Assessing Global Governance in the Area of Human Trafficking Through Civil Society Participation in its Regime Complex

Human trafficking (HT) is a major international problem that states still struggle to tackle. Against this backdrop, this working paper aims at analysing the architecture of HT global governance and at suggesting ways for improvement. It, among others, gives recommendations for actors such as the European Union, which is very active in HT governance. <br><br>In order to do so, the working paper focuses on the participation of civil society organisations in HT global governance, seeing them as important drivers of governance at the global level. After presenting the phenomenon of HT and its governance puzzle, a first section characterises such governance, identifying and describing the HT regime complex by using civil society participation as a proxy. A second section analyses the main actors of the regime complex. The conclusion draws recommendations and identifies paths for further research.


Introduction: the research puzzle
The overall goal of this working paper is to produce knowledge on global governance in the area of human trafficking (HT), with the associated objective to produce policy recommendations for key players and in particular the European Union (EU).First, HT needs to be defined.

What is human trafficking?
Human trafficking implies the traffic of human beings for several different purposes including sexual exploitation, forced labour, organ removal or for other purposes such as criminal exploitation (for drug transport, illegal traffics, etc.).For all these purposes, the economic motivations of the traffickers are a main defining criteria, together with coercion.Synthetizing both aspects, exploitation is the key term used to characterise HT (Bursch, 2014: 78) in the first place, since "human trafficking generates huge illicit profits, seeing human beings as commodities with the purpose of exploiting them in several activities" (Gaspari, 2019: 47-48).While often associated with HT, a "transnational dimension" of such exploitation is not a defining criterion of the phenomenon: HT also very often takes place within state boundaries.
HT is in practical terms a highly ramified issue that implies different recruitment modes (transportation, harbouring and receipt of persons), means (threat, force, coercion, abduction, fraud, deception, abuse of power or vulnerability, giving or receiving of payment) and purposes of exploitation (prostitution or other forms of sexual exploitation, forced labour or services, slavery and similar practices, servitude, the removal of organs) (Rubio Grundell, 2015: 2).Because of the wide scope of the phenomenon (see Box 1 and also Box 4), HT poses a major threat to humanity: "over the last thirty years, human trafficking has become one of the issues of major concern of the international community, of human rights activists, of regional organisations, and governments" (Gaspari, 2019: 47).

Box 1. Key facts on human trafficking worldwide
Data on HT are not easy to gather because of, among others, the difficulty to detect victims (see below).At the international level general estimates are published within the Global Estimates of Modern Slavery: Forced Labour and Forced Marriages, issued in 2017 by the International Labour Organization (ILO), while data on detected victims are found in the Global Report on Trafficking in Persons 2018 published by the United Nations Office on Drugs and Crime (UNODC).According to the ILO, in 2017, out of the 24.9 million people trapped in forced labour worldwide, 16 million people were exploited in the private sector for domestic work, construction or agriculture; 4.8 million persons were in forced sexual exploitation, and 4.1 million persons were in forced labour imposed by state authorities (ILO, 2017: 10).Women and girls are disproportionately affected, accounting for 99% of victims in the commercial sex industry, and 58% in other sectors.According to the 2018 UNODC report, the number of victims detected in 2016 was 24,000 (UNODC, 2018: 21), while they total 225,000 over the period 2005-2018.Women and young girls represented around 70 per cent of the total number of detected victims.While data on HT is difficult to compile, comparing the two sets of data reveals the existing gap between the number of detected victims versus the real total number.Sources: ILO, 2017; UNODC 2018.
While several studies depict HT when taking place under exceptional conditions such as conflicts (UNODC, 2017;Bird and Reitano, 2019) that aggravate the trafficking dynamic, focusing exclusively on these studies might be misleading as trafficking is actually taking place everywhere: "one thing is certain: no country and area of the world are immune to human trafficking.It is not a problem so distant Robert Schuman Centre for Advanced Studies Working Papers from us, instead it is among us, behind the corner, even within our national borders" (Gaspari, 2019: 48).
However, despite the fact that HT is commonplace, solving the issue is not an easy task.One important problem is identifying the victims: "trafficking in persons continues to be, to a large extent, a 'hidden ' and underreported crime" (GloAct, 2018: 1).This difficulty is well known: "a review of the literature pertaining to human trafficking reveals that human trafficking is a difficult crime to detect and prevent" (Portland State University, 2011: 2).One problem is that networks of individuals are involved in trafficking practices, blurring and diluting the supply chain and complicating the identification of perpetrators.Another key problem is the criminalisation of victims, meaning that they are less inclined to look for assistance from national authorities: "victims of trafficking are often treated as criminals by authorities: after being abducted in their countries of origin they may be arrested, detained, and charged in countries of destination for working illegally, frequently as prostitutes, and for having false documentation" (Portland State University, 2011: 3).This means that some victims of trafficking can be victimised twice (Villacampa, 2019) and "the lack of trust between victims and authorities has led to the increased victimisation of victims of human trafficking through the criminalisation of irregular migration (which effectively turns the victims into criminal offenders)" (Fiducia, 2015: 1).
Paradoxically, the lack of detection of HT renders it a worthwhile activity: "this highly profitable trade poses a relatively low risk of capture or conviction when compared with trade in drugs and arms" (Usman, 2014: 282).Developing governance instruments to correct this paradox is therefore crucial.

