Skip to main content

Addressing demand in anti-trafficking efforts and policies (DemandAT)

Final Report Summary - DEMANDAT (Addressing demand in anti-trafficking efforts and policies (DemandAT))

Executive Summary:
Should consumers be held responsible for exploitative practices that may be involved in production of consumer goods or services they consume? If yes, in what ways? How can consumers be encouraged to change their consumption behaviour? Can campaigns encouraging the general public not to donate to or buy token goods from children reduce trafficking of children into begging? Do campaigns encouraging the public to report suspicions of trafficking to the police or specialised NGOs result in tangible results? These types of questions are discussed in the anti-trafficking field as referring to “demand” or “demand reduction”. In fact, “to discourage the demand that fosters all forms of exploitation of persons” is a legal obligation of State Parties to the UN Anti-Trafficking Protocol (also referred to as the Palermo Protocol). Similarly, article 18 of the EU Anti-Trafficking Directive states that “Member States shall take appropriate measures, such as education and training, to discourage and reduce the demand that fosters all forms of exploitation related to trafficking in human beings”. The EU Anti-Trafficking Directive does not offer clear instructions with regard to the meaning of demand and how exactly demand can be reduced or discouraged. What kind of measures should discourage demand and how they are intended to work remains unclear. In the absence of further guidance, states took various approaches. In practice, one finds a wide range of measures described as “addressing demand” by actors in the field, although there is usually no clear explanation regarding what kind of demand is being referred to and how exactly reducing demand leads to reducing “all forms of exploitation”.

Focusing on demand in relation to trafficking has been recognised as having the potential to shift the focus of anti-trafficking policies towards those who profit from trafficking and therefore might have some responsibility for the occurrence of trafficking. In order to provide clarity on the above-mentioned issues, the project “Demand-side measures against trafficking” (DemandAT) was launched in 2014 under the EU’s 7th framework programme. The project, involving a multidisciplinary team of researchers and practitioners from nine institutions in seven European countries, aimed to examine demand-side approaches in the context of trafficking in human beings, to clarify the meaning and relevance of demand in different policy areas as well as to contribute to a better formulation of ant-trafficking policies addressing demand. The study combined theoretical analyses of the concept of demand and different types of policy interventions with a mapping of policy debates and implemented policies across a large range of countries, as well as different fields in which trafficking occurs with in-depth studies of particular fields (domestic work, sex work) and approaches (supply chains, law enforcement, campaigns). Although measures in selected fields and policy approaches are intertwined – there are anti-trafficking campaigns in the area of addressing trafficking in the global supply chains, and law enforcement is relevant in all the fields selected – having separate case studies on areas in which trafficking occurs and on approaches to trafficking presented the opportunity to ask filed specific questions and critically investigate approaches employed.

DemandAT is a research project focusing on measures aimed at addressing demand in the context of trafficking. Discussions on whether addressing the demand in particular markets (in which production, distribution or consumption practices may involve exploitation) leads to reduction of trafficking in human beings were undertaken in particular case studies, but no single case study was aimed primarily at this question. The combined results of the different strands of research undertaken in the project have a number of important implications for the relevance, scope and usefulness in addressing different types of trafficking on the one hand, and for policy formulation and policy evaluation in the field of anti-trafficking, on the other.
Project Context and Objectives:
Should consumers be held responsible for exploitative practices that may be involved in production of consumer goods or services they consume? If yes, in what ways? How can consumers be encouraged to change their consumption behaviour? And what is their actual leverage to shape conditions under which services are provided or goods are produced? What about firms and large institutional procurers purchasing goods and services through supply chains? What is their leverage and in what ways can they be encouraged, or indeed be obliged, to address trafficking in their supply chains? What about other fields in which trafficking occurs? Can campaigns encouraging the general public not to donate to or buy token goods from children reduce trafficking of children into begging? Do campaigns encouraging the public to report suspicions of trafficking to the police or specialised NGOs result in tangible results? These types of questions are discussed in the anti-trafficking field as referring to “demand” or “demand reduction”. In fact, “to discourage the demand that fosters all forms of exploitation of persons” is a legal obligation of State Parties to the UN Anti-Trafficking Protocol (also referred to as the Palermo Protocol). The Protocol does not offer additional explanation regarding the meaning of demand, with the exception of an enumeration of types of measures to address demand: “legislative or other measures, such as educational, social or cultural measures, including through bilateral and multilateral measures”.

Article 18 of the EU Anti-Trafficking Directive states, citing almost verbatim from the Palermo Protocol, that “Member States shall take appropriate measures, such as education and training, to discourage and reduce the demand that fosters all forms of exploitation related to trafficking in human beings”. As the UN Anti-Trafficking Protocol the EU Anti-Trafficking Directive does not offer clear instructions with regard to the meaning of demand and how exactly demand can be reduced or discouraged. Thus, what kind of measures should discourage demand and how they are intended to work remains unclear. In the absence of further guidance, states took various approaches. In the field of anti-trafficking there are various measures labelled as “addressing demand”, although there is no clear explanation regarding what kind of demand is being referred to and how exactly reducing demand leads to reducing “all forms of exploitation”.

Focusing on demand in relation to trafficking has been recognised as having the potential to shift the focus of anti-trafficking policies towards those who profit from trafficking and therefore might have some responsibility for the occurrence of trafficking.

In order to provide clarity on the above-mentioned issues, the project “Demand-side measures against trafficking” (DemandAT) was launched in 2014 under the EU’s 7th framework programme. The project, involving a multidisciplinary team of researchers and practitioners from seven European countries, aimed to examine demand-side approaches in the context of trafficking in human beings, to clarify the meaning and relevance of demand in different policy areas as well as to contribute to a better formulation of ant-trafficking policies addressing demand. The study combined theoretical analyses of the concept of demand (Vogel, 2017a) and different types of policy interventions (Boswell and Kyambi, 2016) with a mapping of policy debates (Cyrus and Vogel, 2015; Cyrus 2015) and implemented policies across a large range of countries (Rogoz et al., 2016), as well as different fields in which trafficking occurs with in-depth studies of particular fields (domestic work, sex work) and approaches (supply chains, law enforcement, campaigns). Although measures in selected fields and policy approaches are intertwined – there are anti-trafficking campaigns in the area of addressing trafficking in the global supply chains, and law enforcement is relevant in all the fields selected – having separate case studies on areas in which trafficking occurs and on approaches to trafficking presented the opportunity to ask filed specific questions and critically investigate approaches employed. DemandAT is a research project focusing on measures aimed at addressing demand in the context of trafficking. Discussions on whether addressing the demand in particular markets (in which production, distribution or consumption practices may involve exploitation) leads to reduction of trafficking in human beings were undertaken in particular case studies, but no single case study was aimed primarily at this question. The combined results of the different strands of research undertaken in the project have a number of important implications for the relevance, scope and usefulness in addressing different types of trafficking on the one hand, and for policy formulation and policy evaluation in the field of anti-trafficking, on the other.

In practice, the research was conducted in three interconnected phases spread over three and a half years of project implementation. The first phase of the research developed a theoretical approach for demand in the context of trafficking and analysed empirical literature in different disciplines, fields and countries. Results of this phase, particularly on the theoretical approach, were used as a basis for the in depth case studies conducted during the second phase of the project. The main idea was to look at demand in the context of various forms of trafficking and from different perspectives in order to address different forms of exploitation (which may occur in different markets) and different ways in which demand in the context of trafficking may be addressed. As mentioned above, the fields selected were domestic work and prostitution, while the approaches analysed were supply chains, law enforcement and anti-trafficking campaigns.

One of the main contributions of this project to the field of anti-trafficking comprises theoretical and conceptual clarifications with regard to demand in the context of trafficking in human beings, and subsequently a comprehensive analysis and discussion of the role of demand in anti-trafficking. Conceptually, the study suggests limiting the meaning of the concept of demand to its core economic meaning, i.e. the willingness and ability to buy a good or a service (Vogel, 2017a). This approach requires that reference to demand is made with the correlative supply and/or with specific reference to a market – what good or service is purchased and/or on what market? Confining ‘demand’ to purchasing plans on markets does not mean that market interactions are most important for understanding trafficking in human beings. A trafficked person is transferred into and kept in an exploitative relation. Thus, a non-market relation is at the core of understanding trafficking in human beings. By limiting the meaning of demand the focus is then on how market transactions (such as the purchase of specific goods by consumers) are linked to situations of trafficking, and as a corollary, how demand-side measures targeting participants in the market transaction (e.g. through campaigns promoting consumers to buy goods certified to have been produced under fair labour conditions) or the rules under which market exchanges occur (e.g. by criminalising the particular types of exchanges) impact on trafficking. Consequently, phrases such as “demand for trafficking in human beings” do not make sense and should be avoided, since trafficking is not a good or a service.

