulation) rather than with regard to addressing the cost issue (mutual recognition). cooperation to address these market problems: mutual recognition and meta-regulation. It was argued that mutual recognition is especially suited to address cost issues, while meta-regulation might contribute to leveling the playing field within the market of VSS. The paper analyzed these mechanisms and showed that cooperation in the market of VSS is currently more developed with regard to addressing the credibility gap (meta-regulation) rather than with regard to addressing the cost-issue (mutual recognition).

In the spring of 2013 the United Nations Forum on Sustainability Standards (UNFSS), a joint initiative by five UN agencies (FAO, UNIDO, ITC, UNEP and UNCTAD), was launched. The UNFSS is a platform created to generate knowledge and information on voluntary sustainability standards (VSS) with a particular focus on their potential contribution to development. The initiative is a recognition of the increasing importance of VSS in international trade. As the website of the UNFSS notes, VSS, although legally non-binding, can or in some areas already have become a de facto market entry hurdle (see also Wouters & Geraets 2012). Indeed, they potentially impact international trade in different ways. First, they can act as new non-tariff barriers diminishing export opportunities for those who face constraints in complying with stringent standards. Secondly, they can generate relocation effects. In several manufacturing industries, factories, or capital sustaining them, are highly mobile and are searching constantly for locations with the lowest input costs. As Levi et al. (2012, p. 22) note: "When challenged by workers forming unions or pressured by MNCs trying to induce compliance with private regulatory schemes, many factories will simply shut their doors without paying severance to workers and re-locate." Third, as Maertens and Swinnen (2012) argue, they might also create opportunities for trade, opening up high-standards markets for those who comply with the standards.
The rise of transnational VSS has also been well documented in the academic literature. The literature includes references to the rise of civil regulation (Vogel 2008;Levi-Faur, 2012) or to the 'certification revolution' (Conroy 2007). The present paper first of all maps this revolution. It shows that over the last decades the number of VSS proliferated resulting in more than 400 systems of which many are operating transnationally. The first part of the paper analyzes the emergence and proliferation of VSS and then assesses the transnational diffusion of VSS. It ends with a discussion of the main drivers of VVS development. The second part focuses on the diversity of systems. The term 'sustainability standards' captures a variety of initiatives. The main distinguishing features are introduced and discussed and an empirical analysis of the standards' institutional diversity is presented. What emerges out of the first two parts is a market of many different VSS. 2 This proliferation and diversification of VSS is creating problems such as a credibility gap, increased costs to gain certification by multiple VSS systems and a potential race to the bottom of VSS systems. These problems point to the need for increased cooperation between VSS systems. The third part of the paper discusses these problems and looks at initiatives which aim to foster cooperation in the market of sustainability standards. In this context the paper focuses on two mechanisms specifically suited to address these problems, namely mutual recognition and meta-regulation. Mutual recognition is especially suited to address cost issues, while meta-regulation might contribute to leveling the playingfield and preventing a race to the bottom. The paper analyzes these mechanisms and argues that cooperation in the market of VSS is currently more developed with regard to addressing the credibility gap (meta-regulation) than with regard to addressing the cost issue (mutual recognition). Throughout the paper we will provide some examples and provide an empirical analysis on the basis of the Ecolabel Index Database. 3

The Emergence and Proliferation of VSS
In 1924 Rudolf Steiner gave eight lectures at Koberwitz (Silesia) on 'Spiritual foundations for the renewal of Agriculture'. He outlined the basic principles of biodynamic farming. An experimental circle of anthroposophical farmers implemented Steiner's principles in their daily farming practices (Paull, 2011;2011a). In 1927, a co-operative was formed to market biodynamic produce. One year later, in 1928, the first standards for quality control were formulated and the first sustainability label, the Demeter symbol, was introduced. When a farm conformed to the standards, the products could use the Demeter label. Three years later almost 1000 biodynamic firms were certified in Germany. Voluntary sustainability certification was born. Seventy-five years later, in 1999, a new Demeter logo was developed and Demeter was operating world-wide certifying more than 3500 products including bread, wines, clothes and toilet articles. It took many years before new labels emerged. In 1954 Spiel Gut was established to promote sustainable toys. In the 1970s some additional biological agricultural labels emerged, such as Bioland in 1971, California Certified Organic Farmers (1973 and the Soil Association organic standard (1973). In 1978, the first government-led voluntary sustainability standard emerged in Germany, Blue Angel. In the early 1980s one can observe the emergence of three to four new initiatives on an annual basis (see Graph 1). Source: Ecolabel Index Database (N = 380) From 1989 onwards VSS were established more frequently. From the mid-1990s onwards one can observe a strong increase in the emergence of the number of VSS (Graph 1) and the survival/persistence of VSS (Graph 2). Indeed, as Graph 2 shows, one cannot only observe an annual increase in the emergence of VSS in the last two decades but also that many VSS survive, persist and are still active in 2011.

Graph 2: Number of Cumulative VSS: 1940-2011
Source: Ecolabel Index Database (N = 380) Moreover, VSS are becoming active in many economic sectors and several are active in more than one economic sector, which explains the fact that Graph 3 has 1218 observations. Most VSS are active in the food sector (145), followed by textiles (103), building products (101) and forest and paper products (80). In other words, for most products several VSS are available from which to choose if one intends to apply for a certificate.

