The UK’s approach to tackling modern slavery through the aid programme

Introduction

Overview and approach

Since the adoption of the 2015 Modern Slavery Act, ending modern slavery has become an important objective for the UK government – both domestically and internationally through the aid programme. Modern slavery is included as an objective under the Sustainable Development Goals (SDGs): Target 8.7 aims to eradicate forced labour, end modern slavery and human trafficking and eliminate child labour in all its forms.

ICAI has carried out a review of the UK’s effort to tackle modern slavery internationally through the aid programme, focusing on the period since November 2014, when the UK’s first modern slavery strategy was launched. The review assesses modern slavery programming undertaken by the three responsible departments (the Department for International Development, the Home Office and the Foreign and Commonwealth Office), as well as the UK’s work with a wider set of partners, including other donors, multilateral partners and countries, to eradicate modern slavery around the world.

This literature review was prepared as part of the ICAI modern slavery review. It provides a brief overview of the literature on modern slavery, and provides evidence from the literature to respond to a number of specific research questions set by the review team. The ICAI review was undertaken in the early months of 2020. At this time, the impacts of the COVID-19 pandemic on modern slavery, although expected to be far-reaching, were not yet reflected in the literature. Consequently, they do not feature in this literature review. Corresponding to the research questions, the literature review is organised into five chapters:

1. What is modern slavery? This section explores definitional issues and typologies.
2. Prevalence issues: This chapter summarises published data on the prevalence of modern slavery and the major trafficking routes.
3. Causal issues: The third chapter explores what is known about the root causes of modern slavery and the business models – including criminal enterprise – that give rise to the problem.
4. What works in tackling modern slavery? This section explores the evidence base on what kinds of intervention are effective in tackling modern slavery.
5. Victim/survivor issues: Who are the most vulnerable to modern slavery?

The research questions are set out in full in Table 1.

The purpose of the literature review is to summarise available knowledge on modern slavery and to assess the quality of the evidence. It covers published materials on modern slavery, human trafficking and forced labour. It includes academic literature, publications from international organisations, government documents and media reports.

The methodology involved searches in social science databases, Google Scholar and within the websites of international organisations and research institutes that are known to work in the areas of modern slavery, forced labour and human trafficking. Search terms were derived from the five core areas and research questions and inclusion criteria were based on whether the publications provided useful information on the research questions.

The literature review should be read in conjunction with the main review, available on the ICAI website.

Table 1: Research areas and questions for the literature review

The methodology involved searches in social science databases, Google Scholar and within the websites of international organisations and research institutes that are known to work in the areas of modern slavery, forced labour and human trafficking. Search terms were derived from the five core areas and research questions and inclusion criteria were based on whether the publications provided useful information on the research questions.

The literature review should be read in conjunction with the main review, available on the ICAI website.

What is modern slavery?

This section explores the nature of modern slavery and its constituent parts, analysing definitional issues (which are themselves a subject of debate in the literature), the typologies included within the modern slavery label, how those separate phenomena relate to each other, and the significance of modern slavery to international development.

The most cited definitions of modern slavery and the areas of controversy

Modern slavery is an area fraught with definitional difficulties. In recent years, ‘modern slavery’ has emerged as an umbrella term for a range of related but distinct issues and phenomena (Jordan, 2011). These include:

• Bonded labour: where people are forced to work to repay a debt.
• Descent-based slavery: where people are born into a class or caste of slaves.
• Forced labour: where people are compelled to work through violence, intimidation or coercion, often for no or inadequate payment.
• Early and forced marriage: where children, usually girls, are married without their consent or forced into sexual and domestic servitude.
• Human trafficking: where people are exploited through the use of violence, deception or coercion and forced to work against their will. People can be trafficked for many different forms of exploitation, such as forced labour, forced prostitution, forced begging, forced criminality, domestic servitude, forced marriage and forced organ removal.

As an umbrella term, ‘modern slavery’ has no precise definition. The UK Modern Slavery Act of 2015 consolidated a range of existing slavery and trafficking offences into two offences:

• Human trafficking, defined as “arranging or facilitating the travel of another person with a view to exploitation”, whether or not the travel crosses an international border.
• Slavery, servitude and forced or compulsory labour.

Those offences are, however, wide-ranging. The Home Office has produced a typology of 17 modern slavery offences, falling under the categories of labour exploitation, domestic servitude, sexual exploitation and criminal exploitation (Cooper, 2017).

The UK government is also strongly associated with the use of the term ‘modern slavery’ in its advocacy for international cooperation in this area – including by lobbying for its inclusion in the SDGs. SDG 8.7 is a commitment to:

“Take immediate and effective measures to eradicate forced labour, end modern slavery and human trafficking and secure the prohibition and elimination of the worst forms of child labour, including recruitment and use of child soldiers, and by 2025 end child labour in all its forms.”

This use of the term ‘modern slavery’ alongside forced labour, human trafficking and child labour, rather than as an umbrella term, has been a source of confusion.

Some of the constituent elements of modern slavery have precise meanings under international conventions and agreements, as follows:

• Slavery is defined according to the 1926 Slavery Convention: “the status or condition of a person over whom any or all of the powers attaching to the right of ownership are exercised” (OHCHR, 1926). Owing to the concept of ownership, slaves are akin to commodities that can be bought, sold or traded. Slavery is a permanent state that cannot be changed, with the owner dictating all aspects of a slave’s life (Jordan, 2011 p: 2 ). The subsequent Supplementary Convention on the Abolition of Slavery, the Slave Trade, and Institutions and Practices Similar to Slavery added debt bondage, serfdom, servile marriages and child labour to the definition (OHCHR, 1956).
• Human trafficking is defined under the UN Convention Against Transnational Organized Crime, which has a protocol on human trafficking: The Protocol to Prevent, Suppress and Punish Trafficking in Persons, Especially Women and Children (also known as the Palermo Protocol). This contains the most commonly cited definition, as follows:

“The re­cruitment, transportation, transfer, harbouring or receipt of persons, by means of the threat or use of force or other forms of coercion, of ab­duction, of fraud, of deception, of the abuse of power or of a position of vulnerability or of the giving or receiving of payments or benefits to achieve the consent of a person having control over another person, for the purpose of exploi­tation” (United Nations, 2003).

The Protocol provides examples of exploitation, including sexual, forced labour, slavery and removal of organs, among others. Beyond these broad categorisations, exploitation is not further defined. Which practices are considered exploitative is therefore a matter for interpretation in each instance (O’Connell-Davidson, 2010). Under this definition, exploitation or coercion may occur at any point, and need not be present at all points. For example, an individual may travel voluntarily to take up employment, only to find himself or herself later subject to coercion or exploitation.

Forced labour is defined by the International Labour Organization (ILO) as: “all work or service which is exacted from any person under the menace of any penalty and for which the said person has not offered himself voluntarily” (ILO, 2012a:12). Forced labour can

• occur in the private economy, including in the form of forced sexual exploitation, or by the imposition of a state, such as forced prison labour. Forced labour among children is defined as “work performed by a child under coercion applied by a third party (other than his or her parents) either to the child or to the child’s parents, or work performed by a child as a direct consequence of his or her parent or parents being engaged in forced labour” (ILO, 2012a: 17). There are also two specific conventions that address child labour: one on minimum age and one on the worst forms of child labour (ILO, 2017a).

One of the arguments against the use of ‘modern slavery’ as an umbrella term is that it replaces constituent elements with relatively precise legal definitions with a vaguer term, which is harder to define and therefore identify or measure. Gallagher has called modern slavery “a made-up concept to which no international legal definition is attached” (Gallagher, 2017: 91). Terminological difficulties in turn give rise to measurement challenges and inconsistencies in prevalence data, sometimes within the same organisation, making comparisons between studies and over time difficult (Weitzer, 2014; Mahmoud and Trebesch, 2010; Jordan, 2011; Piper et al., 2015; Brennan, 2014; Gallagher, 2017).

Some advocacy organisations welcome the rhetorical force of the term ‘modern slavery’ and its association with historical international campaigns to eradicate slavery. Some have supported expanding the definition to encompass a wider range of human rights abuses than those listed above. However, others argue that this would inflate the prevalence of modern slavery while reducing clarity as to what it encompasses, and lead to growing divergence between modern slavery and the definitions of particular abuses used in international law (Jordan, 2011).

There are similar difficulties around determining who is a ‘trafficking victim’, with usage varying across organisations according to their mandates, including whether they are non-governmental organisations (NGO), United Nations agencies or government departments (Ali, 2010). Individuals may also shift between legal categories as their situation changes. Victims of trafficking are often thought of as illegal migrants, but they also include individuals who migrated legally but were later exploited (O’Connell-Davidson, 2010). There is also debate as to whether the label ‘victim of modern slavery’ is helpful to attach to survivors. It may provide access to services, protection and compensation, but may also cause stigma and psychological ill health. Some victims of exploitative labour practices may resent being described as ‘slaves’. O’Connell-Davidson argues that bringing these concepts, particularly human trafficking, under the umbrella term of ‘modern slavery’ is problematic, as it “closes down, rather than opens up, possibilities for effective political struggle against the unfreedoms, exploitation and injustices that many groups of migrants experience” (O’Connell-Davidson, 2010: 245).

In this literature review, the term ‘modern slavery’ is used as an umbrella term, in accordance with UK government usage and the scope of the ICAI review. However, when describing individual studies, the terms used by the authors will be retained.

 What typologies are used to categorise modern slavery?

The broad typologies of modern slavery typically include human trafficking, forced labour, sexual exploitation and forced marriage. Often the concepts are further refined based on specific characteristics: the age of the individual (child vs adult), their legal status (regular vs irregular, documented vs undocumented, legal vs illegal), the purpose of their exploitation (sexual vs labour vs marriage) or a combination of these characteristics (child labour vs child trafficking). These terms are also not well defined and are used differently by different groups for different purposes. In the most recent forced labour report by the ILO and the Walk Free Foundation, modern slavery is depicted as encompassing forced labour and forced marriage, while the former is subdivided into state-imposed forced labour, private sector forced labour, and forced sexual exploitation (ILO, 2017a).

While human trafficking and slavery are legal definitions, they contain constituent concepts, such as exploitation, that are not clearly defined and can be difficult to identify. However, the ILO has produced guidelines on how to measure forced labour (ILO, 2012a: 14). It divides forced labour into three dimensions: ‘unfree recruitment’, ‘work and life under duress’ and ‘impossibility of leaving the employer’. The ILO definition centres around involuntariness and penalty, which are further subdivided by strength: strong and medium. To be considered to have experienced forced labour, an individual must have experienced both an involuntariness and a penalty indicator in any one dimension, of which one indicator must be strong (ILO, 2012a: 21). The ILO has also highlighted a subset of 11 indicators as ‘key indicators’ aimed to support frontline workers in identifying potential victims of forced labour (ILO, 2013). When comparing studies of forced labour, care must be taken to verify that they are measuring the same indicators.

In the UK, the National Referral Mechanism is used to identify potential victims of modern slavery and refer them to appropriate services. Using confirmed cases of modern slavery, 17 different typologies of modern slavery were developed. These typologies are grouped together in four different categories: labour exploitation, domestic servitude, sexual exploitation and criminal exploitation. Labour exploitation includes three typologies: those exploited for multiple purposes in isolated environments, those who work for offenders, and those who work for someone other than the offenders. Domestic servitude includes three typologies: those exploited by a partner, by relatives, or by unrelated individuals. Sexual exploitation typologies include both child and adult sexual exploitation. Child sexual exploitation can be by a group or by an individual, forced sex work can be in a fixed location or in changing locations, or victims can be trafficked for personal gratification. Finally, criminal exploitation includes six different typologies: forced gang-related criminality, forced labour in illegal activities, forced acquisitive crime, forced begging, trafficking for forced sham marriage, and financial fraud including benefit fraud (Cooper et al., 2017).

Although the vast number of documents examining modern slavery focus largely on sexual and labour exploitation, other forms do exist. In the Palermo Protocol, exploitation such as the trading of organs or begging is also highlighted. These other forms of trafficking are less reported, perhaps due to their even more clandestine nature (UNODC, 2018a).