Referring to demand and supply emphasizes the contracting situation in which an exchange takes place for a price. The analytical value of using demand-and-supply-terminology is highest when it is applied to contracting situations. However, many market contracts lead to a longer-term relation between two parties, and what happens in this relation is then relevant for analysing exploitative situations. In doing so, it is important to consider that exchanges take place in different markets and in different contexts, implying that there is no fixed “demand-side” and that who is on the demand-side and what measures can be taken differs between a consumption context (where consumers are on the demand side), a production context (where employers demand labour directly from workers or from labour providers) and a recruitment context (where individuals demand various brokerage services, such as smuggling or labour matching services and related services, such as credit). Indeed, the need for differentiation is one of the most important overarching results of the different studies undertaken in the context of the DemandAT project.

All of the above entails that the overall research design of DemandAT considered the complexity of the concept of trafficking, as well as the various exploitative types of situations included in the legal definition of trafficking. However, since there are no trafficking situations that do not involve exploitation (Cyrus and Vogel, 2017), one logical way of action for measures aimed to reduce trafficking is to primarily aim at reducing exploitation. Moreover, as Cyrus (2015) explains, the legal definition of trafficking in human beings is both narrow and wide. It is “a very narrow definition, as the three constitutive elements (act, means and purpose) have to occur coincidentally in order to constitute the offence of trafficking in the legal sense (Gallagher 2010: 80). At the same time, the definition is wide because the elements of act, means and purpose each cover a wide array of situations” (Cyrus, 2015: 13). Consequently, trafficking in human beings makes for a poor analytical concept, “as the legal construction refers to a wide array of phenomena, these cannot be observed simultaneously by social scientists” (Cyrus, 2015: 16). For these reasons, but also for specific reasons which are clarified in the sections dedicated to particular sub-studies, the starting point into an analysis of demand in the context of various forms of trafficking was to analyse demand for a good or a service in the context of specific exploitative situations.

In parallel, research into different policy areas in which demand for illegal or harmful goods and services is addressed contributed to a better understanding “of what demand-side measures are and the effects they can attain” (Boswell and Kyambi, 2016: 17). This particular sub-study resulted into a typology of regulatory tools for steering societal behaviour (Ibid.), typology employed later on in other sub-studies on governmental approaches to demand (Rogoz et al, 2016) as well as the role of law enforcement in addressing demand in the context of trafficking (Haxhi, 2017; Bachinger et al., 2017). Results of the study on government responses to demand revealed that like other criminal activities, states address the crime of trafficking in human beings predominantly with command-and control measures involving the detection and sanctioning of socially undesirable forms of demand. However, states can also redesign the legal infrastructure of markets to make undesirable forms of demand less likely. They can use market-based incentives such as taxes and subsidies, or use measures promoting specific values or behaviours through peer pressure, promoting, for example, relevant initiatives by businesses, NGOs or other actors either by incentives (subsidies, cooperation) or through legal obligations (e.g. introduction of reporting obligations on measures taken against risk of THB in companies’ supply chains). Yet measures may also be used for symbolic reasons. Both, research on government responses to demand in the context of trafficking and research on demand arguments in debates in different fields of trafficking revealed that demand is often used in its economic understanding, but not consistently. Analysis of demand arguments found that demand is increasingly employed in debates to explain the framework of trafficking as an activity and as well as to justify particular approaches. These approaches are embedded in political frames that link measures to value-loaded problem definitions.
Project Results:
2.1 Demand-side measures in selected policy fields
In this section, demand-side measures in three selected policy fields are addressed – prostitution, domestic work, and global supply chains.
2.1.1 Prostitution

Sex work is one of the areas DemandAT looked into for several reasons. First, from a historical perspective, both notions, “trafficking in human beings” and “demand” are strongly linked with prostitution policies.

“The history of concepts performed in [the framework of this project] revealed that the term [...] trafficking is introduced in past and present debates with a confusing diversity of meanings, referring among other issues to the kidnapping of girls and putting them into prostitution, fraudulent procurement of unsuspecting women for prostitution abroad, procurement of consenting women for prostitution, abetting of irregular border crossing or fraudulent abetting of irregular migration with the purpose to exploit migrants after arrival. The term demand is introduced in past and present debates in order to refer among other issues to biological drive of males, to a demand generated by a system of state regulation of prostitution, to a demand of brothel owners and pimps, to a demand of male clients to purchase commercial sexual services.” (Cyrus, 2015: 83).

Second, in current debates, the reference to demand in the context of trafficking is discussed to a large extent in relation to the sex work sector. In fact, most policy makers we spoke to in the exploratory sub-study on government responses to demand referred to addressing the demand for sexual services (Rogoz et al., 2016) as one measure they either support or they are aware of. In particular, they referred to the criminalisation of clients of sex workers. The ongoing debates on this issue – particularly on whether the criminalisation of clients of sex workers leads to a decrease in the instances of trafficking in human beings in this sector – “have not led to any form of unified policy stance across” countries (neither among EU Member States nor among State Parties to the Palermo Protocol) (Östergren, 2017b).

The study on the sex work sector, conducted in the framework of DemandAT, aimed to understand how demand for sexual services is addressed in different types of prostitution policies and to compare their potential impacts, particularly with regard to exploitation and trafficking in human beings in the sex work sector. The work package on the sex work sector is comprised of three studies. The first is a comparative research on ways in which demand-side measures operate in three countries with different prostitution policies – Sweden, Germany and New Zealand. The second study is a qualitative research with Swedish men who purchased sexual services. The third study is an analysis of prostitution policy typologies, “proposing a new general tripartite policy typology using the labels repressive, restrictive and integrative” (See Annex 1) (Östergren, 2017b). The three countries were selected as they clearly differed in their approaches to what is legally possible or criminalised with regard to sexual services. Each country study included desk review and semi-structured interviews with relevant stakeholders.

The available evidence is insufficient to draw definitive conclusions with regard to the impacts of policies on the occurrence of exploitation or trafficking in the sex work sector or on the clients’ behaviour towards exploitation in the sector. However, the study is able to draw a number of conclusions as to how different policies address demand for sexual services and how they affect “the relationship between authorities and participants in the sex work sector” (Östergren, 2017b). Research in this sector faced particular challenges. Prostitution policy, addressing the sex industry, is a ‘policy resistant’ sector (Wagenaar et al. 2013) which is difficult to control and manage due to the high mobility of workers, anonymity and stigma which make actors engaging in the sex industry to resist policy measures. It deals with morality politics where opposing values and opinions are at the forefront. In addition, ideology is involved in policy formulation which leads to conceptual confusion, lack of solid policy theory and reliable data.

One major contribution of this particular study is the alternative way suggested to regard prostitution policies and their implications on the field, on the actors involved in the sex work sector as well as on the relationships between relevant stakeholders. Rather than raising the question on which demand-side policy or measure is the “best” to address trafficking in the sex work sector, the research undertaken in the framework of DemandAT suggests to look into “which measures are possible in different policy settings” (Östergren, 2017b).

Similar to policies in other fields, assessing prostitution policies requires focus on total impact . At least four variables should be considered – intention of the overall policy, instruments used to achieve declared goals, implementation and impact (on sector or on individuals involved in that sector). Prostitution policies have been largely researched and there are many attempts to classify them – criminalisation, legalisation/regularization and decriminalization etc. However, while conducting the research, problems with existing classifications emerged. For instance, ‘legalisation and decriminalisation could be said to be two ways of speaking about the same final regulatory situation, (where an act that was previously considered a criminal offence no longer is)’.

For the above-mentioned reasons, an alternative conceptualisation of prostitution policies was offered. “The Swedish policy is repressive: commercial sex is considered to be a negative social phenomenon that should be eliminated primarily with the help of criminal law. Purchasing sex is illegal, as well as all third-party activities. The dominant policy rationale is based on a kind of feminist ideology that equates all prostitution with male violence against women. In contrast to Sweden, New Zealand has a multifaceted understanding of commercial sex and regulates the sector by standard labour and administrative law, but adds specific legislation and codes of conduct for authorities and business owners in order to protect sex workers from violence and exploitation. The policy is informed by a human rights ideology. Finally, Germany combines both restrictive and integrative policy elements. The integrative policy can be observed at the level of national law, since it is not a criminal offence to sell or purchase sexual services, and the stated aim is to move the sale of sexual services from the domain of 'immorality' to that of 'labour'. At state and city levels, however, there is a range of different regulations and practices, ranging from restrictive to integrative” (Östergren, 2017b).