Graph 3: VSS Distribution Across Economic Sectors
Source: Ecolabel Index Database (N=1218) The various initiatives differ from one another (see infra). Most initiatives, however, are characterized by the fact that an organization defines social and ecological standards and that there is a procedure to check that products or production processes conform to these standards (i.e. conformity assessment). When products or production processes comply with the defined standards, a certificate is awarded which may or may not be used for external communication (label).
This collection of voluntary standards comprises many different initiatives. A few examples illustrate the diversity. Some initiatives are government-driven, such as the OECD Guidelines for Multinational Enterprises, which were most recently revised in 2011. These are recommendations, accepted by governments, made to multinational companies regarding socially responsible business. The Guidelines define voluntary principles and standards for responsible behavior of companies, relating to matters such as the environment, fighting corruption, labor relations and competition. Other initiatives are driven by industry (associations) or NGOs, such as Responsible Care in the chemical sector, the charter of employment rights of the clothing giant GAP, and the Clean Clothes Campaign, which strives to achieve better terms of employment in textile production plants in developing countries. Other initiatives result from a collaboration between different types of actors such as the International Labor Organization (ILO)'s Tripartite Declaration of Principles Concerning Multinational Enterprises and Social Policy. Some initiatives which have generated significant academic interest are non-state multistakeholder sustainability certification initiatives. Some authors consider them 'one of the most innovative and startling institutional designs of the past 50 years' (Cashore et al. 2004: 4). Among the most prominent and representative examples of these private regulatory initiatives are the Fairtrade Labelling Organization (FLO), the Forest Stewardship Council (FSC), the Fair Labour Association (FLA), Social Accountability International (SAI) and the Marine Stewardship Council (MSC). The FLO, established in 1997, was founded to enable producers and workers in developing countries to evolve from a position of vulnerability to a position of economic security and self-sufficiency. The core of the system focuses on the concept of a fair price. The FSC, set up in 1993, is an international, multi-stakeholder, consensus-based sustainable forestry initiative. It guarantees that a wood or paper product has been made using material from a sustainably managed forest. SAI is a non-governmental, international, multi-stakeholder and non-profit organization whose mission is to promote the rights of workers worldwide and to improve working conditions by applying socially responsible standards. The SAI standards are based on internationally recognized guidelines, including various ILO conventions, the UN Convention on the Rights of the Child and the Universal Declaration of Human Rights. The FLO grew out of the Apparel Industry Partnership (AIP) initiative of the Clinton administration to protect workers worldwide and provide firms and consumers with the information they need to make informed purchasing decisions. The partnership was composed of apparel and footwear firms, human rights groups, labor and religious organizations, and consumer advocates. The FLA now represents a multi-stakeholder coalition of business enterprises, colleges and universities, and NGOs (human rights, labor, religious and consumer groups). Its mission is to combine the efforts of these stakeholders to promote adherence to international labor standards and improve labor rights worldwide. A final example is the MSC, which grew out of a partnership between WWF and Unilever and which aims to sustainable manage oceans and fisheries.