What are the relationships between migration, trafficking, people smuggling and modern slavery?

Migration, trafficking, people smuggling and modern slavery are distinct but interconnected concepts. Migration involves the movement from one geographic location to another and may be internal within a country or across an international border. Individuals migrate for different reasons and not all would be associated with smuggling, trafficking and slavery. The reasons include a lack of local, decently paid employment opportunities, political instability, natural disasters, the demand for cheap labour and climate change (Chuang, 2006; Wickramasekera, 2002). There have been large numbers of fatalities as individuals try to migrate through irregular channels, most recently the Essex lorry incident in the UK in which 39 Vietnamese nationals died (BBC News 2019a). These incidences highlight the precarious journey many undertake.

Some types of migration are labelled as ‘forced’, due to conflicts such as war, persecution (political, religious, sexuality), or natural disasters. Others are considered ‘voluntary’, such as for family reunification (including marriage), education, work or travel. Such types of mobility have been facilitated by improvements in and ease of international travel. Increasingly, migrants are leaving their home environment for economic and employment opportunities as the demand for imported cheap labour rises and workers in source countries see the potential for higher earnings and improved quality of life through temporary labour migration schemes (Huong et al., 2014).

Labour migration has also become a core economic policy in many low- and middle-income countries and in many destination countries that rely on imported labour. Many source countries have established governmental departments specifically to promote and manage the migration process or to negotiate bilateral agreements with key destination countries (Huong et al., 2014; Auwal, 2010; ILO, 2017a; Wickramasekera, 2002).

International migration can be highly regulated with strict guidelines in both source and destination countries, particularly when the purpose of entry is for anything other than tourism, and if the source and destination countries are not equal economic powers – for example, when low-income countries are sending migrant workers to middle- or high-income countries. Between similar economically ranked countries or those geographically close to each other, the passage can be relatively unrestricted with few formalities required. Examples include the Schengen Agreement states in Europe, the South African Development Community and the open border policy for India and Nepal nationals (Düvell, 2011; Shah, 2009). At the same time, many countries, particularly in the West, have severely tightened regulations on immigration and labour policy in recent years, limiting the option for legal migration and work routes.

Migrants who need to cross international borders but are unable to secure the paperwork to do so legally may resort to moving irregularly, by crossing borders at unofficial locations, using false documents or entering with a different type of visa (Mak et al., 2017). In some cases, this may mean using the services of people smugglers. The UN defines migrant smuggling as “The procurement, in order to obtain, directly or indirectly, a financial or other material benefit, of the illegal entry of a person into a State Party of which the person is not a national or a permanent

resident” (OHCHR, 2000). Smuggling can be solely an illegal transport option used voluntarily by migrants to get to their destination across an international border, and where, once arrived, there is no further obligation on either part. Smuggling can also be linked to trafficking, which, as defined earlier, involves the movement of individuals for the purpose of exploiting them. Migrants may be lured through deception or coercion, such as false promises of decent employment, and are sometimes bonded by debt taken out to fund their journey, or ‘advances’ paid by recruiters or employers that they must repay, leaving them with no way out of the situation. This is where trafficking and smuggling can be conflated, because both involve the movement of individuals across international borders, often through clandestine and dangerous means (Beadle and Davison, 2019; Goodey, 2008).

Migrant workers that obtain work in the destination country, even if they initially do so in a voluntary and legal manner, may still fall into situations of exploitation due to inherent power imbalances between them and their employers. A wide range of exploitative labour practices has been reported across a variety of settings and industries. The most common forms of exploitative practices have included contract breaches, restricted mobility, non-payment of wages, and confiscation of identity documents (Mak et al., 2017; World Bank, 2016; Amnesty International, 2011; Environmental Justice Foundation, 2013). Many migrants invest large sums of money either through selling assets or taking out loans to fund their migration, and returning before repaying the loans may push the household into further financial stress.

Prevalence issues

This section explores the available evidence on the global prevalence of modern slavery, with some commentary on the reliability of the data, and looks at the main trafficking routes internationally and to the UK in particular.

What is the evidence of the numbers involved in modern slavery?

There are very few rigorous studies quantifying human trafficking, forced labour and modern slavery (Weitzer, 2014; Mahmoud and Trebesch, 2010; Makisaka, 2009). Many studies that have attempted to do so have relied on small convenience samples or used qualitative assessments where prevalence cannot be estimated.

Estimating prevalence for these phenomena is challenging due to their hidden nature and the difficulty in obtaining accurate accounts. The lack of conceptual clarity and the interchangeable use of the terms by many mean that it is often difficult to compare studies even if they report findings and outcomes using the same term. Additionally, many victims/survivors are reluctant to speak to authorities or researchers about their experiences. This may be because they come from countries where the police are not perceived to be trustworthy, or they are residing in destinations where their legal status is precarious and may not be willing to disclose their situation in case they are arrested or deported (Silverman, 2014). As a result, collecting data in destination countries where migrants may be experiencing exploitation raises many ethical, legal and practical concerns for both those interviewed and the authorities or the research team.

Data on modern slavery, human trafficking and forced labour therefore comes almost exclusively from those who have returned to their country of origin or from organisations providing services to identified victims of trafficking. Although useful, samples of returnee populations cannot be taken to be representative of those who have not or cannot return home. Additionally, the complex global supply chains under which many industries operate, such as garments, electronics, agriculture, and food packaging and preparation, with complex sourcing of components and multiple layers of subcontracting, make the challenge of collecting good quality data even more challenging. For these reasons, some have argued that prevalence figures are better collected at the micro level, using small samples for specific sectors or types of victim, rather than attempting to produce national or regional estimates (Dottridge, 2017). Others suggest that despite the problems with data, global estimates have their place in policy, practice and advocacy (Robinson et al., 2017).

The most cited estimates of the prevalence of modern slavery come from an ILO study, most recently produced jointly with the Walk Free Foundation and the International Organization for Migration (ILO 2017a). Using data largely from nationally representative household surveys in 48 countries, a total of 71,758 participants aged 15 and older were interviewed. Modern slavery experiences were restricted to those that occurred between 2012 and 2016, based on participants’ reporting of their own or someone in their household’s experiences of forced labour and forced marriage, including the country in which they occurred. Forced marriages included those that occurred in the five years preceding the survey, as well as instances where people reported forced marriage before that period and remain in the marriage (ILO, 2017b). The study found that an estimated 40.3 million individuals were in modern slavery, encompassing forced labour and forced marriage. Of these, approximately 25 million were living in forced labour situations (ILO, 2017a: 9). These estimates are conservative, given the hidden nature of such practices and the difficulty in identifying them. In particular, there were gaps in national data, particularly from the Gulf States, as well as from countries experiencing conflict (ILO, 2017b: 18).

Several other organisations and UN agencies have published data on human trafficking. In 2017, the IOM launched the Counter Trafficking Data Collaborative, which provides a publicly accessible anonymised database of victims of human trafficking. The data comes primarily from organisations that provide services to victims. As of 1 June 2020, over 108,000 individual cases representing 175 nationalities exploited in 164 countries were reported (CTDC, 2020).

The Global Report on Trafficking in Persons collects data from 142 countries. More victims were identified in the most recent report compared to any other period (UNODC, 2018a). This may be due to an actual increase in numbers of victims, better recording systems to document such cases, or improved awareness and action by the authorities, local groups or community members. Despite improvements in measurement and data collection methods, the majority of victims identified in the Global Report on Trafficking in Persons were exploited sexually and most were identified in their home country rather than in the destination country. This may indicate that, among the forms of modern slavery, sexual exploitation is the most widely recognised and reported, but not necessarily the most prevalent. The data also reveals that countries with anti-trafficking legislation show increased conviction rates, although in absolute terms the conviction rates are still very low (UNODC, 2018a: 8).

The Global Slavery Index estimated that, in 2016, there were 136,000 people in modern slavery in the UK, equating to a prevalence of 2.1 per 1,000 people (Walk Free Foundation, 2018: 94). The UK collects data on victims of human trafficking from the police, the Home Office, the Border Force, the Gangmasters Licensing Authority and other national, international and voluntary organisations. These data sources are then collated by the National Referral Mechanism (NRM). The National Crime Agency (NCA) reported that there were 2,156 potential victims of modern slavery in the first quarter of 2019, representing a 10% increase from the previous quarter (National Crime Agency, 2019b: 1). The NCA included potential victims, gathered from the NCA intelligence database, in addition to the NRM. There were 6,993 potential victims referred to the NRM in 2018, the most recent full year report and the first that accounted for the change in methodology compared to earlier years (National Crime Agency, 2019a:1). Of these, over 60% of the potential victims referred were males, just over half (55%) were adults and 45% were minors (National Crime Agency, 2019a:5). The main types of exploitations reported were labour and sexual, and represented 52% and 33% respectively among adults, and 63% and 20% respectively among minors. The potential victims came from 130 different countries, of which the most prevalent were the UK (23%), Albania (14%) and Vietnam (10%), with slight variations between adults and minors. For adults, the top three countries of origin were Albania, China and Vietnam, while for minors they were the UK, Vietnam and Sudan (National Crime Agency, 2019a:7). Just over a quarter of the potential victims were reportedly exploited outside the UK – for example, those who experienced trafficking along the migration route to the UK (National Crime Agency, 2019a: 1).

Besides recording cases or potential cases, another method used to assess the prevalence of modern slavery and human trafficking is the multiple systems estimation (MSE) approach. In countries where national data is robust, with at least 80 victims per year and three or more national sources of victim data, the United Nations Office on Drugs and Crime (UNODC) suggests the MSE approach is a cost-effective way of producing estimates of victim numbers, including those undetected by national authorities (UNODC, 2018a: 14). MSE has been tested in the UK using data from the NRM (Silverman, 2014) and four European countries: the Netherlands, Ireland, Romania and Serbia (UNODC, 2018a: 34).

In the examples from the UK and the Netherlands, the estimated number of cases using the MSE approach far exceeds those actually detected. In Silverman’s example, the MSE approach estimated that there were between 10,000 and 13,000 potential victims of trafficking in 2013. This contrasts with the 2,744 confirmed cases identified in the NCA Strategic Assessment of the same year, highlighting the huge range of estimates that can be produced from different data sources and definitions (Silverman, 2014: 1). Gallagher has questioned this methodology due to the difference in findings between these two countries that share a similar risk context but use different criteria in their victim identification (Gallagher, 2017). These challenges with official data, and particularly data based solely on confirmed cases of human trafficking, have also been identified in other settings (Hynes et al., 2018; Beadle and Davison, 2019). The 2019 UK annual report on modern slavery cites the most robust estimates of potential modern slavery victims as between 10,000 and 13,000, based on data from 2013 (Home Office, 2019: 4).

To date, the most cited prevalence estimates on modern slavery remain those produced by the ILO, most recently in conjunction with the IOM and the Walk Free Foundation. However, it is important to highlight the complexity of measuring modern slavery, human trafficking and forced labour. For example, the Walk Free Foundation has also been producing the Global Slavery Index since 2014, with different methodologies between 2014 and 2016. In 2014 it relied on household surveys, while in 2016 it was based on the Gallup World Poll in 25 countries (Walk Free Foundation, 2018; Gallagher, 2017). In particular, the indicators used to determine a country’s vulnerability score, such as tuberculosis, access to clean water and weapons access, have been criticised as not being relevant to modern slavery (Gallagher, 2017). Similarly, the ILO produced global prevalence estimates of forced labour in 2009, 2012 and 2016 but with different methodologies, data sources, scope and regional groupings each time, making it impossible to compare the estimates directly across the three time-points (ILO, 2012b; ILO, 2009; ILO, 2017a).

The ILO’s detailed guidelines on measuring forced labour using a wide range of indicators are helpful despite the limitations to the indicators and the outcome. For example, exploitation is recognised to be a continuum of experiences, with what the ILO terms as ‘decent work’ at one end and ‘forced labour’ at the other. In practice, however, there is little consensus on when exploitation becomes forced labour (ILO, 2015; Andrees, 2008; Skrivankova, 2010; O’Connell-Davidson, 2010). Despite these challenges, data availability and quality on human trafficking globally have been improving over the past decade with every iteration of a new report including data from more countries (UNODC, 2018a) and improved methods for estimating modern slavery (ILO, 2017b).