Although policies in all countries suffer from inconsistencies, some observations regarding policy effects in different regimes can be made. In Sweden demand is being addressed though criminal law and by public campaigns aimed at changing clients’ behaviour. In Germany, cooperation between clients, NGOs and public authorities is encouraged, demand being addressed though codes of conduct and awareness-raising campaigns. In New Zealand empowering sex workers is done though collaboration between sex workers organisations, law enforcement and other authorities with the aim to prevent crimes against sex workers. In addition, the codes of conduct developed focus, among others, on measures to protect sex workers and prevent violence and exploitation.

Another conclusion reached is that repressive policies tend to be an obstacle for detecting, reporting and preventing crimes. As it comes out from the interviews with Swedish clients, in a context in which most aspects of purchasing sex (although not all) are criminalised, clients tend not to report suspicion of crimes in order to avoid being prosecuted themselves . In addition, reporting a crime has to do with the individual ability, responsibility and ultimately willingness to report a potential crime when in danger of being prosecuted. With regard to the relationship between sex workers and public authorities, this is influenced by the different policy regimes. The comparative analysis undertaking, corroborated with results of other studies, showed that both sex workers and clients in Sweden tend to express less trust in law enforcement authorities or NGOs in comparison, for instance, with those involved in selling and buying sexual services in Germany and New Zealand. This has implications for the ways in which and the extent to which violence is reported and combated.

To sum up, the research on prostitution policies (comprising of three sub-studies on policies addressing demand, clients’ study and a typology of prostitution policies) suggested to shift the focus from what is the best way of action into understanding what impacts different measures in different policy regimes can have on both, the field and the actors involved in the sex work sector. While the study formulates a number of recommendations as they come out of its findings, it does not recommend a particular policy types (repressive, restrictive or integrative) over another. Rather, the study suggests possible measures that are expected in different policy settings and which “pertain to reducing violence, exploitation and THB” (Östergren, 2017b). For a repressive policy regime, compatible measures include amnesty practices for witnesses and whistleblowers so that they may report crimes like rape and trafficking in human beings without fear of being prosecuted for buying sexual services. For integrative and repressive regimes, requirements for workplace security and worker information can contribute to a safer work environment. Firewalls preventing that victims and witnesses of crime may be prosecuted for immigration offences could help. Empowering sex workers to address the structural constrains they face when it comes to safety will reduce their vulnerability to exploitation, including trafficking. It is expected that an inclusive policymaking process in this field – which recognised and engages with sex workers organisations as relevant stakeholders – together with the consideration of multiple measures on several levels – not only demand-reduction – aimed at reducing workers’ vulnerability, will increase workers safety and their access to rights.

This particular conclusion – on exploring a multitude of measures to address workers’ vulnerability has been reached by the research conducted in the field of domestic work. There too, workers face structural constraints in accessing their rights. This is the topic of the following section.

2.2.2 Domestic Work

Trafficking in the context of domestic work makes a particular case which deserves special attention, for a number of reasons. First, it challenges, to a certain extent, the organised crime perspective of trafficking in human beings. Trafficking in domestic work occurs outside organised crime, bringing “into the discussion the family and private households of different socio-economic background” (Ricard-Guay, 2016). Second, “domestic work occupies a unique position at the intersection of the public and private realms, the labour market and the family” (Ibid.) which poses multiple challenges when it comes to finding sustainable solutions in terms of preventing exploitation and abuse. Domestic services can be supplied by family members without payment, by workers who serve several households as self-employed service providers (Ambrosini, 2013) and by labour relations of different kinds in which private households are consumers and employers at the same time. Finally, the demand side in relation to trafficking for exploitation in domestic work is worth paying attention to, especially since demand for domestic workers has grown steadily in Europe in the last three decades. Literature on domestic work considers multiple causes for this increase in demand for domestic workers: changes in roles of family members who traditionally were involved in domestic work (e.g. increased number of women on the market labour), aging populations, as well as a migrant work force available for this type of employment. Ambrosini (2013) speaks of an “invisible welfare” supported by the work of domestic workers who then fill in the gap of a formal welfare regime.

Labour exploitation is one of the main areas of concern when it comes to addressing trafficking in human beings in the domestic work sector. Although it seems that, at least in the EU, more cases are being investigated (EUROPOL, 2016: 23) which leads to a higher number of persons identified as having been trafficked for labour exploitation (EUROPOL, 2016; EUROSTAT, 2013, 2014, 2015), less seems to be done to address THB for labour exploitation than, for instance, to address trafficking in the sex work sector (Ricard-Guay, 2016). THB for labour exploitation is unevenly addressed by different EU countries and exploitation in domestic work adds to the complexity of the issue through the nature of the work (often outside of regular employment frameworks) and the place of employment (private households where labour inspectorates and equivalent agencies have little to no access).

The aim of this study – on addressing demand in the context of trafficking in the domestic work sector – was threefold – to investigate types of situations which may involve exploitation, including trafficking; to investigate the motivations and factors shaping demand for domestic work; and to explore possible gaps in legislation and policies intended to prevent exploitative situations in domestic work. Research was conducted in seven European countries: Belgium, France, Greece, Cyprus, Italy, the Netherlands and the UK. Each of these country studies comprised extensive desk research, interviews with key stakeholders working in different sectors relevant for the domestic work sector. While domestic work is recognised as a field prone to exploitation, this research found that little has been studied in this area – particularly on trafficking in human beings in the field of domestic work (Ricard-Guay, 2016).

Research found four specific situations in which trafficking has been reported:
1) domestic workers employed directly or indirectly (involving intermediaries such as recruiting agencies), under declared or undeclared contract by private households;
2) domestic workers employed by diplomatic personnel;
3) instances in which au-pair workers were in fact full-time care workers;
4) family-based arrangements (arranged/forced marriage, child fostering), without an employment contract.

In addition, most trafficking situations in domestic work involve the following three features:
1) Most trafficking situations in domestic work involve migrant workers, although case law involving nationals was reported in two of the researched countries.
2) In addition, almost all trafficking cases involve live-in domestic workers.
3) Other important features of these cases involve usage of deception rather than coercion, in many situations the person exploited and the perpetrator have known each other before the exploitative situation; this is the case, particularly in situations involving family arrangements.

When looking at the motivations and factors influencing demand for domestic work, research found several types of such factors. First, there are the economic factors – households don’t generate additional profit from domestic workers, but savings from paying less or not paying at all are being achieved. Second, attitudes towards domestic work maintain the devaluated perception of work in this sector which faces also discrimination based on gender (mainly women being involved in domestic work) and well as based on ethnicity (particular ethnic groups perceived to be suited for domestic work) (Ricard-Guay, 2016b). Third, similar to findings in other sub-studies in DemandAT, policies and regulations play a particular role in setting the conditions of demand for domestic work (Ibid.). Specific measures, such as tied-visa for migrant workers or criminalisation of irregular status of these workers, “can generate or exacerbate precariousness” (Ibid.). “The voucher system in Belgium and the chèque à l’emploi in France are seen as contributing to decrease undeclared work in the domestic work sector. Both systems set positive incentives for employers to register the employment of a domestic worker: an easy-to-use hiring/registration process, tax deductions, and government contributions. However the impact of these on preventing trafficking has to be considered against the backdrop that most of trafficking cases continue to occur outside formal arrangements. Furthermore, most situations of trafficking take place in live-in arrangements, while these measures are confined to part-time live-out services.” (Ibid.).

The study formulated a series of recommendations which, if taken into account at a policymaking level, are expected to improve working conditions and reduce domestic workers’ vulnerability to exploitation. Considering the under-declared exploitative situations in domestic work and that domestic work takes place in private households, far from the public eye, it is expected that enhancing detection of exploitation in the domestic work sector would break what was called “cycle of invisibility”. This might need training sessions for law enforcement authorities, but also support for and involvement of trade unions and other organisations working with domestic workers who can report possible situations of exploitation, including trafficking. Enhancing detection requires that enforcing the law and monitoring compliance is also enhanced. Research in this particular sector found that although labour inspectorates have restricted access to private households – even when these households are places of employment – rethinking the role of labour inspectorates in monitoring working activities inside private homes is expected to contribute to detection. “Improving measures can draw on innovative practices that provide alternatives to house visit (e.g. in Ireland, a meeting between the employer and employee is organised in the event that the employer refuses an on-site visit)” (Ricard-Guay, 2016b). In addition, to increase reporting of exploitative situations and protect workers’ rights, enforcing the employment legislation should not be linked with the obligation to report irregular migrants to migration authorities – such a firewall was also suggested for the sex work sector. At the same time, stronger regulations and strict enforcement of these regulations onto recruitment agencies (including penalties for non-compliance) is expected to improve employment conditions in the sector. Since particular attitudes towards domestic workers are found to contribute to the unrecognised value and precarious position of domestic workers, awareness-raising activities regarding the working conditions in this field is thought to contribute to changes in the general public attitudes towards domestic workers. Similar to findings in other sub-studies in DemandAT, empowering workers in the domestic work sector is expected to improve the overall employment conditions and working environments. This requires changes in labour regulations through the creation of incentives to comply for both, employer and employee. It is expected that reducing undeclared work and enhancing workers’ leverage (by eliminating schemes that maintain dependency such as tied-visa programmes) will lead to better working conditions. Similarly, specific regulations and support for domestic workers involved in 24/7care are expected to reduce the abuse and ultimately the exploitative situations, particularly with regard to live-ins. Researchers involved in the study go even further, suggesting that live-in arrangements should be removed and “facilitate change of employer [for domestic workers], as these conditions [live-in] exacerbate the risk of exploitation for domestic” workers (Ricard-Guay, 2016b).

