At first sight, then, the collection of statistics on THB victims seems an impossible task – even more so when they should be comparable across countries. Fortunately, the collection of such statistics is facilitated by two unique characteristics of THB. First, THB is one of the rare types of crime for which an international definition has been adopted; namely the definition in the Palermo Protocol, adopted in 2000. According to the latest count, 170 countries have ratified this Protocol, and 146 countries have wholly, or partially, adopted its definition of human trafficking in their national criminal codes.8 Second, in 2004 the OSCE published a handbook on the identification and referral of THB victims.9 Building on this document, the Council of Europe Convention on Action against Trafficking in Human Beings of 2005 (CoE Convention) obliges parties to introduce “appropriate mechanisms for the identification and referral of THB victims” (so called National Referral Mechanisms). Since the CoE Convention has been ratified by nearly all Member States of the Council of Europe, most European countries can produce statistics on the numbers of persons who have been formally identified as victims of THB in the sense of the Palermo Protocol. A mechanism to identify and refer victims of THB also forms part of the United Nations Model Law on Trafficking in Persons.10
The fact that many countries can indeed report on numbers of formally identified victims of THB does not, unfortunately, mean that these statistics are necessarily comparable across countries. Comparability is compromised by the fact that neither the CoE Convention nor the UN Model Law provides further guidance on the nature of the proposed identification mechanism. In the absence of an international standard for identifying THB victims, statistics on THB victims reflect the nation-specific mechanisms of identification developed in various countries.
The latest Eurostat report on the number of identified THB victims provides metadata detailing the precise sources and meanings of the figures per country. This section’s overview shows a stunning variation in what the concept of an identified victim means within the countries discussed. Some examples, taken from the report of Van Dijk et al.,11 may suffice to convey the scope of variation. As mentioned, Sweden reports exclusively on victims involved in trafficking cases irrevocably adjudicated in a Swedish court. At the other extreme, the Netherlands reports on all persons encountered by State institutions or NGOs and showing any sign of having been trafficked. The Dutch recording mechanism, operated by an NGO on behalf of the Parliamentary Rapporteur on THB, covers a larger number of victims than those formally identified by the police/immigration. The UK collects data on all persons identified as presumed victims of trafficking, and on those conclusively identified as victims. In earlier reports to Eurostat, the UK reported on the subgroup of persons conclusively identified as victims. In reports on 2011 and 2012, the UK switched to the much higher numbers of presumed victims. France has, to date, not established a formal identification mechanism for victims of human trafficking. The French statistics sent to Eurostat are obtained from police administrations, and include persons subjected to illegal pimping.
Considering the vast differences in the meaning of the concept of an identified victim, the Eurostat statistics on identified victims cannot, in our view, be usefully compared with each other. Differential identification mechanisms and practices impact not just on the numbers of victims identified, but also on the types of victims. For example, Germany, France, Hungary and Latvia report very few male victims of THB (see Table 30.2). This does not imply that few males are victimised by labour exploitation in these countries, but, rather, that national identification policies are strongly focused on sexual exploitation, and therefore register mainly female victims. In other words, both the numbers and the type of victims identified per country are heavily influenced by the scope and focus of the national identification mechanisms in place.
Since the Palermo Protocol contains no provision on identification mechanisms, the statistics on numbers of identified victims provided to UNODC by non-European countries are likely to vary even more. Only a minority of non-European countries seem to have introduced special identification mechanisms. By way of example, neither Canada nor the USA has introduced nationwide identification mechanisms. Little is known about the existence or nature of identification mechanisms in place outside Europe.12 In the absence of metadata on this issue, it is difficult to determine the meaning of the statistics on identified victims presented in the Global Reports on Trafficking in Persons of UNODC.
Regional and country statistics
Having made these caveats about the comparability of these statistics, the following tables provide an overview of the numbers of identified THB victims in global regions, selected individual countries, and in the Member States of the EU, based on data collected by the UNODC and Eurostat, respectively.