The ILO’s indicators can be useful for comparing studies with different methodologies, provided the authors report their findings on specific exploitative experiences, in addition to the binary (yes/no) findings (Mak et al., 2017). For example, in a study assessing forced labour experiences among returnee labour migrants in Nepal, the authors were able to compare their findings with those from a study of forced labour in the Malaysian electronic sector, where many Nepalis worked, as both studies reported proportions across indicators (Mak et al., 2017; Verité, 2014).

Migration and migrants in all their forms have become increasingly politicised and destination countries with economies dependent on migrant labourers, including those that are irregular, may be less interested in collecting and reporting such data. The core issues of collecting data on the topic remain problematic, and some forms of modern slavery are less likely to be identified, such as organ trafficking compared to sexual exploitation and increasingly labour exploitation.

Which are the main trafficking routes to the UK?

Globally, all countries are affected by human trafficking, whether as source, transit or destination countries. Many European nations, including the UK, are simultaneously source, transit and destination countries for human trafficking and modern slavery. There is limited published data on the routes used to get to the UK. However, it was noted that Romania was the main source country for four years running (2010 – 2014), and data from 2014 suggests that labour exploitation surpassed sexual exploitation among Romanian nationals (National Crime Agency, 2015: 1). Despite the free movement of European citizens to the UK, 82% of the victims exploited for labour were EU nationals with legal status to live and work in the UK (National Crime Agency, 2015: 1). Potential victims reportedly travelled by air on various budget airlines, or by road using international buses or coaches. Among those trafficked internally within the UK, private vehicles and licensed taxis were also used. Potential victims from Nigeria were reported to have travelled to the UK via Russia, the United Arab Emirates and other African nations (National Crime Agency, 2015). Earlier reviews conducted by the UK Home Office found that the main source regions were Africa, China and South East Asia, and Eastern Europe. Neighbouring European countries such as France, Belgium and the Netherlands operate as transit countries. Individuals travelled to the UK using all transport modes: aeroplane, boat, train, lorry, bus, van, car or on foot (Marsh et al., 2012). Examples of trafficking routes, based on earlier findings from the Commission of the European Communities in 2001, reported by Marsh et al, are shown in Table 2.

Table 2 – Trafficking routes into the UK (adapted from Marsh et al., 2012)

Source: Commission of the European Communities, 2001: 41, cited in Marsh et al., 2012

Another Home Office report examining the routes by which women were trafficked into the UK for sexual exploitation found that the routes used were often a direct flight or a flight with a connection by train through mainland Europe – see Table 3 (Kelly and Regan, 2000).

Table 3 – Trafficking routes identified among women victims in the UK

Source: Kelly and Regan, 2000

These mixed travel mode routes are still common, with recent Vietnamese migrants describing routes similar to the Russian-Ukrainian route (Beadle and Davison, 2019). The Global Trafficking in Persons Report found that trafficked victims came from Central and South-Eastern Europe and that the UK in particular received large numbers of victims from Latvia and Lithuania. Beyond Europe, sub-Saharan African is the core source region among victims detected in Western and Central Europe (UNODC, 2018a: 53).

Migrant workers are trafficked to the UK to work in the agriculture, construction, food processing and packaging, and domestic service sectors. In more recent years, trafficking victims have come from Vietnam to work in nail bars and cannabis-growing operations, which have been widely reported in the media and detected by the NRM. As a result, the trafficking of Vietnamese nationals, including children, has become a priority issue for the UK government. Their travel routes are reported to begin in Russia, via Belarus and Ukraine, and may transit through Poland, Slovakia, the Czech Republic, Germany and/or France before arriving in the UK (Beadle and Davison, 2019: 12).

Causal issues

This chapter explores what is known from the literature on the nature of the modern slavery phenomenon, the different business models (licit and illicit) that result in the exploitation, and the role of both the private sector and organised crime.

What are the root causes of modern slavery?

The root causes of modern slavery are complex and overlapping. Cho modelled macro-level indicators by testing exposures commonly reported in the literature, drawing on three large-scale datasets, and identified characteristics associated with trafficking in source countries as: low income, few economic opportunities and instability in the local economy (Cho, 2012). Other causes were those that compel individuals to leave their home countries, including political instability, conflict, natural disasters, persecution due to religion, ethnicity/caste, gender or sexuality, the growing demand for cheap labour and products, and restrictive mobility options that limit the opportunities for regular labour migration (Chuang, 2006; Wickramasekera, 2002; Huong et al., 2014; Hynes et al., 2018; Jordan, 2011; LeBaron et al., 2018). Many of these are structural factors that are common in low-income settings or among disadvantaged populations.

Political instability, particularly where it descends into armed conflict, is a driver of modern slavery as the warring parties may each be involved in exploiting civilians in sexual slavery and forced marriage (among women and girls) or forced armed combat (among men and boys), most notably the use of child soldiers in many West African conflicts (UNODC, 2018a; UNODC, 2018b). Similar types of trafficking risks have been reported in the media following major natural disasters, where law enforcement has substantially weakened, homes have been destroyed and services overstretched, and where there has been widespread disruption of the social structure in which people live and work (Gyawali et al., 2016; Jones, 2015; Branigan, 2013; Hepburn, 2017). Displacement caused by conflict or disasters also increases the likelihood that traffickers will take advantage of fleeing civilians (UNODC, 2018a). There have been media reports of forced labour and open slave markets in the context of mass displacement (Graham-Harrison, 2017; Hayden, 2019).

At the individual level, international migrant workers in low-waged work are vulnerable to modern slavery due to their reliance on recruitment brokers, their limited awareness of their rights, and fear of losing their job. This is exacerbated by having little support from destination governments to monitor and prosecute labour violations (IHRB, 2013; Amnesty International, 2011; Auwal, 2010; Environmental Justice Foundation, 2013; Beadle and Davison, 2019; Mak et al., 2017; Verité, 2017). Recruitment practices for labour migrants, particularly the practice of charging excessively high fees that force them into debt, have been linked to subsequent exploitation (Verité, 2012; Amnesty International, 2011; Amnesty International, 2017). Studies have consistently found that debt increases people’s vulnerability to forced labour, for example by compelling them to accept worse terms and conditions of employment until the debts are repaid. In many settings, recruitment agents are neighbours, community members or even relatives, which makes it difficult for victims of trafficking to report and pursue them when they realise they have been deceived (Hynes et al., 2018: 24). Migrants almost always discover discrepancies with their contract, terms and conditions after arriving in the destination country, and returning home is rarely an option by then (Mak et al., 2017; Amnesty International, 2011; Andrees, 2008). While these factors on their own may not cause forced labour, many individuals experience them in conjunction with it, which increases their vulnerability.

Many countries that import migrant workforces also have policies that create extreme power imbalances between workers and their employers, leaving migrant workers vulnerable to exploitation. For example, the immigration status of many migrant workers is linked to an employment contract with a specific employer, with no or extremely limited options to change employers. In some settings, employers may even be permitted to confiscate workers’ passports during their employment. The most well-known of these is the Kafala System that operates in many of the Gulf States. Workers are completely dependent on their employers, known as ‘sponsors’, to obtain and maintain their legal status and are prevented from changing employers or even leaving the country without the employer’s permission (IHRB, 2013; Auwal, 2010; Sonmez et al., 2011). In these settings, practices like confiscating identification documents, typically passports, or not allowing workers, mainly domestic workers, even a single day off in a week, are so normalised that employers do not see them as violating the law. Many employers view the system as a necessity to ensure that migrant workers do not run away, as they are accountable for the workers in their employ. Even outside of the Gulf States, exploitative labour and contract breaches are common, including restrictions on movement, non-payment of wages, threats and violence (McCormack et al., 2015; Verité, 2014).

Anti-immigrant sentiment has become widespread in many destination countries, resulting in hostility towards migrant workers (Marks and Olsen, 2015). This has led to strict migration controls that limit people’s options to move legally, forcing many to travel irregularly and work on arrival without legal status or protection. Corporations are able to outsource their production through complex and non-transparent chains of subcontracting across different jurisdictions, in which exploitative labour practices can remain hidden. There is rarely any government enforcement against corporations when workers’ rights are violated within international supply chains (LeBaron et al., 2018). Studies that compared occupational hazards between migrant and native workforces have found that, in many settings, migrant workers are typically paid lower for the same grade of work and suffer more occupational injuries (Ronda-Pérez et al., 2012; Rosano et al., 2012). Pearson argues that, while individuals often migrate for economic reasons, recruitment agents and employers are able to exploit them because of the lack of law enforcement. This is linked to the difficulty in identifying victims and perpetrators, along with the fact that migrant populations are often stigmatised and both governments and host populations are less likely to offer them assistance (Pearson, 2005: 5).

What are the business models of modern slavery and the role of the private sector in both causing modern slavery and ending it?

Business models contributing to modern slavery include those that rely heavily on overseas production or the importing of migrant workers, or have complex supply chains that lack transparency. In many settings where production and labour are outsourced, labour laws and enforcement may be less strict, or workers are employed through third-party recruiters and suppliers that have an incentive to deceive or coerce workers on the terms and conditions of the job, or to charge excessive recruitment fees (Verité, 2017; LeBaron et al., 2018).

The practice of charging recruitment fees is widely recognised as a contributing factor to exploitation (Verité, 2012; Amnesty International, 2011; Amnesty International, 2017; IHRB, 2013). Given the imbalance in power and information between migrant workers and recruiters, there is a complete lack of transparency in terms of what is being paid for, as migrants are rarely given a receipt and almost never one that provides itemised details (Martin, 2018). The Institute for Human Rights and Business (IHRB) suggests companies should fully cover the costs of recruiting migrant workers, so as to reduce workers’ vulnerability to forced labour (IHRB, 2013). Given the complex global supply chains now common in many industries, companies may be indirectly responsible if they do not monitor whether workers are, in fact, still charged recruitment fees when it is against company policy. Businesses should set clear rules and expectations for the recruitment agencies and brokers they use, and should not engage with any that are involved in exploitative practices. Their requirements must be communicated clearly to all prospective and current business partners, stakeholders and workers (IHRB, 2013: 15).

The garment industry is one of the most widely reported industries with a high incidence of exploitative practices. It involves complex supply chains and outsourcing of production to other companies, often internationally. Poor occupational safety and exploitative labour practices have claimed thousands of lives in major factory fires where workers were locked inside and could not escape. A number of international clothing brands subsequently developed the Migrant Worker Guidelines to improve their own and suppliers’ practices (IHRB, 2013). These guidelines follow the core components of the Dhaka Principles for Migration with Dignity, which were intended to promote respect for the rights of migrant workers from recruitment to eventual return home. They have been endorsed by the Confederation of International Recruitment Agencies and the International Trade Union Confederation (IHRB, 2017). Some of the specific principles include stipulating that fees associated with recruitment must be covered by the suppliers and that no deductions may be subsequently taken as repayment of such fees. Furthermore, deductions are only permitted if required by law or with the workers’ permission, although the latter may lead to coercive practices. The principles also state that workers must retain control of their own documents, which would address another widespread exploitative practice in many sectors, frequently carried out to prevent workers from running away or seeking work elsewhere (Frantz, 2008; Andrees and van der Linden, 2005; Amnesty International, 2011; McCormack et al., 2015).