From a demand-supply analysis, in a domestic work situation the consumer, employer and trafficker are often identical or are living in the same household, without complex demand-supply chains. The exact opposite can be said of the globalised production of goods. In global supply chains, production is both fragmented and dispersed, with companies involving subsidiaries in various countries, working under different legal frameworks.

2.2.3 Global supply chain initiatives

The way we live in the rich world, which includes the affluent classes in less developed countries, imposes terrible costs on many of those who grow our food, make our clothes, mine our metals, cut our timber and otherwise sustain our generally comfortable existence. One day it is textile workers dead under a collapsed factory in Bangladesh. The next it is a wave of suicides in Chinese computer factories. The next the revelation that the trainers so many of us wear are put together by underage girls on miserably low wages. (The Guardian, 2014)

This excerpt is part of an editorial response to media reports published by the Guardian documenting that prawns are produced through slavery-like conditions. Literature on global production of goods suggests that global supply chains frequently present a challenge for the promotion of decent work (ILO, 2016) and that exploitation within supply chains has become a major concern for both public and private sectors. Between labour markers (subscribing to production contexts) and markets for final product use (consumption context) there are markets for sale and purchase of intermediate goods. Relevant literature underlines that, in these contexts, the buyer-supplier relationship is not depersonalised, as depicted by market relationships. The various relationships between suppliers and buyers and the power relations intrinsic to these relations “mean that a market framework would fail to capture the central dynamics of supply chains” (McGrath and Mieres, 2017). Literature addressing the issues of coordination and control in these contexts refer to supply chains rather as global commodity chains, global value chains or global production networks (GPNs) (Coe, Dicken and Hess 2008; Bair 2009 in McGrath and Mieres, 2017).
When it comes to the relevance of these issues, the EU has a responsibility to address exploitation within supply chains, as products and services produced or offered through exploitative practices outside of the EU might benefit those living in the EU – consumers, companies, states bodies (through public procurement). Responsibility here is not used in the sense of liability only, but it goes beyond that into what comprises ethical responsibility. In addition, interventions in supply chains represent an opportunity for relevant stakeholders to contribute to and shape the conditions under which labour is being performed outside of the EU and thus outside the EU judicial and law enforcement area.

In order to capture the array of initiatives addressing exploitative practices, so not only trafficking, but also forced labour, contemporary/new ‘slavery’ etc., the study on global supply chains used the acronym TFLS to refer to all phenomena that are indicated with the terms “human trafficking, forced labour and/or (new, contemporary or modern) ‘slavery’” (McGrath and Mieres, 2017). This particular sub-study investigated the landscape of initiatives that address TFLS in and through global supply chains. The aim of the research was to contribute to a better understanding of what is likely to make an initiative successful or unsuccessful in bringing about change (McGrath and Mieres, 2017b). The main questions addressed were: How do the logic(s) of the initiatives relate to the logic(s) of industries and their geographical contexts? What are the mechanisms by which initiatives work and what does this imply about responsibility to address TFLS? What can we learn about initiatives when analysed ‘on the ground’ through case-studies?

A first step of the research was to map the landscape of initiatives at the TFLS-supply nexus. A dataset, comprising 97 selected initiatives, was used for developing a typology of such initiatives. The elements used to then categorise the initiatives based on the remit of the initiatives (TFLS or a wider focus), scale (global, regional, national etc.), industrial scope (multi-sector or industry specific), initiating and implementing actors, main mechanisms employed to address TFLS. In addition, three sector-specific case studies were conducted: electronics in Malaysia, construction in Qatar and agriculture in Florida (US). An International Advisory Board for this particular work package supported the research.

What was found was that many initiatives incorporate addressing trafficking into Corporate Social Responsibility (CSR) – through written guidance, training sessions, standard setting, inserting additional language to existing policy etc. This is considered the outcome of the intersection of two trends: those involved in CSR recognise the importance of addressing TFLS and those involved in policy in areas of TFLS recognise the relevance of supply chains.

The analysis of the initiatives included in the dataset revealed that activities around TFLS in supply chains refer to supply chain monitoring and employer guidance, often in combination (McGrath and Mieres, 2017b). The number of initiatives in the area is growing. This could be an indication that “lead firms” are expected to share the responsibility for the issue, together with governments and other involved stakeholders. The growing activities around labour market intermediaries – requiring suppliers to employ workers themselves rather than through recruitment agencies, to prohibit the payment of recruitment fees by workers and reimburse fees to workers who already paid such fees – is also an indication that lead firms recognise workers’ contribution and that changes in working conditions benefit both, the employees and the employers. The issue of how exactly lead firms will contribute to recruitment costs remains an open one.

Research found that it contributes significantly to the success of an initiative if it involves meaningful workers participation as a key mechanism. This is the case of two initiatives implemented by trade unions (“From Catcher to Counter Program” targeting fisheries and “Fair Games – Fair Play” focused on construction) and one implemented by an alliance between an NGO and a trade union (the “Clean Clothes Campaign”). The Fair Food Program (FFP) is also such an initiative in which workers have been involved in developing an industry code of conduct which is informed from the ground on issues worth to include. The FFP is a partnership between farmers, farmworkers and retail food companies that “ensures humane wages and working conditions for the workers who pick fruits and vegetables on participating farms” The code of conduct developed within FFP is the bases for “negotiated bilateral agreements with national and international retail brands” (Brudney, 2016). Workers are also involved in implementation and monitoring. For those non-complying with the code of conduct there are costly consequences: buyers who signed the code are required not to buy from growers who do not comply with the code. Workers and farm labour contractors are employed directly by growers, in order to avoid that intermediaries exploit the a weak position of workers

Since TFLS practices in supply chains do not take place in a vacuum, it is important to acknowledge the relevance of migration regimes and labour standards as frameworks in which all initiatives operate. In environments where workers are not free to organise, nor to ensure that their safety is protected, it is more likely that TFLS practices will occur (McGrath and Mieres, 2017b).
To sum up, as a result of this research, four main policy implications/principles for intervention in the field can be formulated.

1) Responsibility. The first principle is what was referred to as going beyond compliance. The study suggests that lead firms need to move from their concerns (that they now seem to have) about liability to proactively taking responsibility for transforming labour relations in their supply chains. For example, this may involve sharing the costs for ensuring decent labour conditions in their supply chains; or at the level of policies/laws setting/ detailing responsibilities for subcontractors’ actions across the world. Further proposals have been made to require companies to publicly disclose which suppliers they use.
2) Enforceability. Relevant are the types of enforcement mechanisms build into the initiative, the sanctions for noncompliance and forms of legal access to remedies against companies in case of failed targets or regulations. In spite of available examples of and models of enforceable initiatives, many of them lack any form of enforceability.
3) Worker participation. Unions are taking up this issue at international level through international framework agreements and Memoranda of Understanding (MOUs). The research suggests that also companies, NGOs, governments and others should start from the question of worker participation, considering the involvement of trade unions or other workers organisations on the ground. And if not, initiatives should consider whether there are workers organisations elsewhere that might be working as partners. In a more challenging environments, initiatives should first consider what might be done to create the conditions for workers to organise and be heard.
4) Public regulations. Extending and enforcing legislation which protects workers rights and labour standards remains important for combating practices associated with trafficking, slavery and forced labour. Private initiatives complement and reinforce public regulations, but they cannot act as substitutes for public law. Initiatives to combat trafficking, forced labour and slavery will therefore be more successful in an overall environment of commitment to workers’ rights, including migrant workers, as well as labour standards for all (McGrath, 2017).

One element which, if present, initiatives are more likely to be successful, relates to their enforceability. If regulations do not stipulate enforcing mechanisms and significant consequences for those who do not comply, the probability of success is rather low. The issue of enforceability of various regulations aimed, among others, to prevent exploitation and trafficking from happening alongside compensating those who have suffered from trafficking is also addressed in two interconnected sub-studies in DemandAT, presented in the next section on specific types of intervention.