Table 30.1 Total number of registered victims of THB per global region in 2012, by gender and age
Table 30.1 shows that by far the highest numbers of THB victims are identified in Western and Central Europe (comprising mainly EU Member States). The world totals of identified victims are dominated by the European results. The breakdowns show that over 60% of identified victims are women, and that 15% are minors, mainly girls.
Figure 30.1 Total number of registered victims of THB in ten selected countries in 2012, by age
Source: adaptation from UNODC 2014
While a majority of trafficking victims are subjected to sexual exploitation, other forms of exploitation are, according to the UNODC Global Report, more often detected than before. The numbers of identified victims of (trafficking for) forced labour – a broad category which includes, for example, exploitative practices in manufacturing, cleaning, construction, agriculture, catering, restaurants, textile production and domestic work – has increased steadily in recent years. Some 40% of the victims detected between 2010 and 2012 were trafficked for forced labour. Trafficking for exploitation that is neither sexual nor forced labour, such as trafficking of children for armed combat or for involvement in petty crime or forced begging, is also increasing.
There are considerable regional differences with regard to the various forms of exploitation. While trafficking for sexual exploitation is the main form detected in Europe and Central Asia, forced labour is the main form in East Asia and the Pacific. In the Americas, the two types are detected in near equal proportions. The gender ratio is less skewed in regions where more cases of labour exploitation are detected.
The UN’s Global Report on Trafficking in Persons, itself, does not present THB statistics or is on individual countries. Figure 30.1, derived by the authors from the dataset of UNODC, shows the total number of victims, adult and children, from ten selected countries.
By far the highest rates, for both adult and child victims, are reported by some West European countries (notably the UK and the Netherlands). By comparison, the absolute numbers registered in Canada, Japan and the USA, for both children and adults, are extraordinarily low. The comparatively low numbers for the latter countries suggest insufficiencies in their methods of identifying victims of THB. As noted, Canada and the USA lack nationwide identification mechanisms.
In recent years Eurostat has, as mentioned, made an effort to collect statistics on THB from all EU Member States and associated countries. Table 30.2 shows the absolute numbers of victims, and rates per 100,000, in 2012, overall and according to gender, as published by Eurostat.
Table 30.2 Number of registered victims of THB in Europe in 2012, per 100,000 inhabitants, and by gender
Countries Total number of victims Victims per 100,000 inhabitants Males per 100,000 inhabitants Females per 100,000 inhabitants
* * *
EU-28 10998 2.2 1.0 3.0
* * *
Belgium 157 1.4 1.5 1.3
Bulgaria 579 7.9 2.0 13.5
Czech Republic 100 1.0 N/A N/A
Denmark 66 1.2 0.4 2.0
Germany 626 0.8 0.1 1.4
Estonia 22 1.7 2.3 1.1
Ireland 48 1.0 0.
7 1.3
Greece 94 0.8 0.3 1.3
Spain 125 0.3 0.0 0.5
France 751 1.2 0.0 2.2
Croatia 11 0.3 0.2 0.3
Italy 2631 4.4 2.5 6.1
Cyprus 37 4.3 2.6 5.9
Latvia 144 7.0 0.3 3.2
Lithuania 14 0.5 0.2 0.7
Luxembourg 7 1.3 0.4 2.3
Hungary 57 0.6 0.1 1.0
Malta 8 1.9 0.0 3.8
Netherlands 1711 10.2 2.4 17.9
Austria 103 1.2 0.3 2.1
Poland 246 0.6 N/A N/A
Portugal 5 0.0 0.1 0.0
Romania 1041 5.2 3.8 6.5
Slovenia 67 3.3 0.2 1.3
Slovakia 40 0.7 0.2 0.6
Finland 75 1.4 1.4 1.3
Sweden 88 0.9 0.1 0.2
United Kingdom 2145 3.4 2.8 3.6
Iceland 2 0.6 0.6 0.6
Norway 48 1.0 0.6 1.4
Switzerland 60 0.8 0.1 1.4
Montenegro … … … …
Serbia 79 1.1 0.5 1.6
Turkey 55 0.1 0.0 0.1
Source: Eurostat 2015
Table 30.2 shows huge variation in the absolute numbers of victims, as well as in the rates per capita, identified across European countries. The highest rates per capita are reported by the Netherlands, Bulgaria, Latvia, Romania, Cyprus, Italy and the UK. The gender gaps show large variations across the EU Member States, too, with Belgium, Estonia and Finland standing out with higher rates among males than among females. As said, the latter finding points to differences in detection priorities rather than in the extent of the phenomenon.