The Transnational Diffusion of VSS
Besides their exponential growth, VVS are also strongly diffused transnationally since they mainly aim to govern supply chains which operate across borders. Table 1 assesses the global adoption of four leading VSS systems. These are FSC (column 1); PEFC (Program for the Endorsement of Forest Certification, column 2); SAI (column 3) and GLOBALG.A.P. (column 4). To be sure, this selection is not representative but presents the use throughout the world of four of the most established and academically analyzed schemes. These schemes operate on a global scale (see also Abbott & Snidal, 2009) and make information on their adoption publicly available. In order to identify trends Table 1 reports the results of two measurement waves with a gap of three years. The first mapping was conducted in October 2010 and a second mapping in October 2013. The presence of a VVS in a country is indicated by a color and the letter P. We have four columns for the four VSS schemes. The cells in blue are based on the mapping of 2010 and indicate that the VSS in question is active in that particular country; the cells in yellow are based on the mapping of 2013. The cell in red indicates that the VSS was present in that country in 2010 but no longer in 2013. Table 1 reveals several interesting facts. First, VSS keep on expanding across the globe, with the only exception of Cuba, which lost its GLOBALG.A.P. certified entities. Second, the truly global diffusion of VSS schemes is limited. In a truly global scenario each of the systems would operate in each country and hence all of the 904 cells would contain an indication of the presence (P) of the VSS schemes in the country. Table 1 shows that VSS schemes, in 2013, are present in only 352 out of 904 possible cases (38,9%). It should be noted that this constitutes a strong increase in comparison to 2010 when 254 out of possible 904 cases (28%) had a VSS present. Thirdly, the table shows that there are several countries in which no VSS scheme is active, while in other countries all four are active. A closer look reveals that VSS schemes are mainly adopted in developed countries. The 28 EU Member States 4 account for 103 of the 352 (29%) cases in which the selected VSS are active. This is a small decrease compared to 2010, when they took up 33%. In many least developed countries no VSS systems are active. In addition, in countries in which VSS can be observed there is a significant variation in the number of entities certified. For SAI this varies from one facility being certified in several countries such as Tanzania, Nepal, Kenya or Jordan, to a few hundreds in Italy. Similar figures hold for GLOBALG.A.P., which certifies only one or a few farms in Zimbabwe and Zambia, and a few thousand farms in Germany and Greece. In general, the increase of countries in which VSS are active is mostly explained by the fact that the four VSS are getting a hold on the ground in more countries, albeit often with only one or two facilities certified. But once active within a country, the adoption of VSS by other facilities can go quickly. 5 Indeed, in some cases the number of certified entities increased enormously over the three year period. Moreover, overall figures of adoption of VSS on a global scale indicate a strong demand for VSS and a strong increase in VSS adoption. For example, in 2012 GLOBALG.A.P. had 123,115 facilities certified worldwide (Annual Report, 2012), compared to 112,576 in 2011(Annual Report, 2011. The FSC has two types of certificates, those who certify forests (FM) and those who certify all operators in the supply chain (Chain-of-Custody, CoC). The number of forests (FM) certified grew from 1000 in January 2011 to 1225 in September 2013. The number of CoC-certified facilities grew from 18000 in July 2010 to 26886 in September 2013, an increase of almost 50% (FSC Facts and Figures, 2013). This rapid growth is driven by a strong demand for products which are certified according to sustainability standards. This demand is especially driven by large retailers. Marks & Spencer, for example, committed 'to become the world's most sustainable major retailer by 2015' (Marks & Spencer 2010: 3). This strategy, by 2020, will include building in environmental and social principles in all 2.7 billion individual products. In most cases this will be done by using recognized external environmental or social certificates, i.e. using a VSS system (Marks & Spencer 2010: 14). Overall, the figures show a wide diffusion of VSS and a significant demand for VSS. Guam Peru P P P P Argentina P P P P Guatemala P P P Philippines P P P P Armenia P Guinea P Pitcairn Aruba Guinea-Bissau Poland P P P P Australia P P P Guyana P Portugal P P P P Austria P P P P Haiti Puerto Rico P P Azerbaijan Honduras P P P Qatar P Bahamas Hungary P P P Réunion Bahrain P P P Iceland Romania P P P P Bangladesh P P P India P P P P Russian Federation P P P Barbados Indonesia P P P P Rwanda Belarus P P P Iran P P Saint Helena Belgium P P P P Iraq Saint Kitts and Nevis Belize P Ireland P P P Saint Lucia Benin Isle of Man S. Pierre Miquelon Bermuda Israel P P P P S. Vinc Grenadines P Bhutan Italy P P P P Samoa Bolivia P P P Jamaica P San Marino Bosnia&Herzegovina P P P Japan P P P P Sao Tome Principe Botswana Jordan P P Saudi Arabia P P P P Brazil P P P P Kazakstan Senegal P British Virgin Islands P Kenya P P P Serbia P P P P Brunei Darussalam P Kiribati Seychelles P Bulgaria P P P P Korea, DPR Sierra Leone Burkina Faso P Korea, R P P P P Singapore P P P Burundi Kuwait P Slovakia P P P Cambodia P Kyrgystan Slovenia P P P P Cameroon P P Laos P P Solomon Islands P Canada P P P P Latvia P P P P Somalia Cape Verde Lebanon P P P South Africa P P P P Cayman Islands P Lesotho Spain P P P P Central African Rep Liberia P Sri Lanka P P P P Chad Libyan Arab Sudan Channel Islands Liechtenstein Suriname P P Chile P P P Lithuania P P P P Swaziland P P China P P P P Luxembourg P P P Sweden P P P P Colombia P P P P Macedonia P Switzerland P P P P  This adoption overview does not provide information on the nature of organisations (small versus large) that are participating in these schemes. From previous research (Marx & Cuypers, 2010; we know that, for example in the case of FSC, mainly large-scale producers were participating in FSC 6 . As a result, recent focus has turned to small-scale forest owners and wood producers to join FSC. These efforts to engage small-scale forest owners are mainly driven by FSC itself or technical cooperation units in the context of development cooperation. FSC is providing for group chain-of-custody certification which enables small-scale producers to join forces and financial and technical assistance to small forest owners for forest management certification. Large donors, like GiZ, are providing support to engage small-scale forest owners in VVS dynamics. These efforts might result in significant increased adoption by small-scale producers. Similar trends and developments might occur in other VSS. It should be noted that this scaling-up should not only focus on providing assistance to small-scale owners but on removing perverse incentives which generate market distortions in some markets (ie subsidies which favor large-scale industrialized production).