2.3 Specific types of interventions
In the following section, specific types of intervention against trafficking – law enforcement on the one hand, and campaigns, on the other are examined.
2.3.1 Law enforcement

Trafficking in human beings is a crime, defined in the international legislation and transposed in various different national ones. Eventually, trafficking cases come down to the binary decision on whether a situation entailed the crime of trafficking or not, a decision which resides with law enforcement authorities (who gather evidence in this sense) and ultimately with the judiciary. Analysing whether and how demand in the context of trafficking is addressed by law enforcement authorities came therefore as a necessary study.

Labour strand

Exploitation is one of the defining elements of trafficking in human beings. In theory, without the purpose of exploitation, there is no trafficking crime. The definition of trafficking includes labour exploitation. Labour inspections and comparable agencies monitoring compliance with labour law standards and imposing sanctions a particular role have widely been regarded as key institutions in addressing situations of labour exploitation.
The aim of the sub-study on labour inspectorates and comparable agencies was to examine law enforcement in labour markets regarding the reduction of demand in the context of trafficking in human beings for the purpose of labour exploitation. More precisely, the study was interested to understand whether and to what extent labour inspectorates are expected to address demand and whether labour inspectorates influence or regulate demand in this context. In addition, by employing a regulatory state approach, the study looked into modes in which labour inspectorates approaches can be categorised in terms of their regulatory patterns. Case studies of relevant agencies were conducted in five EU countries. A case study was defined as an organisation or an actor involved in supervision or enforcement of labour or other laws aiming to reduce exploitative practices in employment (Bachinger et al., 2017). Case studies included Labour Inspectorates in Austria and in the Czech Republic, the Inspectorate for Social Affairs and Employment (Inspectie Sociale Zaken en Werkgelegenheid (SZW)) in the Netherlands, the Gangmaster Licensing Authority (GLA) in the UK and the Labour Inspectorate (Arbeitsschutz) and the Authority Monitoring Unreported Employment (Finanzkontrolle Schwarzarbeit, FKS) in Germany. In addition to extensive research on these case studies , a basic mapping questionnaire was sent to all EU MS and to Switzerland. All case studies involved extensive literature review and interviews with key stakeholders.

The study found two types of labour inspection systems – a comprehensive and a dual system. In the first system, found in the Czech Republic and in the Netherlands, labour inspectorates deal with both, protection and prevention in the area of occupational safety and health (OSH) and criminal investigations (although these areas are assigned to different departments). In dual systems, like in Austria and Germany, the tasks of monitoring OSH issues and those of monitoring employment and payment issues (including fiscal and social welfare fraud and undeclared unemployment) are separately addressed. The first is dealt with by labour inspectorates, while the second is dealt with by financial police. The different authorities in dual systems follow different principles. While labour inspectorates see their role as protectors of employees in the workplace, the financial police seems to see their role as to “ensure fair and equal terms [for] the participants in economic life and to protect the financial interests of the state” (Ibid.). However, how these principles play out on the ground vary. The study found that authorities under research are not specifically tasked with investigating trafficking for labour exploitation, but that this is rather a side aspect embedded in their routines. When suspicion of THB-LE arises, they pass on the information to the police. But their role is much broader. For instance, the Financial Monitoring Unit to Combat Illicit Employment (FKS) in Germany fulfils labour standard controls such as in minimum wages, but also social security tax fraud. Due to its police-like functions they are also in charge of alerting the public prosecutor to other cases of crimes, including those relating to irregular migration, even though this is no priority in their work.

One finding of the study is that these authorities report suspicions of trafficking or labour exploitation to the police, which is the main actor when it comes to identifying and investigating trafficking, together with prosecutors. Although labour inspectorates are involved in the efforts to tackle trafficking for labour exploitation in all countries under study to a greater or lesser extent, other priorities dominate their approach, particularly combating of undeclared and bogus forms of employment and in fact, the focus is to a great extent on undeclared (“illegal”) migrant labour (Vogel, Rogoz and Kraler, 2017). Labour inspectors see their main task as combating different forms of labour exploitation. They argue that if opportunities for exploitation are reduced through the deterrent effect of inspections and sanctions, incentives for recruiting persons into exploitative labour relations are reduced.

The GLA has a different approach than authorities in the other four countries. The GLA has inspecting and sanctioning powers in a limited number of sectors and operates through a licensing system in these sectors. This approach was both praised for its presumed deterrent and steering effect in these sectors and blamed for its rather too specific focus which leaves relevant sectors outside of its scope. Labour inspection services addressing labour exploitation expect to reduce the grounds for trafficking for labour exploitation more effectively with a broad approach to the reduction of exploitation than by focussing on trafficking only.

In addition to issue such as limited resources, limited mandates or lack of political will to allow/design mandates for labour inspectorates to take a more active role in addressing trafficking, some issues with regard to the effectiveness of their approaches have been raised. In cases in which inspections are reduced to fulfilling paper work, situations of labour exploitation can only rarely be identified. In cases in which labour inspectors are expected to identify illegal employment and signalise the irregular status of employees to responsible authorities, it has been argued that both employers and workers will not be willing to cooperate with labour inspectors. In addition, there are conflicting interests inherit to labour relations, but these are increased by the fragmentation of labour markets, regulations and law enforcement actors.

All these authorities address trafficking mainly through command and control strategies, and to a limited extent through peer pressure approach. The effectiveness of their approach depends on several factors: scope of regulations and power of labour inspectorates to impose sanctions and remedies, the likelihood of inspections and detection (relevant for the deterrent effect), as well as the quality of sanctions (penalty level, appropriateness etc.). The next section looks into the other actors mandated with enforcing the law – the security sector.

Security strand

The security sector, for the purpose of this research, included public authorities such as: the police, gendarmerie, intelligence services, border guards and immigration services, prosecutors and judges, as well as specialised counter-trafficking agencies, National Rapporteurs and equivalent monitoring mechanisms, as well as NGOS interacting and working with public bodies maintaining general public security, law and order (Haxhi et al., 2017). This particular sub-study looked at the role of the security sector in addressing demand in the context of trafficking in human beings.

Addressing demand, according to international law , subscribes to actions intended to prevent trafficking in human beings. Security sector actors have often argued in interviews, particularly those involved in the criminal justice sector, play a key role in addressing demand in the context of trafficking by “making human trafficking a less profitable economic undertaking and by lowering impunity” (ICAT 2014). Demand in an economic sense – as suggested in this study – did not play a big role for law enforcment actors interviewed in this study. Legal market activities that are related to trafficking in human beings are not in the focus of interest of law enforcement actors. Instead, law enforcement actors seem to regard demand in relation to trafficking not in economic terms, but as a “source of criminal activities” (Haxhi et al., 2017). The analytical framework employed in this particular case study was structured “on the criminal law basis, i.e. the security sector’s role is to prevent, deter, disrupt, prosecute and convict criminal or law-violating activities” (Ibid.).

In addition to extensive desk research, main research methods used were standardised questionnaires sent to 26 European countries (out of which 12 were returned) and in-depth case studies conducted in selected countries: Czech Republic, the Netherlands, Switzerland and the UK. Each case study included desk review and semi-structured interviews with relevant stakeholders. The selection of case studies was a combination between legal systems, geographical location within Europe, relevance of the demand-side as it came out from the initial literature review, as well as the expected degree of cooperation between security sector actors and labour laws enforcement actors.
One of the results of this research is that security sector’s role in addressing demand in the context of trafficking is strongly linked with the national legislation, including mandates and jurisdiction. Recorded differences between national approaches are also depending on legal and administrative structures of countries.

Some of the stakeholders interviewed mentioned lack of resources as a major challenge for the security sector’s general approach to trafficking. According to some interviewees, greater resources mean, in addition to financial and human resources, “better distribution of tasks and/or case investigation and reporting, as well as police intelligence on the streets and one-on-one face-to-face status reports” as well as “more detailed and extended operations with joint investigations” (Haxhi et al., 2017). All interviewees indicated that funds for specialised anti-trafficking training session are insufficient.

Awareness-raising through community policing or through information campaigns have been mentioned as possible measures which security sector actors can use in their work. Although not a case study in this research, community policing was reported a relevant measure implemented in Northern Ireland, where police is trying to reduce Organised Crime Groups activity through “continuous interventions, stakeholder engagement and a sustained investment to change mind sets across” (Ibid.) the general public. In countries where prosecutors work together with the police in investigations, whether a case is build up as a trafficking case depends among others on the prosecutor’s awareness of trafficking. A way of promoting a more equal approach to trafficking cases, in the Netherlands for instance, results of cases are published on internally shared websites.
To sum up, for the security sector, addressing demand in the context of trafficking translates into investigating, prosecuting and convicting those internationally involved in criminal activities. The role of the security sector in addressing demand is linked with legal provisions in a particular jurisdiction and in particular areas – e.g. it depends whether law enforcement authorities at the level of cities and/or cantons/or federal states have mandates implementing and elaborating legal provisions to regulate specific activities such as begging, selling and purchasing sexual services etc. One important limiting issue when it comes to security sector’s activity relates to the “crime knows no border” problem (Ibid.). Due to the fast changing modus operandi of organised crime, including those organised crime groups involved in trafficking in human beings, it has been argued that “the security sector is most of the time only reacting to changes in the structures and operations or organised crime instead of proactively being one step ahead of them” (Ibid.).