HT governance: the need for multiple dimensions
The challenge of developing governance to tackle HT is not new.However, recognizing the scope of the phenomenon, it has recently taken on new dimensions with a multiplication of the policy levels involved (local, national, regional, international), the increasing relevance of actors beyond states and the growing complexity of issues (classical security threats, for instance, becoming embedded in broader dilemmas such as human rights or gender inequalities).
While trafficking has become more complex, governments continue to apply classical recipes to deal with it, such as creating networks of top-down international binding agreements with the underlying aim to "take (intergovernmental) networks to fight (trafficking) networks" 1 .In 2000, states negotiated a key landmark agreement on HT: the Protocol to Prevent, Suppress and Punish Trafficking in Persons, Especially Women and Children.
The Protocol has been significant in providing the first comprehensive definition of HT, covering all trafficking forms (Gaspari, 2019: 6).In particular, it states that "trafficking in persons shall mean the recruitment, transportation, transfer, harbouring or receipt of persons, by means of the threat or use of force or other forms of coercion, of abduction, of fraud, of deception, of the abuse of power or of a position of vulnerability or of the giving or receiving of payments or benefits to achieve the consent of a person having control over another person, for the purpose of exploitation.Exploitation shall include, at a minimum, the exploitation of the prostitution of others or other forms of sexual exploitation, forced labour or services, slavery or practices similar to slavery, servitude or the removal of organs" (United Nations, 2000: art.3).
However, intergovernmental politics are proving inefficient.In particular, most states have chosen to reassert control over borders to stop trafficking flows, leading to counterproductive effects.The everincreasing number of deaths of migrants is a constant and tragic reminder of the need to find alternative solutions (Gallagher, 2001;Miller and Baumeister, 2013).In addition to death, "many smuggled migrants can fall into the hands of traffickers and be exploited to repay the debt they incurred with smugglers to cross borders, being in a situation of debt bondage" (UNODC, 2016: 60, cited in Gaspari, 2019: 50).Some underline "the role governments in destination countries may play in causing trafficking through imposing restrictive migration regimes that render migrants vulnerable to traffickers" (O'Brien, 2016: 205).
Intergovernmental policies have also run the risk to be disconnected from field realities, with governments struggling to reach their targets, and more precisely the victims of trafficking (being vulnerable populations such as women, children or local and indigenous communities).
Human rights issues have always been an important dimension of HT (Gallagher, 2009) and recent efforts have been made to include them in global governance.HT interferes with several issue areas within a regime complex (see below) and "efforts must be made to ensure better understanding of the coherence between human rights law, refugee law, labour law and other relevant bodies of law, and the need to bring diverse but complementary instruments to bear on the trafficking challenge, not only as reference tools but as an international framework that needs to be implemented on the ground" (ICAT, 2012: 14).The Sustainable Development Goals (SDGs) (see Box 2) show a promising, multiple understanding of HT, that has however still to be implemented.

Box 2. Human trafficking and the SDGs
Out of the 17 SDGs, trafficking in persons is specifically mentioned in three targets under the three goals: 5 (Gender Equality), 8 (Decent Work and Economic Growth) and 16 (Peace Justice and Strong Institutions).
Target 5.2 -Eliminate all forms of violence against all women and girls in the public and private spheres, including trafficking and sexual and other types of exploitation.
Target 8.7 -Take immediate and effective measures to eradicate forced labour, end modern slavery and human trafficking and secure the prohibition and elimination of the worst forms of child labour, including recruitment and use of child soldiers, and by 2025 to end child labour in all its forms.Target 16.2 -End abuse, exploitation, trafficking and all forms of violence against and torture of children.UNODC is responsible for its implementation.This means that, through the SDGs, governments recognise the multi-dimensions of the trafficking issue.They also recognise the particular vulnerability of children as targets of human trafficking.Sources: United Nations, 2015.For more on the SDGs and HT see (ICAT, 2018) The EU as a leader in HT governance?
While intergovernmental politics are stagnating, the EU aims at being a leading global actor on trafficking issues."the European Union has dealt with human trafficking since the 90s.It entered into legislation of the EU indirectly since the abolition of internal frontiers and the creation of a space in which movements of capital, people, services, and goods became free, which would have made transnational crime easier" (Gaspari, 2019: 61).The European Union has developed several policy instruments to tackle HT (see Box 3) and, as in other fields (Oberthür and Rabitz, 2014), is trying to play a leadership role in HT governance, especially at the international level.The European Union plays a central role due to the importance of the phenomenon of trafficking within and across its borders (see Box 4), that is evidences thanks to the important trafficking tracing capacities of the EU.Data confirms the importance of the European continent with Europe being: "the area of the world with the highest number of citizenships of human trafficking survivors... with 137 different detected nationalities" (UNODC 2016: 5, cited in Gaspari, 2019: 56).

Box 4. Key facts on human trafficking in Europe
-20 532 victims were registered in the 28 EU Member States over the two years 2015-2016.
-68 % of registered victims were female -56 % of registered victims were trafficked for sexual exploitation -95 % of registered victims of sexual exploitation were female.
-23% of registered victims who were trafficked for sexual exploitation were children.-26% of the registered victims were trafficked for labour exploitation.-18% were trafficked for 'other' forms of exploitation.
-23 % of registered victims were children.-44% of registered victims were EU citizens.
-Over 70 % of traffickers were male -Around three quarters of actions against supposed traffickers concerned trafficking for sexual exploitation.
-Over three quarters of supposed traffickers held EU citizenship.Source: Eurostat, 2018.The EU also plays an important role because of its special position as a defender of human rights, although its HT policies have been questioned.Just as other governments in the past: "the incorporation of trade in human beings into 'migration', especially irregular migration, is one of the main reasons for the lack of success of EU anti-trafficking policies" (Rubio Grundell, 2015: 4).This has led the EU to consider the issue through policy and criminal matters, rather than through protection and assistance.The situation has now improved and human rights are becoming more central to EU policies: "the legal framework currently in force marks a significant shift in EU anti-trafficking measures: they have moved from a criminal and migration control approach to one that also includes human rights concerns" (Rubio Grundell, 2015: 8).
The EU wishes to engage with international players (European Commission, NA), which makes it an ideal candidate to express new views and strategies for HT global governance.The willingness to engage is reciprocal with several United Nations agencies having given advice to the EU on its Human Trafficking Directive (United Nations, 2011).
In this role, the EU therefore embraces a unique proactive stance (Simmons and DiSilvestro, 2014), orienting its foreign policy actions towards promoting bottom-up governance, whereby all actors, and especially actors from civil society (CS), are invited to contribute to governance efforts.On HT, the 2012-2016 EU Strategy (see also Box 3) recognised the need to, among others, widen the knowledge base and give more rights to victims (European Commission, 2012).
Researchers on the EU have noted that: "it is important to involve a broad spectrum of individuals from different backgrounds.Therefore, civil society organisations involved in anti-trafficking and nongovernmental organisations, who have extensive knowledge of the reality of the trade in human beings, must be included in all stages of the policy-making process, from agenda-setting to implementation.It is crucial, however, that measures are taken to ensure that the EU's civil society interface does not systematically favour certain voices from the anti-trafficking scene, while excluding others" (Rubio Grundell, 2015: 9).These calls are constantly reiterated, most of the time with little effect.Indeed, while the premise to include more CS actors is a promising one as they play an important role in shaping, implementing and enforcing the adopted measures, there exists very little knowledge of their involvement in HT governance.