The Drivers of VSS Emergence
The rise and proliferation of VSS initiatives is driven by several interrelated factors. Some of these standards were developed to address certain policy gaps and failures, such as a lack of social and environmental protection or food safety (Hachez & Wouters, 2011;Levi et al., 2012) or to protect key business interests such as reputation or liability. In the social and environmental field NGO's boycotts and reputation protection provided the main incentives to establish VVS (Gereffi et al. 2001;Bartley, 2007;Marx, 2008). Brand protection is a key issue for many leading companies. The proliferation, diversification and internationalization of NGOs establishing themselves as genuine counter-powers through the use of media strategies, have forced companies to take civil society concerns into account (Bartley, 2003;O'Rourke, 2003). Gereffi et al. (2001) note, for example, that protests and direct actions against brand name retailers are only 15 years old, but are regarded as extremely powerful tools to force retailers to take environmental, social and safety issues into consideration (see also Bartley, 2003). Joining a voluntary sustainability initiative can be a strategy to protect a firm's reputation, especially when independent external parties certify products and production processes and hence provide assurance on the efforts made by firms to manage for example social issues (Marx, 2008).
In the food sector, a sector with many VSS, the globalization of food markets and the increasing need to market fully safe food has also contributed to the proliferation of VSS (Wouters, Marx and Hachez, 2012). It has become increasingly difficult for single governments to keep track of the range of products present on their domestic markets, and to keep up with the assessment of all the risks associated therewith. In the recent past, public food safety regulations have not been able to prevent some major food crises (see Ansell and Vogel, 2006). Consequently, states have adopted new legislation providing for a system of liability for food chain actors in case of food safety problems. An early example of such regulatory design is the due diligence requirement contained in the UK Food Safety Act (UKFSA) of 1990. The Act puts the liability mainly with retailers and provides that food retailers are not liable for non-compliance with food safety laws if they can demonstrate that they have taken all precautions in this regard. The possibility of being held liable for food safety issues prompted a response from the food industry, resulting in the development of many voluntary standards which acted as proof of due diligence (Henson, 2008;Food and Agriculture Organization, 2007). Mirroring the UKFSA, EU Food Law Regulation (EC) No 178/2002 of 2002 also provides for a quite stringent responsibility threshold for commercial actors involved with food products (Wouters, 2012). Given the important role and position of retailers in current day supply-chains (Hamilton, Petrovic and Senauer, 2012), this liability, which rests on them, results in pushing through 'voluntary' standards down the supply chain to all actors involved in the production process. Major retailer groups have particular influence as they are the 'gatekeepers' of the food markets and concentrate significant bargaining power. The World Bank refers in this context to the 'supermarket revolution' (World Bank, 2007) and Hamilton et al. (2012) to the 'Market Makers'.
From a more theoretical perspective, three major drivers can be identified which focus on preempting government regulation, addressing information asymmetries and direct targeting of firms by NGOs. Bartley (2011, pp. 444-446) has elaborated two possible explanations for the demand of VSS systems and identified two main, partially complementary, theoretical approaches. A first theoretical approach, mainly rooted in political institutionalism, approaches the emergence of VSS from the perspective of social actors (NGOs) and sees VSS as a NGOdriven institutional capacity-building project in order to achieve their aims. Rather than being confrontational towards firms or trying to influence firm behavior via lobbying governments to implement regulatory standards, NGOs are increasingly using a co-operative strategy towards firms, of which VSS are the most prominent example. As Bartley (2011, p. 445) notes, 'private efforts have also been perceived by many NGOs as a way to bypass political roadblocks'. A second theoretical approach, rooted in institutional economics, starts from the perspective of firms and highlights the importance of information asymmetries which arise via NGO actions such as 'naming and shaming' and consumer demand for ethical products. In this perspective, firms have an incentive to solve information problems by creating or joining VSS initiatives (Bartley, 2011). In addition, firms may have incentives to engage in standard-setting and adoption because it increases consumer confidence in product safety and quality, and hence increases consumer demand; or because VSS can be used as a strategic tool to differentiate products, thereby creating market segmentation and softening competition (Swinnen and Vandemoortele, 2011). A third theoretical perspective comes from economics which, following the work of Nobel laureate George Stigler, consider VSS as mechanisms which pre-empt government regulation (Swinnen and Vandemoortele, 2011;McCluskey, 2007). Private firms may choose a level of standards that minimize their own costs before governments or international organizations set standards. McCluskey and Winfree (2009) argue, in addition, that an important advantage of voluntary over public mandatory standards is that the former are more flexible in response to changes in consumer tastes and preferences, and to changes in technology. The discussion of these drivers for the emergence and development of VSS shows that several actors have strong incentives to develop and work with VSS, which structurally embeds them in a bigger constellation of transnational regulatory governance (Abbott and Snidal, 2009).

The Institutional Diversity of VSS
Notwithstanding the similarities in the process of granting sustainability certificates there is significant variation in how this process is organized and which actors and stakeholders are involved. Several authors have analysed this variation (Abbott and Snidal, 2009;Marx, 2013). For the purpose of this paper we focus on four dimensions of variation: who sets standards and how, who assesses the standards ex ante and how is ex post verification facilitated.
The first dimension focuses on who sets the standards and how they are set. Here attention goes to the process of rule development and standard setting, including the actors involved and decision-making processes. The International Social and Environmental Accreditation and Labelling (ISEAL) Alliance Code of Good Practice for Setting Social and Environmental Standards (2010) identifies at least five components which are important in the setting of standards: the identification of stakeholders via stakeholder mapping; the development of a strategy to proactively approach and involve the identified stakeholders; the bringing together of several major stakeholders on a more or less equal representative basis in a process of decision-making; the opening up of the decision-making process to all interested parties not initially identified in the first round of the stakeholder mapping; and the deployment of consensus-based decisionmaking in order to ensure that all interests are included. The Ecolabel Index database provides data on how many eco-labels have an open and consensus-based standard-setting and decisionmaking procedure. In this context, open and consensus-based decision-making is defined and operationalized following the American National Standards Institute's (2012) due process requirements for American National Standards. The American National Standards Institute code (2012) includes specific requirements for due process, meaning that any person (organization, company, government agency, individual, etc.) with an interest has a right to participate in the process. These requirements stipulate inter alia that the process should be open, not be dominated by any single interest category, balance several interests and operate on the basis of a consensus vote.
Concerning ex ante verification, several authors have argued that a standard-setting process can be open and transparent, but that the procedures to assess conformity with the rules can differ significantly, affecting the legitimacy of the systems (Gereffi et al, 2001;Blair et al., 2008). Different verification mechanisms exist. In this context, a first classification was proposed by Gereffi et al. (2001, p. 58 ff; see also Marx, 2011;Van Waarden, 2011). It focussed on the number of independent actors involved in the certification process. First-party conformity assessment is characterised by the fact that a firm enforces compliance with standards defined by the respective firm. In second-party conformity assessment one party develops the standards and a second party demands the standards. Conformity assessment with standards is conducted by either one of the two parties. In the case of third-party conformity assessment, the standards are developed by an independent organization which also accredits organizations to carry out conformity assessments. Accreditation guarantees that the organization that carries out the conformity assessment is competent to do so and operates independently. In this case, three independent parties are involved: an organization applying for a standard, a body setting the standards and an accredited party verifying conformity with the standards.
Concerning ex post verification two components have been highlighted as being relevant, namely the provision of information to the public on the implementation of standards and the accessibility of systems to raise complaints or disputes. The most extensive treatment of transparency and accountability in this context can be found in Auld and Gulbrandsen (2010). They distinguish between two forms of transparency, namely procedural transparency and outcome transparency. Procedural transparency refers to openness of the decision-making process. This form of transparency is closely linked to the standard-setting process outlined above. Outcome transparency refers to openness about the outcome of the standard-setting process and is important in the context of accountability mechanisms where stakeholders use disclosed information to hold actors accountable to their commitments. In organizational terms this is operationalized in two ways. First, VSS need to be transparent by providing accurate and objective information. Information disclosure procedures can inform different stakeholders on the implementation of standards. Publicly available information should include specific information about certification procedures, auditing reports, reports on violations and reports on corrective action plans. This allows stakeholders to assess whether the reported information mirrors real conditions. Second, a well-developed complaint procedure should be put in place to challenge decisions (Ascoly and Zeldenrust 2003).
The Ecolabel database allows us to explore empirically the degree to which these dimensions are present in VSS. Each dimension is either present (P) or absent (A) in a VSS. For a population of 426 VSS Figure 1 provides the results. Figure 1 brings the four dimensions together in a property space of 16 cells. Since a dimension is either present (1) or absent (0), four dimensions lead to 16 possible combinations (2 to the power 4) of dimensions. Figure 1 maps the 426 VSS over these 16 cells and shows that many cells contain existing VSS indicating a significant variation in their institutional design.  (155) have open and consensus-based standard-setting procedures and 3 rd party conformity assessment, but lack ex post verification tools, as is indicated by the red cell. Finally, only 12 VSS (yellow cell) have all four components and hence a very elaborated institutional design for developing standards, assessing conformity and providing mechanisms for ex post verification. In sum, 1/3 rd of the systems have no enforcement mechanisms and only 3% have a fully elaborated standard-setting and enforcement mechanism in place. In-between, a significant number of systems have an open and consensus-based standard-setting procedure and well-developed 3 rd party conformity assessment procedures. This diversity generates a clear credibility gap and several references can be found that VSS are pure 'green-washing'. One commentator even refers to the 'wild west of labelling' (Entine, 2013).