KnowTheChain is a partnership between Humanity United, the Business & Human Rights Resource Centre, Sustainalytics and Verité, to assess corporate strategies to address forced labour in their supply chain. In 2016 it conducted its first assessment of 60 companies in three high-risk sectors: information and communications technology, food and beverage, and apparel and footwear. It followed up with a second assessment in 2018 where the number of companies assessed nearly doubled, to 119 (KnowTheChain, 2019: 10). Its 2018 assessment found that, overall, few companies addressed exploitative recruitment practices or supported workers in their supply chains to exercise their rights or report grievances. The ability to assert their rights and a formal process to address grievances are arguably the most tangible options for improving workers’ conditions (Marks and Olsen, 2015). KnowTheChain also found that companies were more effective if they directly monitored subcontracting by their suppliers and specified which subcontractors they should use, as it is often lower down the supply chain where exploitative labour practices occur. It is interesting to note that the companies which made progress between the 2016 and 2018 assessments were not limited to those subject to media attention, and included those headquartered in Asian countries where transparency in reporting is uncommon (KnowTheChain, 2019). This suggests a fear of negative publicity may not be the primary driver for some companies. A better understanding of what motivates corporations to act and support their migrant workforce, which may differ across different contexts, is an important step to enhance buy-in for such interventions.

Cases of exploitation that have been uncovered and reported in the mainstream media have prompted action by companies seeking to minimise damage to their brand. In one case, a clothing brand discovered that one of its suppliers used exploitative recruitment practices, which led it to review its own recruitment process. The brand then chose to work with only one recruitment agency in the source country and deployed a senior manager trained in compliance standards to oversee the recruitment process. Another case involving a major brand addressed the routine practice of confiscating workers’ passports. Despite being told by local suppliers that the practice was legal, it independently consulted local law firms in the different countries and found that it was only legal in one country. It then put mechanisms in place to educate suppliers on the national labour law and implemented a monitoring system to assess change. Similar reports were found with other global brands (IHRB, 2013).

The UN Guiding Principles on Business and Human Rights provide an overarching frame­work to support companies to tackle human rights abuses within their business, including the development of toolkits to help businesses identify human trafficking down the supply chain. One example is End Human Trafficking Now, which uses the US government’s ‘Trafficking in Persons’ and ‘List of Goods Produced by Child Labor or Forced Labor’ reports to help companies assess their own policies and practice. The US government has issued an executive order, ‘Strengthening Prevention of Human Trafficking’, that prohibits federal contractors, subcontractors and their employees from engaging in exploitative practices, such as confiscating identity documents. These measures tackle specific practices including fraudulent recruitment practices, charging of recruitment fees and confiscating employees’ identification documents. However, to date its effectiveness has not been assessed (IHRB, 2013).

As an alternative approach, companies have been encouraged to conduct trials to assess whether ethically recruited workers are more productive than those recruited through coercion or debt bondage. Indicators such as turnover rates, sickness rates and training costs can be used to assess the productivity benefits of ethical recruitment. If improvements in worker productivity and reductions in absenteeism could be demonstrated, companies might be encouraged to adopt more ethical practices. However, in settings where exploitative practices are legal and widespread, global brands may still need to use their name and buying power to leverage and encourage better practices with local suppliers (IHRB, 2013: 12).

What is the role of organised crime in modern slavery?

Modern slavery across its various forms may be associated with organised crime in various ways. Both the Palermo Protocol and the UN Smuggling Protocol are nested in the UN Convention Against Transnational Organized Crime (UNODC, 2000). Many sectors that benefit from exploitative labour practices, human trafficking and other types of modern slavery, such as brothels or cannabis operations, are illegal in many countries and are often run or controlled by criminal operations. Others, such as car wash operations and nail bars, use trafficked victims to run otherwise legitimate businesses that operate openly. The recruitment, transport (particularly across international borders or using multiple transportation modes), confinement, control and exploitation of individuals are likely to be connected to organised criminal networks at some point along the way (Aronowitz et al., 2010).

However, evidencing connections between modern slavery and organised criminal networks is difficult, given the hidden nature of both phenomena. The criminal enterprises involved are likely to be loose and fluid networks of individuals operating at different points in the trafficking chain, rather than hierarchical organisations. Little is known about the nature of the criminal enterprises, as it is usually only those lower down the criminal network chain that are arrested. It has also been suggested that corruption may be linked to trafficking (Zhang and Pineda, 2008). But corruption as a concept differs across cultures and social contexts and may not be recognised as such in all settings (Richards, 2004). Aronowitz and colleagues adopted a broader interpretation of ‘crime’ and described a range of criminal activities that can occur during the trafficking process, specifically during recruitment, transportation, exploitation, victim disposal and managing criminal proceeds. During recruitment, these could include false promises, kidnapping and forging documents, either directly or through corrupt officials. During the transportation stage, individuals may be assaulted physically or sexually, and deprived of their freedoms, while traffickers may engage in bribing officials. In the exploitation stage, crimes may include coercion, threats, extortion and forced participation in criminal activities. Victims may then be disposed of through sale to another trafficker, abandonment or even murder. Finally, trafficking involves money laundering and tax evasion, possibly with the assistance of corrupt officials. These types of crimes demonstrate that criminal networks have become increasingly sophisticated, diverse and organised, with the ability to smuggle victims through multiple transit countries (Aronowitz et al., 2010).

The lack of coordination between government departments to arrest and prosecute the criminal gangs makes the trading of humans a lucrative market where the risks of detection and arrest are comparatively low (Nadeau, 2018). There is a general disconnect between immigration authorities and police, which limits opportunities to recognise some of these activities as linked to organised criminal networks. When authorities make arrests, it is often the victims who are arrested and charged, as authorities fail to recognise them as potential victims of trafficking due to their involvement in illegal activities at the time of the arrest. These victims are often undocumented, face language barriers and are generally fearful of authorities, as highlighted among Vietnamese victims of trafficking in the UK (Beadle and Davison, 2019). Nevertheless, some success has been documented. Most notably, in 2019 the UK successfully prosecuted a Polish human trafficking ring operating in England after over 88 victims came forward. This case was considered to be the largest modern slavery prosecution in the UK and possibly in Europe (CPS, 2019).

What works in tackling modern slavery?

This chapter explores the state of the evidence base on how to tackle modern slavery. It explores the theories of change behind common interventions and the emerging evidence on effective programming. It then offers some reflections on the strength of the evidence base.

What are common theories of change for ending modern slavery?

Among the many interventions attempted to tackle modern slavery, few have articulated an associated theory of change (LeBaron et al., 2018; Cyrus and Vogel, 2017). A theory of change is important to depict the intervention rationale and logic, ideally through a problem analysis, and to explain how the activities have been selected, through which pathways they are meant to lead to the desired outcome(s) and whether the outcome(s) are specific and measurable. Additionally, theories of change should include contextual information to understand how the proposed intervention should be delivered, by whom and through what avenues. In part, the absence of strong theories of change is also due to a lack of rigorous investigation into whether activities proposed have been found to be effective in other settings. Interventions without a plausible theory of change may deliver their activities successfully without having a demonstrable impact on the underlying issue (LeBaron et al., 2018).

While in the UK ‘modern slavery’ is an umbrella term that captures a range of phenomena, in practice these take different forms in different countries and industries. Any theory of change therefore needs to be specific to the particular manifestation of modern slavery that is being addressed, based on a detailed understanding of the dynamics of each illicit labour market in the specific context.

The ILO described four elements to achieve the SDG to end forced labour through implementing the 2014 Forced Labour Protocol: prevention, protection, remedies and enforcement. Prevention efforts aim to address the factors that put people at risk and to raise awareness among the general public and employers, improve recruitment practices, work with businesses and address risk factors by focusing on the informal economy. Protection involves improving identification of those already in forced labour situations. Remedies support access to justice and compensation for victims. Enforcement is achieved by strengthening the labour and justice system to improve detection and prosecution rates (ILO, 2018).

The literature provides some commentary on the plausibility of the implicit theory of change underlying a number of common intervention types.

One of the most common intervention types for preventing modern slavery is awareness raising, with the assumption that improved knowledge and awareness will lead individuals to avoid exposing themselves to situations where they are at risk of exploitation. However, awareness raising has rarely been articulated into a plausible theory of change linking increased knowledge to behaviour change. Across public health, in areas such as risky sexual behaviours, poor diet, lack of physical activity and substance misuse, it has been demonstrated that knowledge and information are insufficient to change people’s behaviours. Additionally, in Davy’s assessment of 49 evaluations of anti-trafficking interventions, evaluation of awareness-raising activities was linked to assessing short-term recall and does not capture long-term changes in behaviour. There is a need to include longer-term evaluations of these types of intervention among the target population to understand impact beyond short durations of intervention activities (Davy, 2016).

Some interventions are designed to reduce individuals’ need to migrate in the first place by improving livelihood options and offering microcredit schemes (Marshall, 2011: 5). If successful, this would reduce their risk of exploitation. But in settings where the local earnings are several times lower than the potential economic gain from migrating, it would be hard to convince individuals not to try working elsewhere. In any case, the evidence suggests that labour migration increases with income for low-income countries up to a certain level, before it begins to decline (De Haas, 2007; Clemens, 2014). Therefore, in low-income contexts, economic development programming is unlikely to reduce irregular migration or vulnerability to modern slavery except over a long time span.

In response to reports of migrant domestic workers’ extreme exploitation in destination countries, including extreme physical and sexual violence, a number of Asian countries restricted women’s migration options, for example by introducing bans preventing women under a certain age from migrating as domestic workers or migrating to specific destinations. However, the evidence suggests that restricting regular labour migration options has the perverse effect of pushing women to migrate through irregular channels, increasing their risks at every stage of the process (ILO, 2015; Richards, 2004). Policymakers need to consider the potential unintended consequences of policies and interventions designed to promote safe migration.

It has been argued that a rights-based and victim-centred approach that reduces vulnerability to exploitation would better contribute to reducing or ending trafficking and modern slavery (Hynes et al., 2018). Ruhs suggests a core rights approach to protecting labour migrants by concentrating on selected basic rights, including keeping control of one’s own identity documents, having equal access to healthcare and to the legal system for protection, and equal employment conditions (Ruhs, 2017). It may also include the ability to unionise and engage in collective bargaining, as a common form of exploitative practice is to isolate workers and limit their freedoms of movement and association, which impacts on their ability to seek support when needed (Marks and Olsen, 2015; IHRB, 2013). Enabling workers to unionise might empower them to advocate for changes to employment conditions that reduce their risk of exploitation (Pearson, 2005). As highlighted previously, modern slavery differs across settings, sectors and population groups (adults vs children, male vs females), and any strategy to address modern slavery needs to be adapted to the local context. There is no evaluation evidence as yet as to whether rights-based approaches are effective.

One plausible theory of change emerging from the literature is that modern slavery can be reduced by increasing access to safe and well-managed regular migration. A previous ICAI report on UK efforts to tackle irregular migration in the central Mediterranean noted that the Department for International Development (DFID) had a ‘migration theory of change’ (unpublished), which advocates this as a potential solution to trafficking, smuggling, and the humanitarian crisis linked to unsafe sea crossings. There is at present no appetite among European countries to increase the options for regular labour migration into Europe. However, DFID was reportedly exploring options for promoting regional migration options within Africa (ICAI, 2017: 14).

What works in tackling modern slavery?

The evaluation evidence is scarce, but the literature provides some guidance on promising approaches. Local and international organisations have invested significant financial and human resources in programmes aimed at broad goals such as to ‘prevent human trafficking’ or promote ‘safe migration’. Despite this, there is extremely limited evidence on effective interventions, whether these address sexual or labour exploitation (van der Laan et al., 2011; Bryant and Landman, 2020). In a review of human trafficking intervention evaluations, Bryant and Landman found that, although the number has been increasing, it remains low and many evaluations are kept internal by the commissioning organisations. The quality of the available body of evaluations is also questionable – most are process rather than impact evaluations, and rely on qualitative assessments and case studies, with limited efforts to triangulate the data or test the reliability of sources. Most are inconclusive on the question of impact. The lack of explicit theories of change undermines the evaluability of modern slavery programmes (Bryant and Landman, 2020). Complex interventions that address social issues among a mobile population are challenging to design and evaluate, particularly as many interventions include multiple activities. Definitional problems, the hidden nature of the modern slavery phenomenon and difficulties in measurement make it very difficult in practice to assess the impact of any intervention on the incidence of modern slavery (Marshall, 2011; Dottridge, 2014; Bryant and Landman, 2020).