Why CS is important and research is needed
CS actors are key for HT.Firstly, because the victims of HT play a role at the beginning of the supply chain, suffering from misappropriation, and could, if empowered, react against traffickers.The 'situated knowledges' (Haraway, 1988) that these victims of trafficking hold are key and are therefore likely to be the basis of important policy solutions to be applied at the international level.In HT, victims are known for being "better suited to formulating contextually appropriate policy and enforcement responses" (Weitzer, 2014).
The problem is that so far the views of victims have been marginalised as they represent minorities: the local and gender dimensions of victims are particularly important, with women as the main victims of HT (Kempadoo et al., 2012;Green, 2012;Chong Gutierrez and Clark, 2016; see also Box 4).The fact that "the majority of literature on human trafficking originates within feminist schools of thought" (Usman, 2014: 283) has contributed towards taking these gender dimensions into account and should be encouraged.
Secondly, CS organisations (CSOs) are essential as gatekeepers with respect to state policies and behaviour.Indeed, "responses which rely on the state, or approach the issue through a criminal lens, should be treated with caution" (Bird and Reitano, 2019: 1).In particular, current governance efforts based on states come with a number of problems.One of these problems can be governmental corruption (Portland State University, 2011: 2), or, as mentioned above, the risk that the rights of victims can be violated by governments who penalise victims twice (Bachaka, 2017).
Moreover, an efficient approach to combat HT "relies heavily on well-functioning government systems with strong civil society networks and community infrastructureschooling, healthcare, employment opportunities and effective criminal justice systems" (Bird and Reitano, 2019: 9), which are not always available within states.These elements are in particular problematic for a number of weak states, and for specific situations such as conflict areas: "international efforts to combat trade in Robert Schuman Centre for Advanced Studies Working Papers persons must recognise the limited role that states can be realistically expected to play in responses to TIP in conflict areas, where they have by definition at least partially lost control" (Bird and Reitano, 2019: 9).This has led civil society to participate to the implementation of policies, very often replacing the fundamental role of states through "the altruistic work of civil society organisations" (Rubio Grundell, 2015: 10).However, CSOs are not necessarily invited to define state policies.To fill this gap, this working paper is rather interested in looking at the participation of CSOs to decision-making up-stream, and not so much at the down-stream effects of CS organisations once policies have been adopted.
Investigating the engagement of CS is also key as there is currently a risk to engage only a limited number of CS organisations.The lack of comprehensive discussion generates some inaccurate perceptions of trafficking.In a recent study of national civil servants dealing with trafficking victims, Schwarz shows how "for most interviewees, the exclusive meaning of human trafficking was sex trafficking" (Schwarz, 2019: 487).This incomplete perception of HT and HT victims can be relayed by specific CSOs as explained by experts on communication: "(some CSOs) campaigns construct a narrow understanding of the problem through the depiction of 'ideal offenders'.In particular, a strong focus on the demand for commercial sex as causative of human trafficking serves to obscure the problematic role of consumerism in a wide range of industries, and perpetuates an understanding of trafficking that fails to draw a necessary distinction between the demand for labour, and the demand for 'exploitable ' labour" (O'Brien, 2016: 205).This also poses important problems for victims who "believe they cannot access victim services unless they fit the dominant idea of a 'pure victim'" (Hoyle, Bosworth andDempsey, 2011: 32, cited in O'Brien, 2016: 206).
The problem is that we are currently lacking an overall description and assessment of which CS actors are involved in global HT governance.As governments, and especially the EU, are trying to rethink HT politics, this appears as an opportune time for "mapping the field" to provide an overview, serving as a "map which might guide more holistic forms of policy co-creation, outreach and engagement" (Kidwell, 2020: 343).This is what this working paper is about.
It is organised as follows.A first section (Section 1) presents the state of the art on civil society participation to global governance, with a specific focus on HT.While the project has an empirical basis, it also aims at developing a new methodology to characterise global governance architecture according to the participation patterns of non-state actors.This methodology is presented in Section 2. A third section (Section 3) characterises HT global governance, in particular looking at the HT regime complex and using civil society participation as a proxy.A fourth section (Section 4) characterises the main actors of the regime complex.

State of the art: HT global governance as a regime complex and civil society participation
The working paper builds on three strands of the academic literature.

Intergovernmental politics, regime complexes, and the recent call for polycentricity
Until the beginning of 2000, international issues were organised around "international regimes" (Krasner, 1983), taking the shape of unique international organisations or treaties.However, with the proliferation of international institutions and the broadening of their scope, no such thing as a unique international regime for each international issue exists anymore.To describe this change in global governance patterns, Kim (2019) refers to three complementary phenomena: institutional fragmentation, polycentricity, and complexity.
Most international issues are now evolving within what has been labelled "regime complexes" defined as "a network of three or more international regimes that relate to a common subject matter; exhibit overlapping membership; and generate substantive, normative, or operative interactions recognised as potentially problematic whether or not they are managed effectively" (Orsini et al., 2013: 29).The specificity of regime complexes is to have effects that are structural, meaning that they go beyond the mere sum of effects of their different elements.They also put the focus on interactions between international institutions: "with the rise of research interest in regime complexes and governance architectures over the past decade, the analytical focus has started to shift away from networks of actors to networks of institutions" (Burch et al., 2019, cited in Kim, 2019: 7).
Many regime complexes have emerged around a high number of topics in international relations (refugees, climate change, etc.) including human trafficking.Indeed, HT issues are now dealt with by no less than four international regimes on: (i) migration; (ii) labour; (iii) human rights; and (iv) organised crime, involving a very diverse set of international arenas including, among others, meetings of the International Organization for Migration (IOM), the International Labour Organization or the United National Office on Drugs and Crime (Gomez-Mera 2016, see also below, Section 2.2).This working paper precisely focuses on this regime complex.
The current challenge of global governance, HT governance included, therefore becomes the coordination of all these international institutions: "in other words, just like states had to be brought together with the help of international institutions, the myriad of international institutions themselves are in need of coordination for their effectiveness.This is an ultimate policy objective of the emerging field of research on global governance networks" (Kim, 2019: 7).
Despite the multiplicity of arenas and institutions, intergovernmental politics have however proven to hardly cover the scope of regime complexes (Krisch, 2017).This has led to the recognition of the importance of "polycentricity" (Jordan et al., 2018) as a new governance paradigm, whereby all actors from CS, and not just governmental ones, at all levels, can participate to governance efforts, by promoting awareness and action at all scales.