Cooperation in the Market of VSS
What emerges is a diverse and growing (both on the supply and demand side) market of VSS. This proliferation of initiatives leads to several (unintended) consequences. First, users (both consumers and businesses) of VSS are getting increasingly confused about the number of VSS and for what they stand. A survey with opinion makers in the field of sustainability standards concluded by the ISEAL Alliance in 2011 showed that 'the large number of standards has a series of implications: Overlaps between standards systems, confusion and difficulties in differentiating between them and a lack of consumer awareness and demand for specific certification schemes." (ISEAL-100 report, 2011). Similar results were recently published in a report by Seifert and Comas (2012) who surveyed several producers demanding VSS. This results in a credibility and legitimacy gap on the side of the VSS systems. The well-developed systems increasingly need to distinguish themselves from the non-credible systems.
Second, as indicated, several retailers are demanding compliance with VSS as a prerequisite to buy certain products. However, retailers and other intermediaries differ on the preferences they hold with regard to VSS. Some, for example, will ask wood products based on the FSC label while others might prefer the PEFC label. Since they do not mutually recognize each other's systems (infra), producers or forest owners often need to apply for the two labels, thereby increasing the costs of getting certified. Many systems also do not allow for multiple labeling with regard to labels which in essence certify the same standards. As a result, sometimes different packaging needs to be developed, which adds to the increased costs.
Finally, competition between systems can lead to a race to the bottom of standards and enforcement mechanisms in order to increase market share. Little information is available on how competitive dynamics influence the stringency of systems, but the rise of many VSS in certain sectors indicates that some of these VSS aim to capture quick gains. Van Waarden and van Dalen (2013) show that in the sector of Halal certification several VSS were recently introduced which were not very stringent and basically granted certificates with hardly any conformity assessment.
These concerns raise the issue of how (more) cooperation in the world of VSS can be achieved. Cooperation in this context refers to cooperative efforts to harmonize practices and reduce confusion and information gaps. Two possible cooperation mechanisms are discussed below, namely equivalence or mutual recognition, and meta-regulation. Equivalence and mutual recognition mechanisms are well-suited to address the issue of increased costs following multiple certification requirements. Meta-regulation is best suited to address the credibility/legitimacy gap and prevent a race to the bottom of VSS systems. Given the competitive and dynamic nature of the VSS market it is hypothesized that equivalence or mutual recognition holds currently little potential to act as a cooperative mechanism to reduce costs in the market while meta-regulation might generate more interest to address the credibility gap. This lack of mutual recognition will continue to generate possible negative consequences on the level of producers (ie no cost savings or administrative simplification).