One long-standing approach to preventing and protecting against human trafficking and modern slavery is awareness campaigns. These can be delivered through print media, such as billboards and posters, or mass media, such as television and radio, and may also engage celebrities. According to the ILO, this is the most common prevention strategy, used in 154 countries (ILO, 2018: 5). To reach a younger population, which is traditionally a high-risk group for sexual trafficking from Eastern and Central Europe, some campaigns have been delivered in schools (Nieuwenhuys and Pécoud, 2007).

Awareness-raising campaigns have also been used to provide information for aspiring migrants, including on the need to have the right documentation and use reputable recruitment agencies, or provide links to advice or support services such as hotlines. They advocate for safe, regular migration by highlighting the dangers of irregular migration and the risk of being trafficked or exploited in the process. Critiques of this approach argue that such campaigns are based on the idea that aspiring migrants lack information on legal migration options, or on the risks they face during the journey or at their destination which, were they to know beforehand, would lead them to take different actions or not embark on the journey. As a result, only negative experiences involving human trafficking and exploitation are publicised in such campaigns, frequently featuring rescued victims of trafficking or those who managed to escape. Nieuwenhuys and Pécoud argue that awareness campaigns reflect donor countries’ inability to control their own borders and do not recognise the structural and contextual factors that drive individual choices (Nieuwenhuys and Pécoud, 2007). They further assert that these campaigns are a result of “a confused situation in which security and humanitarian considerations, undocumented migration and human trafficking, national and human securities, immigration and asylum, and criminal and victim statuses get inextricably mixed” (Nieuwenhuys and Pécoud, 2007: 1692).

Robust evaluations are also rare among awareness-raising programmes that target the general public and training programmes for professionals that may come into contact with potential victims. Van Dyke asserts that data from these interventions tends to be collected for monitoring purposes, examining whether the funded activities have been conducted and the numbers reached, rather than whether they led to any change in behaviours or attitudes (Van Dyke, 2017). Additionally, the effectiveness of anti-trafficking interventions more generally is rarely sufficiently evaluated. Davy assessed the evaluations of 49 anti-trafficking interventions implemented at the national, regional and international levels, and found that aims and objectives were often vague and unmeasurable, the evaluation approach lacked rigour, and greater emphasis was placed on process evaluations with insufficient information reported for the reader to assess the quality of the evaluation (Davy, 2016).

In a randomised controlled trial on the effects of a mass media awareness campaign on human trafficking in Nepal, the authors evaluated different types of messaging (positive versus negative), in different formats (poster, graphic novel, radio drama, or audio-visual which combines the novel and radio), delivered to either individuals (all three formats) or groups at the village level. Individuals were exposed to all three formats of messaging, while village groups were only exposed to the audio-visual format. The authors collected data at two time-points, approximately nine months apart, and found that the campaign had increased knowledge about the different forms of human trafficking, including the fact that men can be trafficked. Participants in the trial were able to recognise those who have been trafficked, including themselves, but this did not lead to an increased awareness of the prevalence of human trafficking in their own communities, suggesting they may feel their own communities were somehow less affected than other communities. The evaluation also identified that knowledge increased more through individual rather than group exposure, with the conversations and discussions in groups having sometimes led to misunderstandings. However, group exposures were more effective when assessing support for anti-trafficking policies (Archer et al., 2016). This suggests there is a need to spend time refining and tailoring message content, the format of delivery and the target sample before rolling out any campaign.

Cyrus and Vogel (2017), assessed supply-side campaigns (that is, campaigns designed to influence the choices made by potential victims) and similarly reported that little is known about their effectiveness as they are rarely evaluated. They argue that awareness-raising activities that do not lead to action are ineffective and that the goals of such campaigns should be to reduce exploitation rather than the vague aim of reducing human trafficking. They also found that there was often a lack of understanding of how campaigns work, and that campaigns rarely take advantage of lessons learned through previous campaigns.

Furthermore, the work of both Archer et al. and Cyrus and Vogel found unintended consequences through misunderstood messages. In the case of Nepal, it led to participants becoming more sympathetic towards victims of sex trafficking but an increase in the blaming of victims of labour trafficking (Archer et al., 2016). Cyrus and Vogel found that poster campaigns were misunderstood by members of the public, with a quarter of callers thinking the posters were advertising phone sex (Cyrus and Vogel, 2017).

Demand-side interventions, designed to establish better recruitment practices and protection mechanisms through employment regulation, the banning of recruitment fees and the confiscation of identification documents, were relatively common, but evaluation data was again rarely available (Holden, 2008; IOM, 2005; ILO, 2006; USAID, 2009; IHRB, 2013). In 2014, the ILO launched its Fair Recruitment Initiative, a multi-stakeholder initiative that aimed to address exploitative recruitment practices through the promotion of laws and policies, including bilateral agreements between source and destination countries; government-to-government recruitment schemes; as well as through the empowerment of workers by helping them to find reputable recruitment agents. Allowing workers to unionise and take part in collective bargaining may also decrease the demand for human trafficking and modern slavery. According to the ILO, few countries have addressed the governance of labour recruitment comprehensively (ILO, 2018).

Victim services for those rescued or otherwise identified as having been trafficked may include temporary shelter, medical and psychosocial support, repatriation and reintegration assistance. In some cases, the provision of small grants may help with resettling and legal assistance if prosecution is an option. In a study assessing the work of anti-trafficking organisations worldwide, Limoncelli found that the vast majority were engaged in awareness raising, followed by policy advocacy, with far fewer providing direct services (Limoncelli, 2016). Services offered must be victim-centred with the aim to support their recovery, and not negatively affect their agency to improve their situation, such as shelters that are run like detention centres (Gallagher and Pearson, 2010).

Some intervention efforts have targeted law enforcement officers’ ability to recognise potential trafficking victims, through training and awareness raising, and can then connect victims to appropriate services (Ageros and Pathilath, 2009; Chemonics International, 2009; Datta et al., 2009; Thomas, 2011; IOM/Dhaka, 2008; Gopalan and Livingston; Gera, 2011). The conduct of enforcement officers when facing potential victims of human trafficking is likely to affect both the identification and the recovery of victims. In one evaluation, the author concluded that the intervention “…resulted in a shift in attitudes among officials from denial towards a broad acknowledgement and a sense of urgency to act on the critical problem of irregular migration and its impact on development” (Gera, 2011: 29). Although not formally evaluated, another community-based intervention that trains and supports transport workers to form vigilance groups and work with the local police to intercept potential victims of trafficking was given a favourable review by USAID (Chemonics International, 2009).

Interventions delivered through local community groups in source countries may be able to reduce the incidence of trafficking while at the same time improve the identification of victims. In Nepal, interventions that are embedded within a community, organise social and livelihood activities led by trafficking survivors who are committed to the cause, and are organised through a well-known and respected agency are more likely to be effective in addressing some of the root causes of trafficking (Baumann and Dharel, 2014). Such groups have also been successful in lobbying with the local government to address the issue and provide services (Dharel et al., 2015), while trade unions have been effective in obtaining legal aid or compensation packages for trafficked victims (Dasgupta, 2011; Baumann and Dharel, 2014).

Another common type of intervention is the introduction of laws and regulations. Some destination countries have tightened their labour regulations to prevent exploitative practices, but enforcement (including labour inspection) has been described as non-existent or ineffective. There are few avenues for migrant workers (especially those without legal status) to lodge complaints and, even when they do so, they are rarely supported. These factors, along with the continued supply of workers willing to migrate for work mean that employers are readily able to circumvent labour market rules.

Efforts to tighten criminal legislation have become more widespread since the introduction of the United Nations Trafficking in Persons Protocol in 2003. Between 2003 and 2018, the number of countries with criminal laws that covered ‘most/all forms’ of trafficking outlined in the Protocol has steadily increased from 18% in 2003 to 93% in 2018. This is also reflected in the increased conviction rates observed between 2007 and 2017 (UNODC, 2018a: 46). While the situation has improved, detection and prosecution rates remain low, particularly in middle-income destination countries.

Policies and bilateral agreements between key source and destination countries have been negotiated in an attempt to prevent forced labour, trafficking and modern slavery. There is some anecdotal evidence to suggest that such laws and policies may potentially be effective. An example in Singapore showed that the reported cases of abuse of migrant domestic workers declined between 1997 and 2008 (Ministry of Manpower, 2010). In a Memorandum of Understanding (MoU) between the Malaysian and Nepali governments, a number of important measures were included which addressed specific exploitative practices that Nepali workers who had worked in Malaysia encountered (Mandal, 2018). However, past analysis of other similar agreements has shown that enforcement is problematic and effectiveness therefore limited (IOM Kuwait, 2016; ILO, 2017c, 2018; Amnesty International, 2017). Similarly, the UK and Vietnamese governments signed an MoU on human trafficking with the aim of cooperating and sharing intelligence to protect potential victims of trafficking. The details of the MoU have not been made public and its effectiveness cannot therefore be assessed (Huong et al., 2014).

The United States passed anti-trafficking legislation, the Trafficking Victims Protection Act, in 2000 and the linked President’s Interagency Task Force to Monitor and Combat Trafficking in Person, to coordinate the US government’s response to human trafficking (US State Department, 2000). The US State Department also publishes an annual Trafficking in Persons Report that classifies countries according to their perceived level of effort to address human trafficking and meet the minimum standards outlined in the Trafficking Victims Protection Act of 2000. Countries that do not meet the minimum standards and are not making significant efforts to do so are classified as Tier 3 and are subjected to funding restrictions from the US government (Department of State, 2019).

Organisations such as Amnesty International have argued that improved regulation of prostitution, including decriminalisation, might lead to a reduced incidence of modern slavery and better protection of victims (Amnesty International, 2016; Lee, 2018). However, this is not clear-cut, as several studies have found that some interventions to increase the regulation of prostitution have the unintended consequence of increasing the likelihood of trafficking, as potential victims become more reliant on traffickers or suffer an increase in indebtedness. Laws that limit the options for legal prostitution may increase the demand for trafficked sex workers. Akee and colleagues examined the impact of legal interventions on the demand side of trafficking for sexual exploitation. Specifically, these were policies that addressed victim protection through amnesty in destination countries and laws against prostitution in both source and destination countries. They found that increased legal enforcement in source and destination countries can also increase cross-border trafficking by raising the price and profitability (Akee et al., 2011: 29).

As mentioned previously, interventions need to be designed around a detailed understanding of the specific features of the modern slavery challenge being addressed. The UNODC argues that multidimensional interventions involving a cross-section of stakeholders such as NGOs, government departments, trafficking survivors or returnee migrants are likely to be more effective than those focusing on a single group. Peer approaches may also be promising. However, interventions need to be combined with a rigorous evaluation plan to assess their effectiveness and whether they can be maintained over time, as well as highlighting any unintended consequences (UNODC, 2018a).

What are the knowledge and evidence gaps?

Both the ILO and IOM have identified data gaps that need to be filled before we can monitor progress on SDG 8.7 or the Global Compact for Migration, an intergovernmental agreement adopted in 2018 to promote a comprehensive approach to promoting safe and orderly migration. The ILO has called on countries to provide more detailed data on modern slavery, including the number of trafficking cases disaggregated by sex, age and type of exploitation (UNODC, 2018a). The IOM identified five themes and sets of data needed, including:

• Strengthen regular migration pathways through data on skill shortages, demographic trends, migration flows and recruitment costs.
• Reduce vulnerabilities associated with migration using data on displacement, trafficking, exploitation, migrant fatalities, forced and assisted return and unaccompanied minors.
• Promote migrant integration with data on migrants’ demographic characteristics, skills mismatch, experience of discrimination.
• Improve migration management with data on irregular migration flows, smuggling, return and reintegration as well as administrative data on border control, visa processing time and asylum claims.
• Enhance the development effects of migration using remittance data, diaspora profiles and migration-related development assistance (IOM, 2018: 83).

Although these were drawn up with the IOM’s Global Compact for Migration in mind, these types of data would also enable better assessment of the effectiveness of anti-trafficking interventions.