Transnational studies and their focus on CS actors beyond states
Surprisingly, in academia, calls for polycentricity have so far never been fully operationalised within transnational studies, a subfield of international studies specially dedicated to actors beyond states.The question of the involvement of CS actors and of their potential to fill the global governance gap is not new.But former studies, very diverse in terms of analytical perspectives, all share two important shortcomings.First, in these studies, most scholars still look at transnational participation in respect of unique regimes and not regime complexes, therefore missing the global picture.Instead, this working paper aims at looking at all the different international arenas on HT and at CS cross-participation within them.I have shown evidence with former preliminary research that CS participation within regime complexes follows the specific logics (Orsini, 2017) this working paper builds on.Second, former studies look at isolated cases of CS attempts to contribute to global governance with for instance research on precise transnational advocacy coalitions or epistemic communities (among others: Keck and Sikkink, 1998;Betsill and Corell, 2007).This imposes drastic selection processes with regard to the actorsto be followed with, most of the time, questions on the selection process of these actors.Instead, this working paper aims at a systematic analysis of all CS actors within regime complexes, and at developing a methodology to identify key actors.
While most studies of CS organisations are qualitative, focusing on precise case studies, I am suggesting to build a database of CS participation within the HT regime complex.Moreover, I embrace the objective of studying non-state actors through their strategies within regime complexes, to better grasp how they shape interactions between the international institutions involved in the regime complexes.
Such CS strategies within regime complexes can take three forms: forum shopping, forum shifting and forum linking.Former research has identified forum shopping and forum shifting as two important Robert Schuman Centre for Advanced Studies Working Papers strategies used by governments in a context of institutional fragmentation, which could potentially be replicated by non-state actors (Kautto, 2009: 106, see also Haufler, 2009: 128).More precisely: forum shopping is the strategic use of different institutional settings to make progress on a given agenda (Busch, 2007;Murphy and Kellow, 2013;Raustiala and Victor, 2004: 299).Through forum shopping, actors "seek out the forum most favourable to their interests" (Raustiala and Victor, 2004: 280).
-Forum shifting is the changing of discussion forum, i.e., moving the debate on a particular issue to an arena that would be more favourable to an actor's interest.It entails favouring one venue over the other (Helfer, 2004).
In addition to these two strategies, I established a third one that is located at the other end of the spectrum: forum linking (Orsini, 2013).Some non-state actors may actually link and therefore integrate different forums, by following them all acidulously.By doing so, they vehiculate a common normative frame throughout the different forums and may suggest a division of labour between the different institutions of a complex, for example according to specific topics, governance functions, or geographical areas and jurisdictions.
These strategies create integration and fragmentation dynamics within regime complexes, giving different levels of importance to their different institutions.While we lack methods to measure regime complexity (Kim, 2019), I am suggesting to use non-state actors' dynamics within regime complexes to characterise their main elements and shape (see below).The participation of civil society within regime complexes also raises broader debates on power and participation within global governance.

The literature on power and non-state actors' participation in international negotiations
International negotiators have always had to pay a "negotiation burden" (Muñoz et al., 2009) to take part to the negotiations of single international regimes, as has been more precisely noticed in international environmental politics.It is commonplace that the negotiation of global agreements takes several years, requires dozens of official meetings and informal preparation sessions, gets subdivided into dozens of working groups, and necessitates large negotiation delegations.In addition to quantitative burdens, a strong expertise background is also needed to capture the content of the negotiations and an ethnographic account of conferences of the parties to global agreements is often useful to navigate through the diversity of actors and events composing such conferences (Campbell et al., 2014).
States have been so far at the centre of the study of regime complexes and negotiation dynamics for the excellent reason that they are the ones who take international decisions.Yet, legitimacy issues in global governance are very often also related to the issue of the participation of non-state actors to international political processes.While the participation of states to international negotiation processes is rather evident, because they can expect to have some sort of influence on the final result, and because they have developed well-trained foreign affairs representatives, this is not the case for non-state actors that cannot always warrantee a strong level of political significance, if any at all, and often have few resources to dedicate to international policy efforts.It is evident that not all international political actors have the same capacities to pay the negotiation burden of single negotiation processes.Such a lack of capacities has always posed strong issues of participation, representation and political influence.This in turn, has created issues of legitimacy for the overall political process leading to global agreements.As a result, studies have developed to propose strategies for weaker actors to improve their effective involvement in single negotiations (Chasek and Rajamani, 2001).
A recent important question is that of the effects of regime complexes as structures on international regimes, on power, and on participation dynamics.On the one hand, with the development of regime complexes, i.e. the negotiations of the same specific issue in different negotiating fora, the negotiation burden is increasing, even exponentially, for international negotiators.There exist studies of state politics in a context of regime complexity that have sensed that the most powerful actors were able to choose the direction of the negotiations while weaker actors were left aside (Alter and Meunier, 2009).Even developed countries have faced difficulties in coherently following regime complexes, not so much because of a lack of material resources, but because of difficulties of coordination (Morin and Orsini, 2014).As a result, states create internal coordination efforts, a phenomenon explained by, among others, Scott (2011), who looks at the management of fragmentation by states.All this tends to indicate that regime complexes increase the negotiation burden.As explained by Faude and Groβe-Kreul (2020: 433): "Benvenisti and Downs (2007, argue that the institutional fragmentation of global governance 'operates to sabotage the evolution of a more democratic and egalitarian international regulatory system' and, in the end, yields 'a regulatory order that reflects the interests of the powerful that they alone can alter'". On the other hand, regime complexes are often complex systems that can lead to unexpected outcomes as they increase the windows of opportunity, even for usually weak actors that use them as discursive areas: "they (regime complexes) enable actors marginalised within the international institution producing negative spillovers to demand inter-institutional justifications…in doing so, they enable normative progress in global governance" (Faude and Groβe-Kreul 2020: 433).Kuyper therefore argues that the democratisation of global governance should occur at the level of regime complexes because they empower weaker actors and enhance the realisation of the three core values of democratisation: "equal participation," "accountability," and "institutional revisability" (Kuyper 2014a cited in Faude and Groβe-Kreul 2020: 433).
Because debates on CS and regime complexes are lively, this working paper engages in the discussion and investigates more precisely three research questions with regard to HT global governance: who are the CS actors engaged in HT global governance?What does their involvement tell us about the characteristics of the HT global governance architecture?What does the nature of CS actors in the HT regime complex tell us about power relations within this complex?