Equivalence or Mutual Recognition
One way to cooperate within the market of VSS is via a system of equivalence recognition whereby one voluntary sustainability standard-setting organization formally recognizes the criteria and requirements of another. If such an agreement exists, entities that have been awarded the label in one program may therefore be able to register in another without undertaking certification and verification procedures again. This might significantly reduce costs for those applying for certificates. This system of recognition is directional and not necessarily symmetrical. One standard might recognize another without the other one recognizing the former. In order words, a VSS system can recognize another system or be recognized by another system. When there is symmetry or reciprocity one can refer to mutual recognition. For the field of eco-labels we can analyze the degree and nature of equivalence or mutual recognition since data are available.
In a highly coordinated and cooperative field many VSS would consider other VSS as equivalent.
The field of eco-labels, though, is highly fragmented and isolated. Figure 2 maps which VSS recognize other VSS as equivalent. The direction of the arrow indicates which VSS is considering another VSS as equivalent. The figure shows that the network, on the basis of the data available in the Eco-label Index database, is largely unconnected and that it has a very low density, a measure of the degree to which members of a network are connected. The circle shows the more than 300 VSS which are not recognized by another VSS and which neither recognize other VSS. The overall density, which is the proportion of equivalence recognitions on the potential total of recognitions, is 0.005, which is extremely low. This means that, even though there are a large number of VSS, very few of them utilitize the system of mutual recognition. Though 426 VSS initiatives are included in the dataset, only 90 instances are registered where one system recognizes the system employed by another VSS. Moreover, only 64 of the 426 VSS are recognized by others; this indicates that certain VSS are recognized by multiple VSS. Furthermore, only very few VSS apply mutual recognition. Mutual recognition occurs only 7 times among the total of 90 unique relationships (arrow on both sides of the line).

Figure 3: Network Graph of the VSS with equivalence recognition
When we zoom in on the ones which are connected (see Figure 3), we can identify some small clusters of VSS which play a role in equivalence recognition. The color and magnitude of the dots (a VSS-system) indicates to how many other VSS systems they are connected. The bigger the dot the more VSS systems recognize that particular VSS or are recognized by that VSS. Figure 3 presents the VSS initiatives that recognize or are recognized by other VSS initiatives. An arrowed line connects two VSS initiatives in the diagram. The system at the start of the line is the labeling system which "recognizes" and the VSS at the end of the line -indicated by an arrow pointing to its name -is the initiative "recognized".
The figure shows that five VSS play a central role in equivalence recognition of which Environmental Choice New Zealand is the most central. It is recognized by several other VSS and recognizes several other VSS. Environmental Choice New Zealand is the official environmental label of New Zealand owned and endorsed by the government of New Zealand. Other central VVS are PEFC, Greener Product Certification Seal, Energy Star USA and Bio Hotels. Energy Star USA is a government-backed program which is the national symbol for energy efficiency. Bio Hotels is an organic certification system and identifies hotels that only use organic certified products that meet the standard of some other VSS such as Bioland or similar VSS. In case of PECF equivalence recognition mostly refers to other forest certification systems, which previously operated independently but now operate under the global umbrella of PEFC and which are in the process of merging into PEFC. In total eleven VSS are recognized by two other VSS, and 48 VSS are recognized by one other VSS. The limited equivalence recognition which takes place in the market of VSS shows that this is mainly driven by government-led initiatives or incentives to merge and does not apply to the many private VVS systems operating. The latter remain highly competitive and operate in an un-coordinated fashion. It should be noted that one can observe some recent progress on a regional level with for example the emergence of the East African Organic Products Standards.
This highly fragmented picture corresponds to results from case studies which show limited cooperation and coordination between several very similar VSS. For example, in 2005 an initiative was launched in relation to labor rights and workers' rights with a view to organizing cooperation between the Clean Clothes Campaign, the Ethical Trading Initiative, the Fair Labor Association, the Fair Wear Foundation, Social Accountability International and the Worker Rights Consortium in a pilot project in Turkey. The 'Jo-I n' pilot project (Joint Initiative) aimed to promote closer cooperation between the various initiatives so as to increase the impact of the systems. This pilot project did not lead to closer cooperation, and the various systems continue to operate separately (Fransen, 2011;Marx, 2011). The obstacles for closer cooperation related to several issues including the stringency of standards, monitoring procedures and sanctioning. As Fransen (2011) notes, when analyzing private labor governance systems, no real steps are taken to address competition between systems and have convergence on standards and enforcement. In his comparative study Fransen (2011;see also 2012) shows that the prominent systems for private labor governance differ significantly in how they organize auditing in terms of stringency, who does the audits, etc. Similar efforts for cooperation, and their subsequent failure, have been reported in other sectors such as the cut flowers sector (Riisgaard 2009a(Riisgaard , 2009b, which is highly competitive. Even in areas of VSS which have been long established and have significant convergence in standards it took a very long time to establish a full equivalence agreement. A case in point is the organic agriculture and organic certification. David Winickoff and Kendra Klein (2012) compare transatlantic organic food labeling and make the case that organic cultures converged transatlantically to recognize similar process-based and precaution-based standards. Despite "relative convergence", relatively small but intransigent peculiarities remained for a long time, which resulted in some important differences between the US and EU standards with three in particular creating obstacles to equivalence: the approach to the farm unit, the lists of prohibited and allowed inputs, and criteria for livestock operations. It was only in February 2012 that the EU and US concluded a full formal equivalence agreement. This slow development of full mutual recognition resulted in a complexity of international accreditation and certification schemes.
As a result, cooperative measures which aim to reduce costs via mutual recognition are very little developed and there are only few signs of significant incentives to do so in the near future, especially with regard to competing private systems (see also Bowen, 2013 who develops this argument further.