Awareness-raising (either though community policing or though information campaigns) was mentioned as a demand-side measure which has the potential to support law enforcement efforts to address trafficking in human beings. Little is known about the impact of such measures and whether indeed they produce the expected effects (in this case, awareness-raising implies that some action is also expected from members of the public who are expected to support law enforcement’s work in fighting trafficking). How such information campaigns are intended to work is one of the topics researched in the sub-study in communication campaigns.

2.3.2 Communication campaigns

Information campaigns with the aim to reduce trafficking in human beings are supported by various donors and conducted extensively by various implementing organisations. According to Walby et al. (2016), the EC alone spent more than 158 million EURO on 321 anti-trafficking projects between 2004 and 2015, out of which only 5% were externally evaluated. Despite campaigns’ popularity in the anti-trafficking field, little is known about the ways in which campaigns work and to what extent they are effective in bringing about the change they seek.

The research on demand-side campaigns analysed existing evaluations of implemented demand-side campaigns and drew key messages with regard to what can be improved in this context. For the purpose of the research, a campaign was defined as a “series of coordinated communication activities seeking to influence what individuals think and do” (Cyrus and Vogel, 2017). Anti-trafficking demand-side campaigns are considered those addressing “persons as consumers who knowingly or unknowingly pay for work or services of trafficked persons” or those addressing “the general public who is expected to observe seemingly suspicious occurrences of exploitation, often in a consumption context” (Ibid.). Within these demand-side campaigns, the study identified spending-shift campaigns – those aiming to influence spending patterns of target groups – and reporting campaigns – those campaigns asking the members of the targeted groups to report suspicions of undesired behaviour. The main question of the study was “how spending-shift and reporting campaigns can have an impact on situations outlawed in international law as trafficking in human beings and how the effects of campaigns can be assessed” (Ibid.).

This study looked at campaigns which address a situation in what can be called a short supply chain – the consumer of a good or a service and the exploited person are present in the same area (see on initiatives relating to global supply chains are addressed in the section 2.2.3). The study identified 55 campaigns addressing demand in the context of trafficking and developed a typology of these demand-side campaigns. Those campaigns aimed at awareness raising only (with no request for a particular action) were excluded as an impact requires some sort of action by the target group and cannot result from awareness alone. After an intensive search for more information and contacting 21 campaign-conducting organisations, access to evaluation reports for eight campaigns or projects containing campaigns was available. These efforts show that in spite of the accounting efforts presented on the EU anti-trafficking coordinator’s website, it is not easy to locate detailed and evaluative information.

In the absence of robust theoretical considerations according to which effectiveness of campaigns could be assessed, the research proceeded first with the development of a theoretical frame. Researchers reconstructed the intervention logic of a campaign (See Cyrus and Vogel 2017) and developed a stylised logical chain of necessary effects of an anti-trafficking demand-side campaign (see Annex 2). In addition to the intended effects of a campaign, presuming that no interruption in the logical chain of effects occurs, there are also side-effects to consider. Some of these can be positive for the overall aim of the intervention, such as financial donations to organisations supporting behavioural change in the same direction. But there can also be unintended effects, such as stigmatising social groups or gender stereotyping, reinforcing overall attitudes leading to behaviours that the campaign is actually fighting to change.

Research on anti-trafficking demand-side campaigns aimed at behaviour change reached results with regard to both, the effects and the evaluation of such campaigns. Regarding demand-side campaigns’ effects, research found that awareness-raising alone, has no impacts unless it leads to a change in behaviour. In addition, campaigns which focus on cases of trafficked persons only neglect those who are in situations of exploitation and for whom it is not obvious whether they have been trafficked or not. Research found that effectiveness can be increased when campaigns are sector specific but also address exploitation generally.

Research into the evaluations of spending-shift campaigns could not provide for robust evidence that they have produced a change with regard to reducing exploitation and trafficking in human beings.
Regarding the effectiveness of reporting campaigns, particularly in terms of resources spent, research found that they “expose large audiences to a message but generate only a handful of relevant suspicious occurrences per month” (Vogel and Cyrus, 2017).

Campaign messages can be misunderstood. One documented case concerns a campaign asking the public to call a helpline in case of suspicions in which a third of the callers misunderstood the campaign posters as an advertisement for phone sex (ibid: 3). In addition, campaigns can produce harmful side effects induced by the choice of message. Besides reinforcing stigma on particular groups or victimising those affected by trafficking, campaigns can overlook cases of exploitation not considered as “extreme” or can induce the idea that trafficking is “everywhere around us” which can lead to reporting of false suspicions.

Similar to findings in other studies in DemandAT, this study found that for a positive impact on reducing exploitation and ultimately reducing trafficking in human beings, multiple interventions are to be implemented. This finding is supported, among others, through the fact that campaigns rely on other measures and policies, such as an adequate law enforcement response when indication to possible trafficking cases is reported, as well as proper support offered those who are reportedly in situations of exploitation.

As it turns out from the analysis of evaluations, internal evaluations were more learning oriented than the external evaluations. In fact, form the four external evaluations analysed in the study, “only one referred to established standards as developed by professional bodies and promoted by the United Nations” (Vogel and Cyrus, 2017). Finally, the study concluded that designing evaluable interventions is key for improving learning in this area. The focus of the analysis based on evaluation theory and the established standards for evaluations is one of the major contributions of the DemandAT project, as elaborated in one of the following sections.
Potential Impact:
3.1 Strategic Impact

The questions raised by the analysis of demand-side approaches in anti-trafficking are far-reaching. Empirical research undertaken in DemandAT on various policy approaches – such as addressing demand through law enforcement, through initiatives in supply chains or through communication campaigns – revealed key aspects of successful initiatives and signalised their limitations. Since trafficking in human beings is a field in which evidence is difficult to produce and easy to contest, when it comes to impacts of the project it is important to mention that research aimed at providing clear arguments and focused evidence where possible, but whether they are taken up in a way to reduce trafficking in human beings and related offences is up to policymakers and other actors in the field. The project proposal promised to contribute to the following aims:

• Contributing to Improved communication
• Presenting “smart” policy choices and
• Providing arguments for informed choices on selected policy issues
3.1.1 Better communication in the field of anti-trafficking

With regard to improved communication, a clear concept of demand and a consistent use of the term were suggested. The idea was to contribute to improved communication in the field by suggesting that the most frequently evoked meaning of demand in an economic sense should be used consistently: demand as ‘willingness and ability to buy goods or services’ in a market context. The conceptual analysis undertaken in DemandAT showed that referring to demand indicates that someone wants and can buy something at a certain price. However, it showed that trafficking in human beings does not only mean buying and selling of people. It is embedded in a range of different market relations in supply chains of varying length. In addition, the concept is not suitable for describing the relation between a coercive trafficker and a victim of trafficking. Demand and supply refer to a market relation which is fundamentally voluntary, although decisions may be taken under difficult conditions. A coercive relation is not a market relation so that the use of terms like supply and demand seems inadequate . For this reason, one can conclude that “at the core of understanding trafficking in human beings is a non-market relation” (Vogel, Rogoz and Kraler, 2017). This has implications for both, how trafficking is being referred to and what should be taken into consideration when designing anti-trafficking measures.

Analysing demand in the context of trafficking in human beings has shifted the focus of the anti-trafficking actions on those potentially gaining from exploitative practices. In so doing, the enquiry had to look at the element of exploitation, since policies are aiming to address “the demand that fosters all forms of exploitation” (UNODC, 2014). One limitation of the approach is that, with the exception of forced labour, there is no internationally agreed upon definition of exploitation. Furthermore, states do not have an obligation to compensate victims of exploitation, as opposed to compensation for victims of trafficking and special status offered to migrant victims of trafficking (Cyrus 2015). But there are important prospects opened up by looking at demand in the context of trafficking. Vogel (2017) distinguishes between at least three types of markets in which abuses can take place: a market of goods and services in which clients demand a particular good or service from a producer or a service provider; a labour market in which the producer/provider/employer demand labour from workers or demands intermediary goods from other producers; and there is another market in which abuse (including exploitation and trafficking) can occur, market in which migrant workers demand irregular border crossing services from smugglers or demands, for instance, a credit from formal or informal lenders. When designing a demand-side measure to address trafficking in human beings, the research conducted in DemandAT suggests to always spelling out the market in which the measure is supposed to work, as well as how exactly it is expected that the particular measure will address trafficking.