Methodology: characterising regime complexes through CS participation
To tackle the above-mentioned challenges, the research is subdivided into three objectives.The first objective is to identify all CS actors involved in HT.This objective is descriptive and presents itself as a necessary step for the next two objectives.The second objective is to trace the participation of CS actors across the different elements of the HT regime complex and to characterise its architecture.This is important to get a sense of how HT issues are dealt with at the international level, and where.As explained by Kim, regime complexes have a certain level of organisation: "institutional complexity, emerging from self-organisation of the myriad institutions involved, is often organized" (Kim, 2019, 12), as I intend to reveal by studying CSOs.The third objective is to identify the main players of the HT regime complex.These actors have comprehensive knowledge of the HT issue at the global level and therefore play an important role in the complex.While characterising the global governance architecture of HT and CS participation, our final aim is to draw potential recommendations for the EU to better engage in HT governance globally.
Before detailing a clear methodology for each of these objectives, one also needs to consider how to interpret the results regarding CS participation to HT global governance, if these results are only developed on the HT case study: how to know if the level of involvement of CS actors is low or high?If the complex is integrated or fragmented?If its key players represent a high diversity of different interests or not?As explained by Kim, "the issue of measuring global institutional complexity is an area where there is little consensus.This is understandable, as there are no agreed ways of measuring complexity in the broader science of complex systems" (Mitchell, 2009, cited in Kim, 2019: 13).Because there exists no current scale to measure regime complexity, and in order to still enable such interpretation, I decide to include a comparative dimension in the study, comparing the results obtained Robert Schuman Centre for Advanced Studies Working Papers for the HT regime complex with the ones obtained for another, well-established, regime complex: the regime complex on genetic resources (GR).

Using the GR regime complex as a benchmark
Following Kim's reasoning in stating that "defining a network of institutions as either, for example, dense or sparse in absolute terms has proven difficult.Therefore, network analysis of what a single network looks like at a particular moment in time has not led to conclusive findings about topological properties of the system in question.When characterizing the structure of a global governance system, a more sensible approach would be to make a cross-system or longitudinal comparison for a relative assessment" (Beckfield, 2010;Kim, 2013;Gomez and Parigi, 2015;Greenhill andLupu, 2017, cited in Kim 2019, 11).I therefore decide to develop a comparison across regime complexes, comparing CS dynamics within the HT regime complex with CS dynamics within the GR regime complex, used as a baseline.
Indeed, the GR regime complex is the first to have been identified.It is a key regime complex, subject to a high number of studies and presenting a rather integrated and robust structure (see among others Morin and Orsini, 2014;Rabitz, 2018).The GR regime complex is currently evolving at the crossroads of the international regimes for agriculture, environment, property rights and trade (Raustiala and Victor, 2004;Elliot, 2017), made up of numerous international arenas such as the Nagoya Protocol to the Convention on biological diversity, the World Intellectual Property Organization Committee on genetic resources, or the International Treaty to the United Nations Food and Agriculture Organization.
Both the HT and GR have moreover similar starting dates, meaning that their maturity can be compared over time.While the precise shape of the GR regime complex has not been clearly described, it is known to be integrated and robust.In any case, it will still be possible to compare one regime complex to the other: "one can only define a system structure in comparison to another or to itself at a different point in time.This is to say, we may only reasonably suggest that a global governance network is relatively more fragmented, polycentric, or complex than anotherbut not in absolute terms.Therefore, comparative or longitudinal analysis is imperative to put observed structural features into perspective" (Kim, 2019: 17).

Methodology for the first objective: identifying all actors involved in HT
Objective 1 requires compiling an inventory of all civil society actors taking part to decision making within the global governance structure on HT.In order to do so, I build a database 2 on CSOs participating as observers to the negotiations of the different international institutions shaping the HT regime complex.This database is meant to be a systematic and comprehensive inventory of CS actors.The lists of participants to most of the international negotiation meetings included in the database are publicly available upon request to the corresponding international secretariats.The temporal scope of the database is defined as 2000-2019.2000 is chosen as a starting date as it marks the adoption of the HT Protocol.
The international institutions included in the HT database were selected on the basis of a number of criteria.First, they are known to be part to the HT regime complex (see Gomez-Mera, 2016;Aradau, 2013).Second, data is available on CS participation to their negotiation meetings through the lists of participants.Third, these institutions are party to the Inter-Agency Coordination Group against Trafficking in Persons (ICAT) created within the United Nations by the Economic and Social Council resolution 2006/27 and General Assembly Resolution 61/180.More precisely, they are ICAT members, members of the ICAT working group and have been ICAT chairs.Fourth, they are regularly mentioned in international reports, documents and academic publications on HT.