In addition, evidence on the demand side of human trafficking is needed. This should aim to capture the perspectives of multiple groups of stakeholders: those who employ trafficked victims, those who purchase the products or services produced, and those involved at various stages of the process (Pearson, 2005). Such research should examine labour protection rules and regulations because of their important potential in the prevention or reduction of human trafficking and forced labour (Pearson, 2005). Additionally, the views of recruitment agents or brokers are also needed, especially as some may not be aware that the people they recruit would end up in forced labour and modern slavery conditions.

There is a need for more research and evaluation on how to address the multiple layers of demand, from employers to consumers as well as the wider environmental drivers that contribute and shape such exploitations, focusing therefore on destination countries. Apart from sexual exploitation, where interventions to address demand have been attempted, for forced labour, few consumers know how the supply chains of the many products they purchase operate and would not be able to ascertain whether the products were produced by exploited labour. Pearson argues that research and interventions must encompass those “companies and individuals further up the supply chain that recklessly disregard whether trafficking or the use of forced labour is occurring in the production of their goods” (Pearson, 2005: 4). Chuang maintains that as long as individuals are compelled to leave their homes to work elsewhere, no intervention can eradicate human trafficking (Chuang, 2006: 137). Indeed, migrant workers who supply the market may themselves contribute to generating the demand for low-wage work (Anderson and O’Connell-Davidson, 2003).

Another under-researched aspect of forced labour, trafficking and modern slavery is the experiences of victims in destination countries. Research conducted while victims are at destination would provide a more complete picture of their situation than retrospective accounts, which are prone to recall bias. Furthermore, some labour sectors, such as domestic work, fishing, manufacturing and construction, are particularly vulnerable to modern slavery as they are typically hidden from the general public. As a result, collecting data is often impossible without putting workers at risk of harm, which violates ethical research procedures. Migrants in destination countries often have precarious legal status and may be unwilling to take part in research due to fear of traffickers, or of arrest and deportation (Beadle and Davison, 2019: 18). The diaspora community of source countries could be used to support the development of community-based interventions that can support and assist victims of trafficking. It is also important to recognise that the diaspora community and current or past migrants themselves may be involved in the trafficking of their compatriots, and to ensure that this is considered in any community-based intervention (Beadle and Davison, 2019: 11).

There is also a lack of research into the reintegration process of potential victims of trafficking, including assessing the types of support needed and available in both the short and the long term. This is particularly true among female victims of trafficking in conservative societies, which may view any migration as synonymous with sexual exploitation, leading to stigmatisation. It is unclear how accepting their own families may be upon their return home. Longer-term effects of shelter and residential programmes should also be evaluated as they are often offered by NGOs as temporary measures, either when individuals first return to their home country, while taking legal action against recruitment agencies, or until they are ready to return to their home village. Victims of trafficking who are minors need special attention and authorities need to improve their protection mechanisms. In the UK, large numbers of Vietnamese children who were taken into care subsequently went missing (Beadle and Davison, 2019: 10). They may then be at risk of being re-trafficked, possibly even by the same traffickers.

There are also geographic and regional gaps in data availability. In sub-Saharan Africa and the Middle East, sparse and unsystematic data limits our understanding of the specific regional and corridor patterns. Countries could benefit from capacity building so that the responsible officials better appreciate the need for good data collection approaches and are supported in collecting and maintaining such data (UNODC, 2018a: 7). Furthermore, there is a lack of data on domestic or internal trafficking, as authorities, NGOs and community groups may be less aware of instances when potential victims do not cross an international border.

Echoing other researchers’ insistence on the need for victim-centred approaches, in Ali’s analysis of different sources of data on human trafficking, he argues that:

“There is a need to harmonise policies through shifting from criminal justice systems to a victim-centred approach. This approach must go into the framing of anti-trafficking movement where the main goal will be based on rediscovering the ‘trafficking victims’ where emphasis is placed on victims, victimisation and social reactions to the victims. It is vital to generate proper knowledge based on overarching social correlates of victimisation in human trafficking and its impact on society” (Ali, 2010: 65).

The impact of public opinion of migrant workers, who are often stigmatised by the host population, should be assessed. The ILO includes the general public as one of the target populations in awareness-raising activities (ILO, 2018). Improving awareness of forced labour and human trafficking in the general population could shift public opinion and lead to greater empathy between the host and migrant populations (Pearson, 2005). However, such a task is itself challenging given the current anti-migration climate across many Western societies, including Europe, Australia and the United States.

Finally, the way in which trafficking research is conducted needs to be more robust. Surtees and Johnson recommend basing each study on strong conceptual foundations, informed by literature review, and being fully transparent about the data collection and analysis process, including ethical issues. Where possible, they suggest exploring ways to collect longitudinal data, capturing the lifetime experience of victims (Surtees and Johnson, 2019: 25). When conducting intervention evaluations, independent evaluators should be used and their relationship with the organisations and programmes should be reported for greater transparency (Bryant and Landman, 2020; Hames et al., 2010).

Victim/survivor issues

This section explores the evidence on modern slavery in the literature from the perspective of victims and survivors. It looks at the types of individuals who are vulnerable to modern slavery, and at who the perpetrators are. It assesses the costs to victims and society at large, and considers how other identity categories, such as age, gender and ethnicity, relate to modern slavery.

Which types of individuals are more likely to be victims and perpetrators?

According to the IOM’s Counter Trafficking Data Collaborative (CTDC, 2020) data, three-quarters of all victims identified were adults. Child victims were commonly recruited by their family and relatives and were trafficked into begging, domestic work, peddling and petty crime. The highest proportion of adult female victims identified were in the 18-20 age group, while men were older at 30-38, and were most commonly exploited in construction, domestic work and agriculture. Among both adult and child victims, the most common forms of control were psychological abuses, restrictions in movements, threats and deception. Children were more often controlled with psychoactive substances, while adults reported more cases of confiscation of earnings (CTDC, 2020).

According to the Global Report on Trafficking in Person, most victims identified globally were adult women, accounting for nearly half of all those identified (UNODC, 2018a: 10). Girls and adult men represented one-quarter and one-fifth respectively, followed by boys. Sexual exploitation was the most commonly detected type and represented 83% of the cases experienced by women. Among men this was reversed, with 82% reporting having been trafficked for forced labour, and 10% for sexual exploitation (UNODC, 2018a: 10). There were also geographic differences, with relatively

more boys identified in Africa than the global average, while South Asia and the Middle East identified more adult men (UNODC, 2018a:10).

Individuals who have experienced forced labour worked in a variety of sectors. Men typically worked in the fishing, manufacturing and construction sectors, while women worked in the domestic and garment sectors or in the sex industry (Pocock et al., 2016; Mak et al., 2017; ILO, 2017a). In an analysis of individuals in post-trafficking services in the Mekong region, no single profile of a trafficked person could be found as the range of demographic characteristics, migration history and experiences of exploitation varied widely (Kiss et al., 2015).

Studies that have included recruiters and employers as samples are prone to bias as those who agree to be interviewed are likely to be very different from those who do not, or would not let researchers into their homes (for domestic workers), factories or dormitories where workers reside. Workers that require their employers’ agreement or approval to take part are unlikely to be honest about their situation as the potential for repercussion is real. Thus, the possibility of generalising the findings to a wider sample of employers or workers should be cautioned. Nevertheless, such perspectives are necessary to understand both the demand and the supply side of modern slavery, and may lead to sector-wide improvements.

Men were the main suspected perpetrators, including those who were investigated, arrested, prosecuted and convicted, although one-third were women according to data from 2016. There were also geographic differences in terms of where perpetrators were convicted, with relatively more women convicted in Eastern Europe and Central Asia compared to other regions, while relatively more men were convicted in Central America and the Caribbean. The majority of convicted traffickers in Western and Southern Europe were regional nationals (60%) rather than nationals of the country in which they were convicted (UNODC, 2018a: 37).

Restricting the CTDC data to the UK only, there were more men identified as victims compared to women in 2016 (UNODC, 2018a: 51), while longer-term data indicates that women still predominate, representing 61% of those identified as having been trafficked. Victims in the UK came from Slovakia (33%), Romania (22%) and Ukraine (13%). The most common type of exploitation was labour (56%), followed by sexual (35%) and others (9%) (CTDC, 2020).

Using the 17 typologies of modern slavery developed by the Home Office for the National Referral Mechanism, Cooper and colleagues developed a profile of the victims and perpetrators for each typology, including information on how the victims were recruited, transported and exploited. Given the range of typologies, the profiles of both victims and perpetrators also varied, as did the specifics of their exploitation and control method (Cooper et al., 2017). This reinforces the idea that there is no set profile of victims or perpetrators. Similarly, Pearson cautions against the continued focus on victim profiling that emphasises their personal background characteristics and argues instead that it would be more useful to understand the recruitment, working and living conditions, including laws and policies that enable trafficking and forced labour to occur (Pearson, 2005: 2). It is important to remember that data from identified victims is extremely limited and cannot be used to generalise across modern slavery victims.

What are the costs of modern slavery to survivors and society at large?

Studies have found that individuals who have experienced modern slavery in its various forms are likely to suffer from a range of mental, physical and social ill-health, including violence, depression, anxiety, post-traumatic stress disorder and suicide attempts (Turner-Moss et al., 2014; Al-Maskari et al., 2011; Such and Salway, 2017). However, there is limited reliable evidence in the literature on the economic value of the harm suffered.

One evidence assessment commissioned by Public Health England, summarising the results of 17 studies, found that the harms caused by modern slavery varied according to the type of exploitation suffered, with physical injury particularly associated with forced labour and sexually transmitted diseases associated with sexual exploitation. Poor mental health was found to be prevalent among all types, particularly post-traumatic stress disorder, stress and anxiety (Such and Salway, 2017).

One study used ‘the economic and social costs of crime’ methodology to assess the total costs per victim of modern slavery in the UK. It used Quality Adjusted Life Years – a standard measurement unit used in public health economic evaluation to measure losses to both the length and the quality of life. It estimated that the duration for victims in domestic servitude ranged up to 24 months, while those in labour and sexual exploitation experienced a shorter median duration of approximately nine months. Costs were estimated for three common types of exploitation: sexual exploitation, labour exploitation and domestic servitude. Costs were divided into those incurred by the victim as a consequence of modern slavery (physical and emotional harm, lost time and output, and health costs) and those incurred by the state (victim services and law enforcement). The findings estimated an average cost per victim at just under £330,000 across the three types of exploitation, with the highest cost for domestic servitude (£390,000). The largest share of these costs came from physical and emotional harm to the survivors, followed by their loss of time and output (Reed et al., 2018: 5).

How do gender and other social characteristics relate to the issue of modern slavery?

There has been a traditional focus in the literature and in development programming on the trafficking of women and girls. This focus is reflected in the name of the Palermo Protocol (The Protocol to Prevent, Suppress and Punish Trafficking in Persons, Especially Women and Children). However, for some categories of modern slavery, such as forced labour, the survivors are more likely to be men (UNODC, 2018a; Cooper et al., 2017). Despite greater understanding of men’s experiences in recent years, there is still an emphasis on women in international programming (Hynes et al., 2018; Pocock et al., 2016). While there is no doubt that women are at a greater risk of sexual exploitation, such emphasis may also do a disservice to women as those who experienced labour exploitation are often assumed to have been sexually exploited. Such perceptions might make it difficult for women to access services and support for fear of further stigma and discrimination, including when they return home.

Similar concerns with stigma affect male victims. In many settings, men are seen by society as less vulnerable, particularly given the traditional association between trafficking and sexual exploitation of women. Research among trafficked men in Eurasia and South-Eastern Europe found that men who were unable to escape from a trafficked situation felt great shame (Surtees, 2008). Such beliefs may result in them being reluctant to be labelled as victims and to access support services. Men may choose not to share their experiences with others to avoid revealing their experiences of exploitation and, in turn, may present a misleading picture to aspiring migrants (Walter et al., 2004; Datta et al., 2008; Dannecker, 2013).