As mentioned above, the project combined theoretical analyses of the concept of demand (Vogel, 2017a) and different types of policy interventions (Boswell and Kyambi, 2016) with a mapping of policy debates (Cyrus and Vogel, 2015; Cyrus 2015) and implemented policies across a large range of countries (Rogoz et al., 2016), as well as different fields in which trafficking occurs with in-depth studies of particular fields (domestic work, sex work) and approaches (supply chains, law enforcement, campaigns). With regard to policy choices, the combined results of the different strands of research undertaken in the project have a number of important implications for policy formulation and policy evaluation in the field of anti-trafficking. Some of these results, as well as arguments for informed choices, are summarised below.

3.1.2 “Smart” policy options

Empirical research on initiatives addressing TFLS practices in and around global supply chains showed that the context in which a transaction takes place is relevant. Initiatives addressing exploitation in global supply chains are specific to the industries and level at which they operate. Markets where exploitation occurs are relevant because of the nature of the goods/services transactions and the contexts in which are being produced/offered/consumed. Domestic work, for instance, is employed inside private homes and it can be particularised to the type of domestic work required (e.g. care work for dependant elderly). Research found therefore that the context and the nature of the work are relevant for addressing exploitation in the domestic work sector. For this reason, one of the conclusions of the project was that policies focusing solely on the demand–side cannot be expected to provide solutions to human trafficking. As relevance of demand varies according to different areas where THB occurs, addressing demand has to be done in parallel with other preventive measures as well as with protection measures of those potentially affected and in partnership with all stakeholders engaged in tackling THB.

Awareness raising is considered to be a preventive measure in the anti-trafficking policies. Research on awareness raising campaigns conducted within DemandAT shows that although a large audience is targeted, only a few members of the public change their behaviour (e.g. a small number of people reports potential exploitative situations to anonymous hotlines). Therefore, awareness raising as such should be considered against its potential impact and not only as a label in prevention of trafficking. Research found that awareness raising alone is insufficient as an intervention. To sum up, demand-side measures have to be custom made for particular markets. Involvement of those affected by exploitation and/or trafficking in the elaboration, implementation and monitoring of demand-side measures is more likely to lead to progress in addressing exploitation, including trafficking. In conclusion, ‘demand-side’ policies and measures shape purchasing conditions in a market context. Such measures need to be combined with other anti-trafficking efforts.

With a comprehensive mapping exercise in different countries and fields, the project provided for an inventory of policy choices that are addressed as demand-side, taking also stock of different understandings of demand in the respective contexts. Insights from other fields – illegal drugs, tobacco and irregular employment – were imported. It became particularly clear that demand side measures – whatever the explicit or implicit definition in the respective field was – are no stand-alone measures. In other words, anti-trafficking policies do not function in a vacuum, but operate within national and international legal contexts. These contexts are highly relevant. Transferring insights from one policy area to another or from one country to another is difficult. In addition, trafficking in human beings rests on exploitative relations that are embedded in many different market contexts. Exploitation in agriculture and construction, in domestic work and prostitution is embedded in markets with different characteristics. For example, domestic work and sex work are characterized by an environment in which most activities are performed unpaid within relations inside private rooms – an environment that offers options to hide the exploitation of workers who clearly aim at gaining an income with their activity. For construction and agriculture, fluctuations in labour demand and the predominance of temporary workers are more relevant circumstances. In some situations policies focusing on the demand-side are harmful (such as some measures employed in repressive policy regimes when it comes to regulations in the sex work sector). 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. Extreme cases are found concerning domestic workers of diplomats – an environment where tied working visa is combined with virtual immunity of potential exploiter. One conclusion of the study was that policies addressing demand in the context of trafficking need to be contextualised to specific policy areas and not used as a general measure for all types of exploitative situations. Particularly as in some trafficking situations the demand-side is not relevant (e.g. demand in the context of trafficking for the purpose of exploitation through begging). Without a meaningful use of the concept of demand in particular anti-trafficking policies (explaining how the expected outcome is achieved by addressing demand), ‘demand’ can become just a label for policies that would be in place anyway in the context of THB. Demand is an economic concept and this research suggests making use of it in an economic way, by referring to the market in which the measure is expected to have an effect. In conclusion, anti-trafficking policies, including demand-side measures, need to be context specific and custom made for particular markets.

All empirical work in this project had to deal with the fact that while much is written on measures against trafficking, empirical evidence on the impacts is rare and difficult to produce. In such a field, conceptual and theoretical clarity and close scrutiny of the available evidence is necessary for improvement. This study has shown that in-depth analysis of case studies with all relevant stakeholders as well as case law analysis can highlight mechanisms and features relevant for the emergence of trafficking in human beings. Particularly relevant are those case studies where there is solid evidence, according to international established standards, of reducing exploitation (Vogel, Rogoz and Kraler, 2017). Evaluability is key for better learning from past interventions and improving future ones. Evaluation must be planned in advance and resources allocated accordingly for both, learning-oriented self-evaluation and external professional evaluations (using established standards as developed by professional bodies). Through publicising evaluation reports it is expected that organisations learn from one another. For this, evaluation should be recognised as a success in itself, as learning from past interventions is crucial for improving future ones.

Considering the above and based on the empirical results of the various sub-studies conducted in the study, demand-side and other interventions against trafficking should be both context specific and evidence based. Context specific policy making should acknowledge that policy impacts relate to a wide range of contextual factors including national, sector specific and organisational factors. One implication of this is that transferability of measures from one area to another is difficult and sometimes the transfer of a possibly successful measure from one context not helpful in other contexts. Effects of policies need to be critically assessed against agreed-upon standards and also by considering the contexts in which they operate, such as, for national policies on labour migration. In conclusion, this research suggests making best use of available evidence and improving evaluation of measures.

Research on labour inspections in domestic work showed that people are trafficked into exploitation of various and increasing degrees. A sole emphasis on extreme situations is unhelpful in identifying forced labour and trafficking in human beings. With regard to trafficking for labour exploitation, there is a need to look beyond the point of contacting (demand, supply, price) and to look at the relational aspect of the employment arrangements. Vulnerabilities may not be obvious at the beginning of a labour relation, but can evolve or come to the attention of the employer in the course of a relation. In that sense, combatting labour exploitation is a prevention measure against the emergence of more severe forms of exploitation. Access of (migrant) workers to mechanisms for reporting exploitative situations is expected to contribute to detecting and preventing trafficking for labour exploitation. In addition, situations of severe exploitation may not meet all trafficking criteria and situations might not be addressed. If resources are mobilised for addressing exploitation, trafficking cases are less likely to be overlooked.” (Ibid.). In conclusion, as resources to fight trafficking are limited, addressing situations of various degrees of exploitation can contribute to detecting and preventing trafficking.

3.1.3 Promoting informed policy choices

The project sought to impact on the real world by making more informed choices on selected policy issues possible. With regard to five selected issues – prostitution policies, domestic work, global supply chains, law enforcement approaches and information campaigns – the project aimed to provide in-depth consistent arguments, built on best available evidence and investigation and make it accessible for policy-makers. As the results of this project phase are only finalised at the time of writing this report, it is too early to assess whether policy makers actually take an interest in the opportunity for more informed choices based on the results. Project partners took a pro-active stance when it comes with informing relevant policymakers about the results of the project and engaged with a limited number of stakeholders (limited by the budgetary and time constraints of each workpackage) in workshops organised in the second and third phase of project implementation.

3.2 Dissemination activities of the project

DemandAT followed a three pronged approach to dissemination - (1) dissemination of project results through a communication infrastructure established specifically for the project (the project website, announcements on important project results or events through a newsletter tool and dissemination through social media, notably twitter and youtube), (2) engaging with relevant stakeholders in a systematic manner throughout the project, in three ways, through (a) establishing an overall project advisory board as well as a specific workpackage advisory board for WP9 (Global Production of Goods), (b) using the opportunity of expert interviews in various workpackages to engage with and maintain relations with relevant stakeholders, and (c) organising a range of stakeholder workshops, both within individual workpackages in the project’s second research phase and by organising two stakeholder event for the project as whole. Third (3), DemandAT pursued an opportunity oriented approach, using available opportunities (such as academic or policy oriented conferences) to disseminate results. The project’s dissemination activities were continuously monitored and the dissemination strategy regularly adjusted during the project lifetime.