2
The databases elaborated for this project are available upon request.
More precisely, the database on the HT regime complex includes 138 meetings, since 2000, of the United Nations High Commissioner for Refugees (UNHCR), the International Labour Organisation (ILO), the International Organization for Migration (IOM), the United Nations Office on Drugs and Crime (UNODC), and the Office of the United Nations High Commissioner for Human Rights (OHCHR).A list of all meetings included in the HT database is available in Annex 1. ILO lists of participants were available only since 2004.Moreover, the first session of the Regular Session of the Human Rights Council convened for the first time in 2006.The postponed starting dates for both venues will be taken into account when analysing the results.
All these organisations have clear mandates and missions on HT.The UNHCR, among others, issues regularly reports and guidelines, not binding for states, to better interpret international migration instruments in the light of human trafficking (such as the 2006 Guidelines).The ILO has several conventions specifically dedicated to forced labour, such as Convention n°29 of 1930, enhanced by a Protocol in 2014, which entered into force in 2016.The IOM supervises the interactions between migration and human trafficking, while the UNODC supervises the relations between transnational crime and HT.The OHCHR produces guidelines such as the 2002 United Nations Principles and Guidelines on Human Rights and Human Trafficking.The United Nations, under the OHCHR, in 2004, also established through its decision 2004/110 a Special Rapporteur for trafficking in persons, especially women and children 3 .
For comparative purposes (see Section 2.1), I also build a database on CS participation to the GR regime complex.The international regimes included in this second database were selected on the basis of two criteria: (i) their mentioning in the literature on GR (Raustiala and Victor, 2004;Morin and Orsini, 2014;Rabitz 2018); (ii) the availability of the data (meaning that the World Trade Organisation has been excluded as a forum).The database on the GR regime complex includes 68 meetings, since 2001, of the Convention on biological diversity (CBD), of the Food and Agriculture Organisation (FAO), and of the World Intellectual Property Organisation (WIPO).A list of all meetings included in the GR database is available in Annex 2.

Methodology for the second objective: characterising the participation of CS actors in the HT regime complex
Once the databases are built, I analyse CSOs' involvement in the regime complexes and their corresponding effects on the shape of these regime complexes.In order to do so, I analyse the following factors: -Participation.To characterise participation within the identified international regimes and regime complexes, I look at how many delegates from CSOs participated to their negotiation meetings.I also analyse the diversity of these organisations (how many different organisations are there in total for instance).
-Follow-up.To characterise the expertise of CS actors, I analyse to which extent they have a consistent follow-up of the identified negotiations over time.This enables to identify the CS organisations that have a greater knowledge of the negotiations at hand, the ones being more present considered as being more expert on the corresponding issues.
-Integration.To characterise the degree of integration of the regime complexes, I look at how many CSOs are multi-fora, meaning that they follow more than one institutional venue, to get an idea of the degree of integration of the regime complex.
-Shape.To characterise the shape of the regime complexes, I look more precisely at the number and nature of the links created by all CS actors across the international institutions of the regime complexes.This gives an idea of the shape of the regime complex.Coding CS strategies (forum 3 For the mandate and main activities of these organisations, also with regard to human trafficking, see (ICAT, 2010).Robert Schuman Centre for Advanced Studies Working Papers shopping, forum shifting and forum linking) is also used to characterise dynamics within the studied regime complexes.

Methodology for the third objective: identifying and characterising the main players of the HT regime complex
The main players of each regime complex are defined as: (i) the multi-fora CS actors (they are the ones following more fora); (ii) that demonstrate the highest level of participation (they are the ones following more meetings).On the basis of these criteria, I have selected the top 15 CSOs of each regime complex within my databases.
To know more about their behaviour within each regime complex, I coded their period of attendance and their strategy.Period of attendance enables to test the hypothesis of the negotiation burden: if key organisations are the ones following the negotiation process since the beginning, this confirms the need for a long-term engagement and therefore a certain degree of resources.It also enables to test whether regime complexity is an old phenomenon or a recent one (if multi-fora organisations appear recently or not).
To know more about their nature, I code their headquarters and type.I used the Yearbook of International Organizations to code such data (except for a few organisations not included in the Yearbook and for which data was gathered through the Internet).Headquarters is used to question the origin of these groups (global south, global north) and their location (is there a geographical diversity of representation?).Type is used to characterise the nature of their interests and their topic of specialisation.For Type, I use the typology of the Yearbook of International Organizations and of the United Nations 'major groups' categorisation 4 .I now turn to the analysis of the results.