Having characteristics or being affiliated to groups considered undesirable in a given context may increase one’s vulnerability to modern slavery. These factors are likely to be highly context-dependent. In South Asia, for example, certain indigenous groups, such as scheduled castes, often suffer discrimination by other castes. It has also been suggested that those most at risk come from rural areas and live in small communities (Centre for Social Justice, 2015). Data from identified victims in the UK found that some victims were homeless, or had learning difficulties or substance misuse issues when they were recruited (Cooper et al., 2017). A Polish gang described as the biggest modern-day slavery network in the UK targeted homeless people, ex-prisoners and alcoholics from Poland and managed to amass over 400 victims before eventually being caught (BBC News 2019b). Young women and children are often targeted as they are perceived to be weaker and easier to control (Pearson, 2005: 19). However, the specific characteristics that make individuals more or less vulnerable to trafficking and modern slavery are highly context-specific. Therefore, there is a need to explore and assess how different characteristics create or reduce vulnerability in any given target population before embarking on research, programmes and interventions.

Conclusions

Some of the key findings from the literature are as follows.

What is modern slavery?

‘Modern slavery’ has emerged in recent years as an umbrella term for a series of connected but distinct phenomena, including human trafficking and forced labour, and can be for various purposes, including labour, sexual exploitation or forced marriage. There are many typologies in use, including some with definitions under international agreements, but often with constituent elements – in particular, ‘exploitation’ – that are difficult to precisely define. This gives rise to identification and measurement challenges. The use of ‘modern slavery’ as an umbrella term is controversial, with some commentators suggesting that it lacks precision and nuance, given the variety of complex issues it covers.

What is the prevalence of modern slavery?

There are major challenges to obtaining accurate data on the prevalence of modern slavery, given definitional problems and the hidden nature of the phenomenon, with many victims reluctant to speak to authorities or researchers. The available estimates suggest that over 400 million people around the world are in modern slavery, including 25 million in situations of forced labour. All countries, including the UK, are affected by modern slavery, whether as a source, transit or destination country, or a combination. In the UK, the most robust figures, from 2013, suggest there were between 10,000 and 13,000 potential victims of modern slavery. These estimates are likely to be conservative, and other estimates range up to 136,000 victims in the UK (2016). Almost 7,000 potential victims were referred to UK authorities in 2018, of whom 60% were males and 45% were children, with Albania, China and Vietnam the most commonly reported source countries. The trafficking routes to the UK are multiple and often multi-country.

What are the causes of modern slavery?

Modern slavery has complex and overlapping causes. There are risk factors associated with the context in source countries, including political instability, conflict, natural disasters, persecution, and limited options for regular labour migration. Individuals who opt to become migrant workers are at risk of falling into modern slavery due to dependence on recruitment brokers, high levels of debt, lack of awareness of their rights and uncertain legal status, which leaves them vulnerable to prosecution and deportation. They often suffer from severe imbalances of power with employers, leaving them vulnerable to exploitation. Their identification documents are often confiscated. Industries with complex and non-transparent international supply chains, such as the garment industry, are more likely to feature modern slavery, particularly where migrant workers pay substantial fees to recruiting agents. Corruption and organised crime intersect with modern slavery in multiple ways, and are known to be hard to prosecute, given the hidden and often transnational nature of the offences.

What works in tackling modern slavery?

While there have been many interventions that attempt to tackle modern slavery, few have fully articulated a theory of change setting out how the interventions are expected to reduce modern slavery. Some of the most common intervention types – in particular, awareness raising among potential victims to encourage them to adopt less risky behaviour – have a poor record of success. There is a lack of credible evaluation evidence on what works, due in large part to the difficulty of measuring impact. The literature suggests that modern slavery is a complex, multidimensional challenge that requires multi-component interventions, targeting different groups and addressing both the supply of and demand for trafficked individuals. The field is marked by major evidence gaps on the incidence of modern slavery, the characteristics of victims that make them vulnerable, their lifetime experiences (including reintegration), and the drivers of modern slavery in particular countries and industries.

Who are the most vulnerable to modern slavery?

The available evidence suggests most victims are women, targeted for sexual exploitation, but men are more likely to be victims of forced labour. There is a lack of understanding of and focus on other forms of modern slavery, which are likely to be significantly under-reported. The profiles of both victims and perpetrators vary across different types of modern slavery and different industries and contexts. Vulnerable individuals, such as the homeless or people with disabilities, may be particularly at risk. Victims are often reluctant to avoid the stigma associated with being labelled as such, which may make them unwilling to seek assistance. There is limited data on the social and economic costs of modern slavery.

Bibliography

Ageros, B. & Pathilath, B (2009). Terminal Evaluation Report of UNODC R76 Project: Strengthening of the legal and law enforcement institutions to prevent and combat human trafficking in Lao PDR, UNODC, link.

Akee, R. et al. (2011). Transnational Trafficking, Law Enforcement and Victim Protection: A Middleman Trafficker’s Perspective. Discussion Paper Series, Institute for the Study of Labor, link.

Al-Maskari, F. et al. (2011). Prevalence of depression and suicidal behaviors among male migrant workers in United Arab Emirates. Journal of Immigrant Minority Health, 13, pp.1027-1032.

Ali, H. M (2010). Data collection on victims of human trafficking. An analysis of various sources. Journal of Human Security, 6, pp. 55-69.

Amnesty international (2011). Nepal: False promises: Exploitation and forced labour of Nepalese migrant workers, link.

Amnesty International (2016). Amnesty International policy on state obligations to respect, protect and fulfil the human rights of sex workers, link.

Amnesty International (2017). Nepal: Turning people into profits. Abusive recruitment, trafficking and forced labour of Nepali migrant workers, link.

Anderson, B. & O’connell-Davidson, J (2003). Is trafficking in human beings demand driven? A multi-country pilot study, Migration Research Series, link.

Andrees, B (2008). Forced labour and trafficking in Europe: how people are trapped in, live through and come out. Working paper 57.

Andrees, B. & Van Der Linden, M. N. J (2005). Designing trafficking research from a labour market perspective: the ILO experience. In: Laczko, F. & Gozdziak, E. (eds.) Data and Research on Human Trafficking: A global survey. International Organization for Migration.

Archer, D., Boittin, M. & Mo, C. H (2016). Reducing Vulnerability to Human Trafficking. An Experimental Intervention Using Anti-Trafficking Campaigns to Change Knowledge, Attitudes, Beliefs, and Practices in Nepal, Research and Innovation Grants Working Paper Series USAID, link.

Aronowitz, A., Theuermann, G. & Tyurykanova, E (2010). Analysing the business model of trafficking in human beings to better prevent the crime, Organization for Security and Co-operation in Europe, link.

Auwal, M. A (2010). Ending the exploitation of migrant workers in the Gulf. The Fletcher Forum of World Affairs, 34, pp. 87-108.

BBC News (2019a). Essex lorry deaths: 39 bodies found in refrigerated trailer, link.

BBC News (2019b). UK slavery network ‘had 400 victims’ link.

Beadle, D. & Davison, L (2019). Precarious Journeys: Mapping vulnerabilities of victims of trafficking from Vietnam to Europe, link.

Branigan, T (2013). Typhoon Haiyan: Children in Disaster Zone Are Vulnerable, Warns Unicef, link.

Brennan, D (2014). Trafficking, scandal, and abuse of migrant workers in Argentina and the United States. The ANNALS of the American Academy of Political and Social Science, 653, pp. 107-123.

Bryant, K. & Landman, T (2020). Combatting Human Trafficking since Palermo: What Do We Know about What Works? Journal of Human Trafficking, 6, pp. 119-140.

Centre For Social Justice (2015). A modern response to modern slavery, link.

Chemonics International (2009). USAID Anti-Trafficking in Persons Programs in Asia: A Synthesis, link.

Cho, S.-Y (2012). Modelling for determinants of human trafficking: An Empirical Analysis. Economics of Security Working Paper 70, link.

Chuang, J (2006). Beyond a snapshot: preventing human trafficking in the global economy. Indiana Journal of Global Legal Studies, 13, pp. 137-163, link.

Commission of The European Communities (2001). Research based on case studies of victims of trafficking in human beings in 3 EU member states, i.e. Belgium, Italy and the Netherlands, Commission of the European Communities, link.

Cooper, C. et al, (2017). A typology of modern slavery offences in the UK. Research Report 93, Home Office, link.

Council Of Europe (2005). Explanatory Report to the Council of Europe Convention on Action against Trafficking in Human Beings, link.

CPS (2019). CPS secures convictions in largest ever modern slavery prosecution, link.

Craig, G (2017). UK’s Modern Slavery Legislation: An Early Assessment of Progress, Social Inclusion, link.

CTCD (2020). The Counter Trafficking Data Collective, link.

Cyrus, N. & Vogel, D (2017). Learning from Demand-Side Campaigns against Trafficking in Human Beings: Evaluation as Knowledge-Generator and Project-Improver, DemandAT Working Paper No 9, link.

Dannecker, P (2013). Rationalities and images underlying labour migration from Bangladesh to Malaysia, International Migration, 51, pp. 40-60.

Datta, K. et al. (2008). Mobile masculinities: Men, migration and low paid work in London, Queen Mary, University of London, link.

Datta, P. K., Sarker, N. I. & Saha, J. K (2009). External evaluation report on Prevention of Cross-Border Trafficking in Women and Children between Bangladesh and West Bengal, India (C-BAT) Project.

Davy, D (2016). Anti-Human Trafficking Interventions: How Do We Know if They Are Working? American Journal of Evaluation, 37, pp. 1-19.

De Haas, H (2007). Turning the Tide? Why development will not stop migration,  Development and Change, 38(5), pp. 20.

Department of State (2019). Trafficking in Persons Report June 2019, United States Department of State, link.

Dottridge, M (2014). Editorial: How is the money to combat human trafficking spent? Anti-Trafficking Review, 3, link.

Dottridge, M (2017). Global Trafficking Prevalence Data Distorts Efforts to Stop Patterns of Human Trafficking, Anti-Trafficking Review, pp. 161-164, link.

Düvell, F (2011). Paths into irregularity: the legal and political construction of irregular migration, European Journal of Migration and Law, 13, pp. 275-295.

Environmental Justice Foundation (2013). Sold to the sea. Human trafficking in Thailand’s fishing industry, link.

Frantz, E (2008). Of maids and madams. Sri Lankan domestic workers and their employers in Jordan, Critical Asian Studies, 40, pp. 609-638.

Gallagher, A. T (2017). What’s Wrong with the Global Slavery Index? Anti-Trafficking Review, pp. 90-112, link.

Gallagher, A. T. & Pearson, E (2010). The High Cost of Freedom: A Legal and Policy Analysis of Shelter Detention for Victims of Trafficking, Human Rights Quarterly, 32, pp. 73-114.

Gera, W (2011). Regional Programme and Dialogue on Facilitating Safe and Legal Migration from South Asia to the European Union – Final Report, IOM/EU.

Gleason, K. & Cockayne, J (2018). Official Development Assistance and SDG Target 8.7. Measuring aid to address forced labour, modern slavery, human trafficking and child labour, United Nations University Centre for Policy Research, link.

Goodey, J (2008). Human trafficking: Sketchy data and policy responses, Criminology & Criminal Justice, 8, pp. 421-442.

Gopalan, S. & Livingston, L (2008). Terminal evaluation report: Strengthening the law enforcement response in India against trafficking in persons through training and capacity building (Project number IND/S16), link.

Graham-Harrison, E (2017). Migrants from West Africa being ‘sold in Libyan slave markets’, link.

Gyawali, B., Keeling, J. & Kallestrup, P (2016). Human trafficking in Nepal: post-earthquake risk and response, Disaster Medicine and Public Health link.

Hames, C., Dewar, F. & Napier-Moore, R (2010). Feeling good about feeling bad. A Global Review of Evaluation in Anti-Trafficking Initiatives, Global Alliance Against Traffic in Women (GAATW), link.

Hayden, S (2019). Fear and despair engulf refugees in Libya’s ‘market of human beings’, link.