Meta-Regulation of VSS
A second mechanism to foster cooperation in the VSS market which recently is receiving increasing attention is meta-regulation (Verbruggen and Havinga, 2013) or the development of meta-standards (Loconto and Fouilleux, 2013). There is no consensus on what meta-regulation is (Coglianese and Mendelson, 2010), but as Verbruggen and Havinga (2013) point out, the concept involves the use of control and steering mechanisms between different regulatory actors. Several authors have argued that due to the proliferation and diversification of systems there is an increasing need for the independent certification of the 'certifiers' by a multilateral organization or by a private organization. The idea is that meta-regulators determine the rules of the game and provide for minimum requirements to which VSS should adhere. This form of meta-regulation, in absence of government-led action, can potentially partially address the legitimacy gap and the potential problem of a race to the bottom (Bernstein, 2011).
A number of authors argue in favour of multilateral organizations taking on a pro-active role in this field (Sabel et al. 2000;van Waarden 2009). Abbott and Snidal (2009), in this context, refer to the orchestrating role international organizations can play in transnational regulatory governance. To date, however, no initiatives have been launched at the multilateral level with a view to achieving this objective. The UNFSS is the first multilateral initiative addressing sustainability standards. The UNFSS could potentially play this role, but it is currently not in its work programme for the near future. The UNFSS mainly addresses the potential trade or development obstacles VSS may create, with a particular emphasis on their impact on smallscale producers and less developed countries. It will mainly focus on analyzing trade effects.
Within the realm of private VSS some meta-regulatory systems are emerging. Within the food sector meta-regulatory initiatives have been taken such as the Global Food Safety Initiative (GFSI). The GFSI aims to harmonize food safety rules between different initiatives in order to reduce costs for suppliers. GFSI is a standard for other VVS in the food sector and is mainly focusing on the harmonization of the different retail specific standards (Verbruggen and Havinga, 2013). One of the most developed meta-regulatory systems in the realm of VSS is the International Social and Environmental Accreditation and Labelling (ISEAL) Alliance, which is discussed in the next section (Bernstein, 2011;Loconto and Fouilleux, 2013).

Meta-Regulation: the Case of the ISEAL Alliance
The ISEAL Alliance was established in 2000 by eight existing VSS including FLO, FSC and SAI. It is an umbrella members' organization that codifies best practices in the design and implementation of certification initiatives. ISEAL has drawn up a Code of Good Practice for Setting Social and Environmental Standards, i.e. a standard for organizations that set standards, against which existing initiatives can be tested. In addition, members must embrace the credibility principles of the ISEAL Alliance (see Box 1). This set of credibility principles illustrates the importance that ISEAL, as a meta-regulator, attaches to addressing the legitimacy/credibility gap.

Sustainability
Standards scheme owners clearly define and communicate their sustainability objectives and approach to achieving them. They make decisions that best advance these objectives.

Improvement
Standards scheme owners seek to understand their impacts and measure and demonstrate progress towards their intended outcomes. They regularly integrate learning and encourage innovation to increase benefits to people and the environment.

Relevance
Standards are fit for purpose. They address the most significant sustainability impacts of a product, process, business or service; only include requirements that contribute to their objectives; reflect best scientific understanding and relevant international norms; and are adapted where necessary to local conditions.

Rigour
All components of a standards system are structured to deliver quality outcomes. In particular, standards are set at a performance level that results in measurable progress towards the scheme's sustainability objectives, while assessments of compliance provide an accurate picture of whether an entity meets the standard's requirements.

Engagement
Standard-setters engage a balanced and representative group of stakeholders in standards development. Standards systems provide meaningful and accessible opportunities to participate in governance, assurance and monitoring and evaluation. They empower stakeholders with fair mechanisms to resolve complaints.

Impartiality
Standards systems identify and mitigate conflicts of interest throughout their operations, particularly in the assurance process and in governance. Transparency, accessibility and balanced representation contribute to impartiality.

Transparency
Standards systems make relevant information freely available about the development and content of the standard, how the system is governed, who is evaluated and under what process, impact information and the various ways in which stakeholders can engage.

Accessibility
To reduce barriers to implementation, standards systems minimise costs and overly burdensome requirements. They facilitate access to information about meeting the standard, training, and financial resources to build capacity throughout supply chains and for actors within the standards system.

Truthfulness
Claims and communications made by actors within standards systems and by certified entities about the benefits or impacts that derive from the system or from the purchase or use of a certified product or service are verifiable, not misleading, and enable an informed choice.