Dark figure studies
Since statistics on identified victims of THB show only the part of the phenomenon which is detected by the authorities – and this part differs vastly across countries, to boot – such statistics cannot be used as comparative measures of the true volume of THB. Even if their comparability would be improved, statistics on identified victims per country would still show just ‘the tip of the iceberg’. To monitor trends in human trafficking for policy purposes, statistics must be collected on the true numbers of THB victims, regardless of their identification by the authorities.13 A first pioneering attempt to supplement officially registered numbers of victims with data from informal sources was made by UNODC in 2003. Information was collected on incidents of THB reported in public sources, including public media. This innovative project yielded a tentative ranking of countries according to the numbers of victims reported on in public sources – differentiating between source, transit and destination countries.14
In 2005, the ILO published a global estimate of victims of forced labour, including for sexual exploitation. This estimate was based on an extrapolation of the counts of publicly reported victims of forced labour, identified by two independent teams of analysts.15 The results showed a minimum global estimate of 12.3 million victims of forced labour at any point in time during the period 1995–2004. In 2012, the ILO improved on its methodology by combining an estimate of recorded cases based on capture-recapture methodology with the results of survey studies among populations of returned migrants about experiences of exploitation. The 2012 results produced an estimate of 20.9 million victims at any point in time covering the period 2002–2011.16 For the EU, the estimated total of victims amounted to 888,000 at any given time, or 1.8 per 1,000 inhabitants.17 Out of the total number of 880,000 estimated forced labourers in EU Member States, 270,000 (30%) were victims of sexual exploitation, and 610,000 (70%) of labour exploitation. It is worth noticing that this estimated true volume for Europe is many times larger than the 10,000 identified victims per year reported by Eurostat for 2012, and that the majority of these victims have been subjected to labour exploitation rather than to sexual exploitation. Compared to the ILO estimates, the identified victims recorded by Eurostat are indeed just ‘the tip of the iceberg’ –especially so regarding the numbers of victims of labour exploitation.
Surveys among population groups at risk of being trafficked, sampled through respondent-driven sampling techniques, have been pilot tested in the USA.18 These surveys suggested that as many as 30% of Hispanic migrant workers in California are exposed to exploitative practices that qualify as forced labour under American law. The NGO Walk Free produced global estimates based on a combination of analyses of documents and surveys among representative samples of national populations in selected countries.19 The 2016 Global Slavery Index estimates there are 45.8 million people living in some form of modern slavery globally. This estimate is once again significantly higher than that of the ILO of 2012. As can be gathered from the changing, ever increasing, global numbers of THB/forced labour victims, the estimation of the hidden numbers of THB is still very much a work in progress. The expectation is that, with the conducting of standardised victimisation surveys in more and more countries, as planned by Walk Free and the ILO, the estimates will gradually become more robust.
Offender-related statistics
In theory, statistics on offenders can, in most Western countries, unlike statistics on victims, be derived from regular statistical systems of the police, the prosecutors and/or the courts. However, in reality, many States have failed to report data on these issues to UNODC. This is especially the case with statistics on suspected offenders/first contacts (police statistics) and prosecuted offenders (prosecution statistics). The large numbers of missing values, and changes in reporting countries over the years, complicates the interpretation of global and regional results. Table 30.3 shows the numbers of persons arrested for THB (first contact), persons prosecuted and persons convicted, respectively, worldwide and for world regions, in 2010, 2011 and 2012.