Hepburn, S. (2017). Harvey, Irma, and Maria: Natural Disasters and Human Trafficking, link.

Holden, D (2008). Independent Evaluation report – Midterm Evaluation: ILO Combating Forced Labour and Trafficking of Indonesian Migrant Workers (INS/06/M10/NOR).

Home Office (2019). UK Annual Report on Modern Slavery, link.

Huong, D. et al. (2014). The current situations of Vietnamese returnees from the United Kingdom baseline research report, Center for Training and Community Development.

Hynes, P. et al. (2018). ‘Vulnerability’ to human trafficking: a study of Viet Nam, Albania, Nigeria and the UK, link.

ICAI (2017). The UK’s response to irregular migration in the central Mediterranean: A rapid review, The Independent Commission for Aid Impact, link.

IHRB (2013). Fees and IDs: Tackling recruitment fees and confiscation of workers’ passports, Institute for Human Rights and Business link.

IHRB (2017). Migration with dignity: implementing the Dhaka Principles, link.

ILO (2006). Demand Side of Human Trafficking in Asia: Empirical Findings, link.

ILO (2009). Operational indicators of trafficking in human beings. Results from a Delphi survey implemented by the ILO and the European Commission, link.

ILO (2012a). Hard to see, harder to count. Survey guidelines to estimate forced labour of adults and children, link.

ILO (2012b). ILO Global Estimate of Forced Labour 2012: Results and Methodology, link.

ILO (2013). ILO indicators of forced labour, link.

ILO (2015). ILO expert workshop on measuring modern slavery, link.

ILO (2015) No easy exit – Migration bans affecting women from Nepal, link.

ILO (2017a). Global estimates of modern slavery: forced labour and forced marriage, link.

ILO (2017b). Methodology of the global estimates of modern slavery: forced labour and forced marriage, link.

ILO (2017c). The migrant recruitment industry. Profitability and unethical business practices in Nepal, Paraguay and Kenya, link.

ILO (2018). Ending forced labour by 2030: A review of policies and programmes, link.

Institute for Human Rights and Business (2018). Responsible recruitment: addressing gaps in protections for migrant workers, link.

IOM (2005). Labour migration in Asia: protection of migrant workers, support services and enhancing development benefits, link.

IOM (2011). Labour migration from Colombo Process countries. Good practices, challenges and way forward, link.

IOM (2018). Informing the Implementation of the Global Compact for Migration. Improving migration data to implement the Global Compact for Migration, link.

IOM Kuwait (2016). Research on the labour recruitment industry between United Arab Emirates, Kerala (India) and Nepal, link.

IOM/Dhaka (2008). “Counter-Trafficking Interventions in Prevention, Protection and Prosecution for Victims of Trafficking in Persons in Bangladesh”: End project evaluation report, link.

ITUC. Domestic Workers – 12 + 12, link.

Clemens, M (2014). Does development reduce migration? Institute for the Study of Labour, link.

Jones, S (2015). Nepal earthquakes leave bitter legacy as children become quarry for traffickers, link.

Jordan, A (2011). Slavery, forced labour, debt bondage, and human trafficking: from conceptional confusion to targeted solutions, Issue Paper 2, link.

Kelly, L. & Regan, L (2000). Stopping Traffic: Exploring the extent of, and responses to, trafficking in women for sexual exploitation in the UK. In: Willis, C. F. (ed.). London: Home Office.

Kiss, L. et al. (2015). Health of men, women, and children in post-trafficking services in Cambodia, Thailand, and Vietnam: an observational cross-sectional study, The Lancet, 3, pp. 154-61.

KnowTheChain (2019). Three sectors, three years later: Progress and Gaps in the Fight Against Forced Labor, link.

Lee, S. & Persson, P. 2018. Human Trafficking and Regulating Prostitution, NYU Law and Economics Research Paper No. 12-08.

Lebaron, G. et al. (2018). Confronting root causes: forced labour in global supply chains. Beyond Trafficking and Slavery, OpenDemocracy and the Sheffield Political Economy Research Institute, link.

Limoncelli, S. A (2016). What in the world are anti-trafficking NGOs doing? Findings from a global study, Journal of Human Trafficking, 2, pp. 316-328.

Mahmoud, T. O. & Trebesch, C (2010). The economics of human trafficking and labour migration: micro-evidence from Eastern Europe, Journal of Comparative Economics, 38, pp. 173-188.

Mak, J. et al. (2017). What is the prevalence of and associations with forced labour experiences among male migrants from Dolakha, Nepal? Findings from a cross-sectional study of returnee migrants. BMJ Open, (7).

Makisaka, M (2009). Human Trafficking: A Brief Overview. Social Development Notes. Conflict, Crime and Violence, link.

Mandal, C. K (2018). Nepal and Malaysia sign labour pack, link.

Marks, E. & Olsen, A (2015). The role of trade unions in reducing migrant workers’ vulnerability to forced labour and human trafficking in the Greater Mekong Subregion, Anti-Trafficking Review, pp. 111-128, link.

Marsh, K. et al. (2012). An evidence assessment of the routes of human trafficking into the UK, link.

Marshall, P (2011). Re-thinking trafficking prevention. A guide to applying behaviour theory, link.

Martin, P (2016). Labor Recruitment. Background paper presented at the Abu Dhabi Dialogue Senior Officials’ Thematic Meeting on Bilateral and Multilateral Government Collaboration to Ensure Lawful, Fair, and Transparent Labor Recruitment Practices, Abu Dhabi Dialogue.

Martin, P (2018). Recruitment costs. Data Bulletin Series. Informing the Implementation of the Global Compact for Migration. Part II: Data availability, gaps and challenges on selected topics.

Mccormack, S. et al. (2015). The other migrant crisis. Protecting migrant workers against exploitation in the Middle East and North Africa, A policy research report, link.

Ministry of Manpower (2010). MOM FDW and FDW Employer Study 2010, link.

Nadeau, B. L (2018). ‘Migrants are more profitable than drugs’: how the mafia infiltrated Italy’s asylum system, link.

National Crime Agency (2015). NCA Strategic Assessment. The nature and scale of human trafficking in 2014, link.

National Crime Agency (2019a). National Referral Mechanism Statistics – End of Year Summary 2018, link.

National Crime Agency (2019b). National Referral Mechanism Statistics. Quarter 1 2019 – January to March, link.

Nieuwenhuys, C. & Pécoud, A (2007). Human trafficking, information campaigns, and strategies of migration control, American Behavioral Scientist, 50, pp. 1674-1696.

O’connell-Davidson, J (2010). New slavery, old binaries: Human trafficking and the borders of ‘freedom’, pp. 244-261, link.

OHCHR (1926). Slavery Convention, link.

OHCHR (1956). Supplementary Convention on the Abolition of Slavery, the Slave Trade, and Institutions and Practices Similar to Slavery, link.

OHCHR (2000). Protocol against the Smuggling of Migrants by Land, Sea and Air, Supplementing The United Nations Convention Against Transnational Organized Crime, link.

Pearson, E (2005). The Mekong challenge: human trafficking: redefining demand: destination factors in the trafficking of children and young women in the Mekong sub-region, link.

Piper, N., Segrave, M. & Napier-Moore, R (2015). Editorial: What’s in a name? Distinguishing forced labour, trafficking and slavery, Anti-Trafficking Review, pp. 1-9, link.

Pocock, N. et al. (2016). Labour Trafficking among Men and Boys in the Greater Mekong Subregion: Exploitation, Violence, Occupational Health Risks and Injuries, PLoS One, link.

Reed, S. et al. (2018). The economic and social costs of modern slavery, link.

Richards, K (2004). The Trafficking of Migrant Workers: What are the Links Between Labour Trafficking and Corruption? International Migration, 42, pp. 147-168.

Robinson, C., Branchini, C. & Thame, C (2017). Global Trafficking Prevalence Data Advances the Fight against Trafficking in Persons, Anti-Trafficking Review, pp. 157-160, link.

Ronda-Pérez, E. et al (2012). Differences in working conditions and employment arrangements among migrant and non-migrant workers in Europe, Ethnicity and Health, 17, pp. 563-577.

Rosano, A. et al. (2012). Work-related health problems among resident immigrant workers in Italy and Spain, Italian Journal of Public Health, 9 (3) pp. 68-74 link.

RUHS, M (2017). Rethinking international legal standards for the protection of migrant workers: the case for a ‘core rights’ approach, American Journal of International Law Unbound, 111, pp. 172-176.

Shah, N (2009). The management of irregular migration and its consequences for development: Gulf Cooperation Council, link.

Silverman, B (2014). Modern Slavery: an application of Multiple Systems Estimation, link.

Skrivankova, K (2010). Between decent work and forced labour: examining the continuum of exploitation, link.

Sonmez, S. et al. (2011). Human rights and health disparities for migrant workers in the UAE, Health and Human Rights, 13, pp. 17-35.

Such, E. & Salway, S (2017). Modern slavery and public health, link.

Surtees, R (2008). Trafficked men as unwilling victims, St Anthony’s International Review, 4, pp. 16-36.

Surtees, R. & Johnson, L. S (2019). Quality and rigor in tip research in the Mekong region. Assessing the evidence base (2008-2018), Nexus institute link.

Thomas, S. E (2011). Responses to human trafficking in Bangladesh, India, Nepal and Sri Lanka, UNODC Regional Office for South Asia, link.

Turner-Moss, E. et al. (2014). Labour exploitation and health: a case series of men and women seeking post-trafficking services, Journal of Immigrant Minority Health, 16, pp. 473-480.

Ucnikova, M (2014). OECD and modern slavery: how much aid money is spent to tackle the issue? Anti-Trafficking Review, link.

UK Legislation (2015). Modern Slavery Act 2015. The National Archives.

UNDP (2015). Human Development Report 2015. Work for Human Development. New York, NY, USA:

United Nations. Sustainable Development Goals, link.

United Nations (2003). United Nations Protocol to Prevent, Suppress and Punish Trafficking in Persons, especially Women and Children, link.

United Nations (2018). Global compact for safe, orderly and regular migration. Intergovernmental negotiated and agreed outcome, link.

UNODC. Status of ratification interactive dashboard: International Convention on the Protection of the Rights of All Migrant Workers, link.

UNODC (2000). United Nations Convention against Transnational Organized Crime and the Protocols Thereto.

UNODC (2018a). Global report on trafficking in persons, link.

UNODC (2018b). Global Report on Trafficking in Persons 2018, Trafficking in persons in the context of armed conflict, link.

US State Department (2000). International and Domestic Law: office to monitor and combat trafficking in persons, link.

USAID (2009). Audit of USAID/Cambodia’s Counter-Trafficking In Persons Project, link.

Van Der Laan, P. et al. (2011). Cross-border trafficking in human beings: prevention and intervention strategies for reducing sexual exploitation, link.

Van Dyke, R (2017). Monitoring and Evaluation of Human Trafficking Partnerships in England and Wales, Anti-Trafficking Review, pp. 131-146, link.

Verité (2012). Labour brokerage and trafficking of Nepali migrant workers, link.

Verité (2014). Forced labour in the production of electronic goods in Malaysia. A comprehensive study of scope and characteristics, link.

Verité (2017). Strengthening protections against trafficking in persons in federal and corporate supply chains research on risk in 43 commodities worldwide, link.

Walk Free Foundation (2018). The Global Slavery Index 2018, link.

Walter, N., Bourgois, P. & Loinaz, H. M (2004). Masculinity and undocumented labor migration: injured latino day laborers in San Francisco. Social Science & Medicine, 59, pp. 1159-1168, link.

Weitzer, R (2014). New directions in research on human trafficking, The ANNALS of the American Academy of Political and Social Science, 653, pp. 6-24.

Wickramasekera, P (2002). Asian labour migration: issues and challenges in an era of globalization, link.

World Bank (2016). New Development Update: remittances at risk, link.

Zhang, S. X. & Pineda, S. L (2008). Corruption as a Causal Factor in Human Trafficking. In: Siegel, D. & Nelen, H. (eds.) Organized Crime: Culture, Markets and Policies. Studies in organized Crime, Springer New York.