Efficiency
Standards systems refer to or collaborate with other credible schemes to improve consistency and efficiency in standards content and operating practices. They improve their viability through the application of sound revenue models and organisational management strategies.
Source: http://www.isealalliance.org/sites/default/files/Credibility%20Principles%20v1.0%20low%20res.pdf Full members of the ISEAL Alliance are VSS that have demonstrated a high level of compliance with different codes of good practice which have been developed by ISEAL. Granting of membership is based on independent evaluation and peer review processes. Currently the ISEAL Alliance has 14 full members, coming from 8. These include:  4C Association, a membership organization of coffee farmers, trade and industry and civil society actors which develop standards for sustainable coffee production  Accreditation Services International, an independent accreditation body offering international, third party accreditation for several VVS  Bonsucro, a not-for-profit initiative focusing on reducing the environmental and social impacts of sugar cane production.
 Fairtrade International (FLO), a Fairtrade labelling initiative helping producers in the global South to gain Fairtrade certification and develop market opportunities.
 The FSC, an international not-for-profit organization established to promote the responsible and sustainable management of the world's forests (supra, 2.1).
 GoodWeave, a non-profit organization aiming to end exploitative child labour in the carpet industry and to offer educational opportunities to children.
 IOAS, an independent, non-profit organization active in the organic community. It primarily provides accreditation and assessment services to certification bodies working in organic agriculture.
 The MSC, a certification and eco-labelling program for sustainable seafood from wild fisheries (supra, 2.1).
 The Responsible Jewellery Council, an international not-for-profit organization committed to promoting responsible ethical, human rights, social and environmental practices throughout the supply chain in the jewellery industry.
 Roundtable on Sustainable Biomaterials, which aims to ensure that the use of biomass and derived bio-products, including biofuels, deliver on their promises of climate change mitigation, economic development and energy security without causing environmental or social damage, such as deforestation and food insecurity.
 Social Accountability Accreditation Services (SAAS), which supports social responsibility by ensuring the implementation of credible social standards.
 The Rainforest Alliance, an NGO working to conserve biodiversity and ensure sustainable livelihoods by transforming land-use practices, business practices and consumer behaviour.
 The Union for Ethical BioTrade, a not-for-profit association that promotes the "Sourcing with Respect" of ingredients that come from native biodiversity. Its standards reflect the goals of the Convention on Biological Diversity (CBD), the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES), and the Millennium Development Goals.
 UTZ certified Farms, a coffee, tea and cocoa certifier.
In addition, ISEAL Alliance also has a system of associate members, which are VSS that are on a pathway to becoming full member and which have met some baseline requirements; and subscribers who are mainly VSS which align with ISEAL's vision. In total, ISEAL now has seven associate members and approx. 40 VSS subscribers such as GLOBALG.A.P. The interest of several VSS in ISEAL Alliance is becoming evident and its role as a meta-regulator is increasing.
According to Loconto and Fouilleux (2013) ISEAL Alliance attempts to develop an organizational field that legitimizes specific actors, tools and practices in the area of VSS. They aim to create distinctions in the VSS market with the aim of distinguishing between credible and legitimate VSS on the one hand and illegitimate ones on the other hand. As Reinecke et al. (2012, p. 804) point out, ISEAL's coordinating role makes it "a market watchdog protecting the legitimacy of the sustainability standards movement." As a result, several VSS systems are interested in becoming a member, in order to distinguish them from others. On the one hand, this creates more diversity in the market of VSS between members and non-members. On the other hand, it creates more harmonization and coordination for the sub-group which is a member.
One possible explanation why this form of meta-regulation generates interest is that it creates status hierarchies between VSS systems in the VSS market. Being an ISEAL member generates a status which VSS systems can use in the market. Following the work of Joel Poldony (1993Poldony ( , 2005 on status-based models of market competition, one could hypothesize that VSS systems seek status, via ISEAL Alliance, in a market characterized with significant confusion on the quality of the services they are delivering. High status, as Joel Podolny and many others applying his status-based model show, will lead to competitive advantages, which is of key importance in a fastly expanding market. Hence, the interest in meta-regulation which establishes high minimum standards and procedures might be driven by self-interest to gain market share. This status-seeking dynamic is also supported by studies which focus on competition between systems. Contrary to some expectations they do not necessarily lead to a race to the bottom of standards and procedures. Christine Overdevest (2010), for example, showed how FSC is playing a pivotal role in ratcheting up forest standards of rival industry-sponsored certification initiatives in the forest sector.

Conclusion
Voluntary sustainability standards have become a prominent instrument to govern supply chains according to a set of social, environmental and safety standards. Their rise is remarkable and they have generated significant attention both in terms of their potential to 'regulate' markets and their potential impact on trade . This paper engaged in the debate on the development of these systems from the perspective of a whole group of VSS, contrary to focusing on a single VSS. The paper first mapped the emergence, proliferation and global diffusion of VSS. This part showed an exponential growth of these systems as well as a significant, and fast growing, global diffusion. In a second part, it was argued that VSS significantly differ in their design in how they set and enforce standards. This institutional diversity analysis is important in the current debates on the credibility of these systems.
What emerges is a diverse and growing (both on the supply and demand-side) market of voluntary sustainability standards. This proliferation and diversification of VSS is creating market problems which the third part of the paper identified. These problems concern a credibility gap, increased costs to gain certification by multiple VSS-systems and a potential race to the bottom of VSS-systems. Next, the paper assessed two mechanisms which can facilitate more cooperation to address these market problems: mutual recognition and meta-regulation. It was argued that mutual recognition is especially suited to address cost issues, while meta-regulation might contribute to leveling the playing field within the market of VSS. The paper analyzed these mechanisms and showed that cooperation in the market of VSS is currently more developed with regard to addressing the credibility gap (meta-regulation) rather than with regard to addressing the cost-issue (mutual recognition).
The Leuven Centre for Global Governance Studies is an interdisciplinary research centre of the Humanities and Social Sciences recognized as a Centre of Excellence at the KU Leuven. It hosts researchers from law, economics, political science, history, philosophy and area studies. The Centre initiates and conducts interdisciplinary research on topics related to globalization, governance processes and multilateralism, with a particular focus on the following areas: (i) the European Union and global governance; (ii) human rights, democracy and rule of law; (iii) trade and sustainable development; (iv) peace and security; (v) global commons and outer space; (vi) federalism and multilevel governance; (vii) non-state actors and emerging powers. It hosts the InBev Baillet-Latour Chair EU-China and the Leuven India Focus.
In addition to its fundamental research activities the Centre carries out independent applied research and offers innovative policy advice and solutions to policy-makers.
In full recognition of the complex issues involved, the Centre approaches global governance from a multi-level and multi-actor perspective. The multi-level governance perspective takes the interactions between the various levels of governance (international, European, national, subnational, local) into account, with a particular emphasis on the multifaceted interactions between the United Nations System, the World Trade Organization, the European Union and other regional organizations/actors in global multilateral governance. The multi-actors perspective pertains to the roles and interactions of various actors at different governance levels, which includes public authorities, formal and informal international institutions, business enterprises and non-governmental organizations.