Overall, the global picture of the criminal justice response shows few signs of change in recent years. Globally, the number of convictions per 100,000 population remained basically unchanged in comparison to the 2007–2010 period (0.1 per 100,000 population). The global rate of 13,000 convictions in 2012 compares unfavourably with the estimates of the ILO and Walk Free of 20 to 30 million people subjected to slavery at any time. The gap between estimated number of victims and the number of convictions suggests near immunity for traffickers/exploiters in most parts of the world. Europe and Central Asia report most convictions per 100,000 population (around 0.3). South Asia, East Asia and the Pacific report a rate above 0.1; while the Americas, Africa and the Middle East register rates well below 0.1.
The global data on investigations, prosecutions and convictions collected for the 2010–2012 period by UNODC show steep increases in reported cases in Asia, partly due to an increase in reporting countries. No firm conclusions can be drawn from the global and regional data on attrition between arrests and conviction. Remarkably, the total numbers of convictions are often significantly larger than those of prosecutions. Apparently, prosecutions tend to be less well recorded in some countries. As said, global and regional figures are heavily influenced by changing missing values. In order to get a better picture, Table 30.4 presents the data on 2012 of a selection of individual European and Central Asian countries. Some of these countries have duly reported on arrests, prosecutions and convictions.
As can be seen from the data for these selected individual countries, there is, in most countries, significant and incremental attrition from first contacts to prosecutions, and from prosecutions to convictions. The totals decrease from 3,699 arrests/suspects to 2,320 prosecutions, and subsequently to 1,659 convictions.
Table 30.3 Total number of offenders (first contact, prosecuted and convicted) of THB, per global region, in 2010, 2011 and 2012
Table 30.4 Total number of offenders in selected countries from Europe and Central Asia, in 2012 (or last available year)
Total First Contact Total Prosecuted Total Convicted
* * *
Albania … 29* 11*
Austria 231 … 20
Belgium … 83* …
Bosnia 42 … 13
Bulgaria 198 101 110
Croatia 7 2** 5**
Cyprus … 46** 1*
/> Czech Republic … 35* 11
Denmark 9 21 2
Estonia … … 36
Finland … … …
France … … 664*
Germany … 148** 121**
Greece … 171 37*
Hungary 33 18 8
Ireland … 15 6
Italy 522 362* 99*
Latvia 17 12 19
Lithuania 32 24 7
Luxembourg … … 2*
Macedonia 6 60 53
Malta 4 1 1
Monaco … … …
Montenegro … 10* 10*
Netherlands … 311 109
Norway … 22 8
Poland 23 65 22
Portugal … … …
Romania 1342 536 427
Serbia … 52** 39**
Slovakia … 16 12
Slovenia 12 13** 8
Spain 195 … 22
Sweden … 15 5
Switzerland 71 … 5**
Turkey 86 60 49
United Kingdom 32 45 30
Armenia … … …
Azerbaijan … 22 …
Belarus 75 75 37
Georgia … … …
Kazakhstan … … …
Kyrgyzstan … 134** 20**
Moldova … 111** 21
Russia … … …
Tajikistan … 44** 4**
Ukraine 22 16* 28
Uzbekistan 740 730 595
Total 3699 2320 1659
* Data from 2010
** Data from 2011
Source: Authors’ adaptation from UNODC 2014
Even among countries with data on all three stages, between-country comparisons prove to be difficult – especially for the categories of suspected and prosecuted offenders. This is mainly due to the very different meanings of the concepts of arrests/suspects and prosecutions in national laws and practices and, on top of that, to differences in counting rules. However, police and prosecution data on THB should not be disregarded altogether. Such data can potentially provide important information on the nationality of suspects, and on the types of exploitation detected offenders are suspected of. Such information reveals the policy priorities of police forces and prosecution offices in fighting THB. In some countries, for example, no or very few arrests or prosecutions have ever been recorded for forced labour. Increased numbers of such cases would indicate a change in criminal policies.
Routledge Handbook of Human Trafficking Page 77