Characterising the human trafficking regime complex
Table 1 below presents some characteristics of the HT and GR regime complexes, based on the built databases.
Participation within the different institutions of the HT regime complex can be considered as high with, on average, about 80 participants from civil society to each negotiation meeting, compared to 54 for GR.There are however important discrepancies between the different institutional settings of the HT complex.The two extremes are the ILO, with 38 meetings attended by only 9 different non-state actors, whereas the OHCHR has been organising a similar number of meetings (42) followed by 995 different CS organisations.The number of different non-state actors is roughly the same for both regime complexes, suggesting that at some point, the number of non-state actors interested in a particular issue area of global governance becomes stable over time.
Regarding follow-up, there is a high turnover for both regime complexes.32.86% of CSOs involved in the HT regime complex only attend one meeting, while their share is 65.19% for the GR regime complex.This tends to indicate that the follow-up of negotiation meetings by civil society within the HT regime complex is more comprehensive.
Regarding integration, among these organisations, 84.39% of CSOs attending more than one meeting in the HT regime complex attend only one forum.Their share is similar for the GR regime complex Assessing global governance in the area of human trafficking through civil society participation in its regime complex European University Institute 13 (81.25%)even if slightly lower, meaning that integration across regimes through non-state actors is found to be higher for GR.As a consequence, 15.6% of civil society organisations that attended more than one meeting of the HT regime complex are multi-fora, whereas 18.75% are multi-fora in the GR regime complex.This confirms the existence of an HT regime complex that is integrated, even if less than the GR one.190,[9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26]300,9 222,727,497,218,1  Moreover, in the GR regime complex, 87.77% of multi-fora organisations are following two fora out of three, while 12.22% are following three fora.In the HT regime complex, patterns show that the regime complex seems to be less integrated: 81.29% of multi-fora organisations follow two fora out of five; 14.83% follow 3 fora; 3.87% follow 4 fora, and no CSOs follows the five mentioned fora of the HT regime complex.This means that no CS organisation integrates all five fora at the same time.It also means that the links between the different elements of the HT regime complex are mostly drawn by civil society actors on a bilateral basis.This pattern is also true for the GR regime complex.This tends to indicate that CSOs are contributing to regime complexity by mostly linking international fora two by two.It also means that the few civil society organisations that perform multi-fora participation to more Robert Schuman Centre for Advanced Studies Working Papers than two fora are likely to play a key role, benefiting from a privileged position where they can view more comprehensively the regime complexes.It also tends to confirm the negotiation-burden hypothesis for regime complexes, as CSOs cannot easily multiply the number of fora they follow.
Regarding the shape of the regime complexes on the basis of CS participation, it is possible to draw a network representing each regime complex, where institutions are nodes linked by CS participation. Figure 1 presents these CSOs-type ties shaping the HT and GR regime complexes.In the GR regime complex, the CBD, appears as the central institution around which civil society organisations create links.CSOs following the CBD process are particularly well connected on the one hand to WIPO, on the other hand to the FAO.Would three-fora multi-fora organisation not exist, the FAO would be rather marginalised in the complex.This confirms the important role played by multifora organisations in integrating regime complexes.
In the HT regime complex, the OHCHR appears as the central institution around which CSOs create links.This is surprising, especially when one considers that data for the OHCHR was collected from 2006, not 2000.Data confirms the importance of the norm of human rights within HT governance.CSOs following the OHCHR process are particularly well connected on the one hand to the UNHCR, on the other hand to the UNODC.These three institutions are the most central.The IOM is connected to the complex, but mostly only through the three-fora multi-fora non-state actors.To the contrary, the ILO is clearly underrepresented within the complex.This last point is surprising as positive spill-overs exist between the ILO and the other international institutions of the complex: "several conventions adopted within the ILO, for example, reduce incentives for trafficking in persons and thus support the governance objective of the UN Trafficking Protocol.The latter, in turn, supports the governance objective of the ILO because preventing human trafficking reduces forced labor practices" (Gómez-Mera, 2016: 584).
Both the apparent marginalisation of the ILO and the IOM could be problematic as "the evidence showed that it is important to tackle exploitative practices in the labour market at an early stage in the exploitative continuum.Targeted prevention efforts and services to migrant workers are the key to protecting their rights and eradicating the phenomenon from the outset" (Fiducia, 2015: 2).This confirms the problem underlined earlier that restrictive policies towards migrant workers and domestic workers emphasise HT: "research on labour exploitation showed that restrictive immigration policies make workers vulnerable to exploitation and abuse from the side of their employers.This concerns particularly undocumented workers and workers on tied visa" (DemandAt, 2017: 9).Re-tying both elements (ILO and IOM) to the complex would be an interesting approach to take.
Figure 1 also shows that the international institutions within which more negotiation meetings take place are not necessarily the central ones in regime complexes.In the GR regime complex, the CBD is central while it organised much less meetings than WIPO.In HT, the ILO convened many meetings but is marginalised within the regime complex.This confirms that the dynamics of regime complexes are different from the dynamics of their respective parts.
To complement this quantitative approach of civil society organisations in regime complexes, the next section looks at their nature.Regarding attendance, for the HT regime complex, CS organisations appear to have been always engaged for a long period of time, starting their engagement in the early 2000s (all civil society organisations engaged in 3 fora started following the negotiations in early 2000s), with the latest one engaging in 2004, and still following the negotiations now.There is a more visible discrepancy between the follow-up capacity of the different multi-fora organisations in the GR regime complex: this capacity is divided by two, when one compares the attendance and attendance period of the top 4 organisations.This means that the multi-fora group belonging to the HT regime complex is more balanced in terms of follow-up capacity, with all key actors able to ensure participation.This observation confirms that the regime complex is rather integrated in its different elements linked by CS organisations and that participation to regime complexes in general requires a certain level of resources.Multi-fora participation is not practiced by newcomers.

Identifying the key players in the human trafficking regime complex
Regarding strategy, most key CSOs engaged in the studied regime complexes practice forum linking, while within the HT regime complex, several shift towards the OHCHR, underlying the human rights dimension of HT global governance.Within the HT regime complex, it is visible that economic actors navigate the complex differently, being mostly based at the ILO and practicing forum shopping to other fora.These actors could be mobilised and invited to participate more consistently to other fora.
Regarding headquarters, geographical diversity between the global south and the global north is more visible within the GR regime complex, while the HT regime complex is highly dominated by CS organisations based in the global north.Engaging more organisations from the global south could be a solution to tackle the HT problem more comprehensively.Within the HT regime complex, it is noticeable that only three organisations are not based on the European continent: 1/3 are based in Switzerland and 1/5 in the EU.This confirms the key role the EU could play in reaching out to these organisations.
Regarding type, the GR regime complex presents a better mix of types of key actors compared to the HT regime complex.It sees a balanced representation of for-profit and not-for-profit organisations, with business involved in the GR regime complex, while it is very discreet in the HT one.There is a tendency in the GR complex to see business as very present, while groups representing minorities such as indigenous peoples' groups are less active in following three fora.While engaging more businesses, a certain balance should be maintained.
The GR regime complex also sees the involvement of more specialised civil society organisations.In HT, universal non-specialised organisations (such as Amnesty International or Human Rights Watch) are a majority, presenting no specific specialisation in human trafficking, while GR is populated with organisations specifically dedicated to the GR issue (such as the International Seed Federation and Genetic Resources Action International).
In HT several key CS organisations are faith-based organisations.The literature rarely mentions such organisations, while they could actually play a crucial role in preventing HT, by reinforcing local ties: "greater understanding of factors that increase individual and community resilience to trafficking is needed to ensure support is appropriately directed" (Bird and Reinato, 2019: 1).Human rights and youth organisations are equally well represented, confirming the importance of human rights' norms within HT governance and the importance to recognise children and youth as HT victims.Humanitarian organisations are also present, which might, again, play against a conception of human trafficking insisting on its criminal dimension: "seeking to address trafficking in conflict through criminal frameworks can lead to inappropriate responses which detract from efforts to meet humanitarian needs" (Bird and Reinato, 2019: 1).
Regarding less represented types, a few organisations are specialised in migrations but are overall poorly represented.Finally, worker and business organisations are a minority of represented interests within the HT regime complex (2 out of 15).This might be a problem because they are directly concerned by human trafficking, especially with regards to forced labour.Trafficking is most of all an Robert Schuman Centre for Advanced Studies Working Papers economic issue and the marginalisation of the ILO as a forum of the regime complex is coupled by the absence of business interests within the regime complex.