3.2.1 Dissemination of project results through a communication infrastructure established for the purpose of the project

A project website ( was set up in the beginning of the project. Set up as a slim and easily manageable portal, the website was designed to provide basic information on the project, project partners and the project advisory board and to disseminate the project’s main publications as well as shorter content (blogs), written both by project researchers and experts outside the project consortium. In addition, the project website was used to post event and other announcements on related issues. After the end of the project, the website will be maintained for at least another 6 years until June 2023. Website administration moved to ICMPD with the end of the project, the costs for maintaining the website beyond the project are covered by ICMPD from own funds.

In addition to the project website, a twitter account was set up to post announcements on publications and project related events.

Finally, while the project never formally established a regular project newsletter, a newsletter tool was set up by ICMPD to send out publication and event announcements at important points in the projects. Altogether, 6 announcements were sent out via this too to some 1200 recipients.

3.2.2 Engaging with relevant stakeholders in a systematic manner throughout the project

Stakeholder advisory boards

DemandAT systematically involved relevant stakeholders throughout the project. A project advisory board set up prior to the project start involved representatives from key organisations active on trafficking, including Bandana Pattanaik from the Global Alliance against Trafficking in Women (GAATW), Irina Todorova from the International Organization for Migration (IOM), Zuzanna Muskat-Gorska from the International Trade Union Confederation (ITUC), Ruth Pojman from the Organisation for Cooperation and Security in Europe (OSCE), and Kristiina Kangaspunta from the the United Nations Office on Drugs and Crime (UNODC). In addition, Roger Plant, a former Head of the ILO's Special Action Programme to Combat Forced Labour ILO was involved in a personal capacity.

For the research on global supply chain initiatives a separate advisory board was set up and involved Anannya Bhattacharjee, International Coordinator of Asia Floor Wage Alliance in New Delhi, India; Fathallah Omrani from the General Trade Union of Workers in Textile Garment and Clothing Industries in Jordan; Jin Sook Lee, from the Building and Wood Workers’ International (BWI), Geneva; Leonardo Sakamoto from the Brazilian NGO Reporter Brasil; Laura Germino, Greg Asbed and Steve Hitov from the Coalition of Immokalee Workers (CIW), USA; and Andrews Tagoe from the General Agricultural Workers‘ Union (GAWU) in Ghana.

Both the overall project advisory board as well as the specific workpackage advisory board helped to reach out to further stakeholders and disseminate project results. In addition, results were shared with the advisory board members for comments.

Reaching out to stakeholders via interviews

Overall, some 330 stakeholders (researchers, policy-makers, NGO representatives, trade unionists and others) were interviewed for the purpose of this project. Stakeholder interviews were an important means of mapping the stakeholders in the field and apart from relevant substantive information also generated information on relevant contact persons, including interviewees beyond those identified by project researchers. Results of the respective research reports were also shared with interviewed stakeholders and engagement with stakeholders through interviews thus was an important means for engaging with stakeholders overall.

Engaging stakeholders through workshops throughout the project

Two events were organised on the level of the overall project. A small-scale mid-term dissemination event held in Brussels in June 2015 targeted a small invited Brussels based audience and aimed at disseminating preliminary results of the research and focused on conceptual aspects and policies to steer demand. The final DemandAT conference held in May 2017 and attended by about 50 participants focused on key results from the five case studies conducted in the project as well as the overarching theoretical and conceptual results of the project. In addition from these two events catering for the project as a whole a workshop on trafficking for severe labour exploitation was organised by the team at the European University Institute in October 2016, involving several partners from the project dealing with trafficking for labour exploitation.

In addition, the project involved a number of other stakeholder events organised within individual workpackages.

With regard to the study on demand-side campaigns, three workshops with stakeholders from NGOs, donors and evaluation researchers organised by DemandAT researchers at the University of Bremen helped to sharpen the arguments and formulate understandable and convincing summaries of results. NGOs responding to calls were encouraged to learn in a feasible way, with support of a manual for designing evaluable campaigns in response to calls. The project partners joined other voices with a message to donors such as the EU, asking for better and more accessible evaluation of measures against trafficking in human beings.

Similarly, preliminary results of the sub-study on law enforcement, both on the labour strand and the security strand, were discussed with relevant stakeholders in two stakeholder workshops held in Vienna and Geneva, respectively.

The sub-study on domestic work involved national researchers for the eight national case studies whose engagement supported the involvement of relevant national stakeholders. Apart from the overarching workshop on Trafficking for Severe Labour Exploitation organised in October 2015, a specific stakeholder event was co-organised in collaboration with the nationally funded Trafficko workshop, involving a total of 65 participants. That event also involved a side-event featuring the UN Special Rapporteur on Trafficking in Persons, which reinforced the collaboration with the latter, judge Maria Grazia Giammarinaro.
3.2.3 Open access publication strategy with a series of policy briefs specifically tailored to the needs of stakeholders

All major project results have been made available freely accessible via the project website, including – to date – 15 working papers, altogether 14 policy briefs as well as a number of other publications, blogs and presentations. While working papers and project reports arguably mainly serve the needs of the academic community as well as stakeholders used to engage with academic research outputs, the policy briefs, blogs and videos produced in the framework of the project specifically aim at a wider audience, including both expert and non-expert communities. Altogether, 30 blogs have been produced in the framework of the project which, apart from the project website, were also disseminated via other channels, such as the Open Democracy blog platform or the ICMPD expert voice blog series. As a specific output tailored to stakeholder needs the manual on designing evaluable campaigns written by DemandAT researchers at the University of Bremen also should be highlighted. Based on critical research of various campaigns, it translates findings into concrete actions needed to improve the quality of campaigns.

Multimedia outputs

Individual project partners developed a series of videos in order to promote project results. For instance, a short video clip was produced by the European University Institute and the Global Governance Program communication team which included excerpts from 3 interviews conducted with 3 speakers during the targeted dissemination event on the 12-13 of October 2016 at EUI (see

Similarly, Durham University, developed a video on the research on global supply chain initiatives ( highlighting four Principles to Address Trafficking, Forced Labour and Slavery in Supply Chains

For the project as a whole, three short videos present the results of the theoretical and conceptual analysis of the work (Part 1), the results of research on different fields in which trafficking occurs (Part II) and the results of specific means of intervention (Part III). The videos are accessible at

All results reviewed by project partners and by the Advisory Board Members are available on the project website:
3.2.4 A pro-active opportunity based approach

An important element of the project’s overall dissemination strategy was to take opportunity of relevant academic conferences, stakeholder events and initiatives organised outside of the project.

Thus, researchers from the workpackage on Domestic Work presented research at academic conferences in Vienna, Poznan and El Paso, amongst others. Researchers from the University of Durham presented results of their research on global supply chain initiatives at several international conferences both in the US and in Europe. DemandAT researchers at Lund University presented the results of their research on prostitution policies in seminars and international conferences in Oslo, Leipzig, Wellington (NZ), Sydney (Australia), Copenhagen, Brussels (twice) and Vienna. Stakeholders, human rights organisations and academics from many different countries have already begun using the typology, in media interviews, articles, papers, etc. It will also be used by other scholars in upcoming textbooks on prostitution policies. Furthermore, the EU funded research project DESiRE, examining “demand for sexual exploitation” ( ) builds on the conceptual work of the DemandAT project and use the typology for their study on sexual exploitation and trafficking. Finally, not only scholars from the prostitution policy field have shown an interest in the typology, but also scholars studying welfare and drug policies

Furthermore, DemandAT researchers have been invited to participate in trainings, for example by the Academy of European Law (ERA, Trier with a training in Thessaloniki). Several project researchers have also been invited to wider stakeholder conferences, such as an OSCE conference on the Prevention of Trafficking through Supply Chains in October 2016 or a conference organised by the Dutch EU Presidency in early 2016.

Project partners are also committed to continue disseminating project results beyond the end of the project, both in the framework of existing networks or initiatives they are part of as well as in the context of new collaborations.

ICMPD for instance will utilize its involvement in the consultations for the Global Compact for Migration as well as its membership of Alliance 8.7 (referring to objective 8.7 of the Sustainable Development Goals) to feed project results into global policy debates. In addition, its active engagement in various Regional Migration Dialogues as well as other regional networks, such as the network of National Anti-Trafficking Coordinators in South-Eastern Europe allows disseminating results at regional levels. Researchers at the University of Durham on the other hand have been consulted by ILO in view of a new initiative to develop a global business alliance against trafficking.

Overall, this opportunity based approach ensures a broad dissemination strategy targeting both academic and non-academic audiences.
List of Websites:

Albert Kraler, Senior Researcher
International Centre for Migration Policy Development (ICMPD)
Gonzagagasse 1, 5th floor, A-1010 Vienna
Phone: +43-(0)1-5044677-2345