Conclusion and recommendations
In March 2017 the UN Secretary General announced at a UN Security Council meeting that "at a time of division in so many areas, this [the fight against human trafficking] should be an issue that can unite us" (UN, 2017).Indeed, the challenge of HT is still ongoing and requires an efficient coordination between the different elements of the HT regime complex.This research aims to better characterise this complex, identifying its shape by looking at non-state actors' participation to its negotiation meetings.The analysis reveals some strengths of the regime complex, as well as a number of weaknesses.By pointing at potential improvements to correct these weaknesses, the analysis suggests ways to further organise international HT global governance.These policy recommendations are meant to be endorsed by governments, with the EU as a particular target.
Regarding strengths, the analysis of CS participation within the HT complex confirms the good inclusion of CS actors in HT global governance.This inclusion is positive for participatory dynamics and creates a rather dense and solid regime complex.Another trend is the progressive shift towards the increasing inclusion of human rights' concerns in HT governance, confirming former assumptions that "one can easily assume that the contrast to trafficking in persons changed perspective throughout the years, from a mere repressive strategy of the Palermo Protocol in order to fight the crime itself to a more human-rights sensitive approach of soft law international instruments, like the Recommended Principle and Guidelines on Human Rights and Human Trafficking and recent regional legal instruments" (Gaspari, 2019: 63).In particular, the analysis reveals the central role of the OHCHR and of the UNHCR in the HT regime complex.This evolution clearly marks a shift towards the recognition of victims' conditions and rights.
Regarding weaknesses, both the ILO and the IOM appear disconnected from the complex and could be further engaged in HT governance by suggesting followers of these fora to follow other international negotiation processes.In particular, the ILO is the only forum that engages business actors, which are necessary players to stop the HT demand side, especially in relation to forced labour.Economic actors could therefore be mobilised and invited to participate more consistently to other fora than the ILO, to better perform forum linking.Of course, business actors can have reluctances and involving them could be seen as a potential risk to water down HT governance.However, business actors could also be willing, for corporate responsibility reasons, to adopt and implement a code of business conduct to address labour trafficking, and prevent human trafficking in supply chains by publishing information, including supplier or factory lists, for consumer awareness.They could adopt fair trade certification or other ethical business practices that guarantee that no child or forced labour was used in the production of goods.Consumers could therefore be better informed of their responsibility with regards to HT and be interested by these labels: "it is argued that the narrow construction of the 'trafficking offender' establishes a very limited scope of blame for the crime of trafficking, absolving others of responsibility.The target audiences for the campaigns are positioned as a solution to, not potential cause of, the crime of trafficking" (O'Brien, 2016: 208).
Another weakness is the unbalanced characteristics of key organisations engaged in the HT regime complex.The research confirms the existence of a negotiation burden in regime complexes that means that only organisations with a certain level of resources are able to participate consistently over time.Therefore, organisations from the global south tend to be underrepresented in regime complexes and involving them would be a solution to tackle the HT problem more comprehensively, involving also the analysis of the very causes of HT.Indeed, so far, "a large part of the efforts are concentrated towards rescuing victims and prosecuting traffickers" (Usman, 2014: 283), while preventive measures should also be included: "why are people trafficked?What pushes them to fall into the traffickers' trap?" (Gaspari, 2019: 49).
Finally, this paper enables to identify a new research agenda on HT.First, it would be useful to conduct qualitative research, by interviewing each of the identified key actors of the HT regime complex, to better grasp their practices and views on HT global governance.Second, it would be interesting to assess how regional processes are connected to the global processes evidenced here.HT is also a regional issue and several regional initiatives have been developed.On the European continent, the Council of Europe and the European Union are important players.Regarding the former, "Europe is both an area of destination and origin of trafficking in human beings and for this reason it has developed a well-structured system, starting from the Council of Europe (CoE)which is not a European Union institution and numbers forty-seven member States.On 16th May 2005 in Warsaw (Poland), the Council of Europe Convention on Action against Trafficking in Human Beings was opened to signatures" (Gaspari, 2019: 56).Regarding the latter, it would be interesting to look at the participation of CSOs within internal European policy processes (see Box 3), to see how well they are connected to the global level, and to build more bridges between these different policy levels.

Box 3 .
European instruments to fight against human trafficking 2000 -Charter of Fundamental Rights of the European Union -Article 5. 2002 -Council Framework Decision on combating trafficking in human beings.2004 -Council Directive regarding the residency permit issued to third-country nationals who are victims of trafficking in human beings or who have been subject of an action to facilitate illegal immigration, who cooperate with the competent authorities (2004/81/EC).2011 -Directive of the European Parliament and of the Council on preventing and combating trafficking in human beings and protecting its victims (2011/36/EU).2012 -Directive of the European Parliament and of the Council establishing minimum standards on the rights, support and protection of victims of crime (2012/29/EU).2012-2016 -EU strategy towards the eradication of trafficking in human beings to supplement and complete the European Union framework.2017 -EU Communication 'Reporting on the follow-up to the EU Strategy towards eradication of trafficking in human beings and identifying further concrete actions'.Source: European Union websites.

4
CSOs were first categorised by the United Nations in 1992 for Agenda 21 as encompassing 9 categories: women; children & youth; indigenous peoples; non-governmental organisations; local authorities; workers & trade unions; business & industry; scientific and technological community; farmers.

Figure 1 .
Figure 1.Civil society links between the different institutional elements of regime complexes

Table 1 -Civil society participation in the HT and GR regime complexes (2000-2019)
Table 2 below presents the top 15 key CS actors identified in the HT regime complex according to the methodology used for my third objective.Robert Schuman Centre for Advanced Studies Working Papers

Annex 1. Meetings included in the HT database (2001-2019)
Robert Schuman Centre for Advanced Studies Working Papers