Since the concept of a conviction is more uniform across countries, conviction statistics tend to be somewhat more reliable and comparable than the other statistics on offenders. Although this might be true for THB statistics as well, the variation in the numbers of convictions for THB across countries is still unexplainably large. The relatively high numbers in France seems, for example, to be inflated by the erroneous inclusion of convictions for other types of offences than THB, such as pimping.20
As mentioned, the numbers of convictions for THB are generally rather low. One explanation for these low numbers is that persons suspected of having engaged in human trafficking are sometimes convicted for other, less serious offences, when the evidence for THB is deemed too weak by prosecutor or court. Convictions for such ‘lesser offences’ do not usually show up under the published convictions for THB. This can explain part of the attrition in the final stage of convictions. Nevertheless, and as observed in the UNODC 2014 Global Report, the numbers of convictions for THB remain disappointingly low in most countries and, even more disappointingly, show hardly any sign of increasing in most parts of the world.
Statistics on compliance with treaty obligations
This section deals with statistics on the nature of special counter-trafficking policies. This category of THB-related statistics reflects the compliance of States with their international legislative and operational obligations. The relevant treaties, such as the Palermo Protocol and CoE Convention, as well as the EU Directive, typically oblige States to provide identified victims with specialised victim assistance, legal aid and compensation for material and immaterial damages. In addition, States are obliged to offer victims without residential rights a recovery and reflection period, safe repatriation and, under certain conditions, a temporary and/or permanent residence permit.
In annual evaluation reports, the US Department of State assesses countries with respect to their compliance with anti-trafficking policies – as defined by the USA (so-called TIP reports).21 The so-called TIP country reports assess national policies according to a four tier system (1, 2, 2 Watch List and 3). Combining data from the TIP country reports and from the UNODC Global Reports, Young Cho, Dreher and Neumayer have rated 177 countries with respect to their level of compliance with the Palermo Protocol.22 This so-called 3P index has a maximum score of 15. The mean score of all countries is 9.8. The list of best performing countries, with scores of 14 or higher, comprises Armenia, Belgium, Canada, Cyprus, Italy, South Korea, Moldova, the Netherlands and Sweden. The worst performing countries, with scores of 6 or lower, include Algeria, Cuba, Curacao, Haiti and Kuwait.
A distinct feature of the CoE Convention is its rigorous monitoring mechanism: conducted by a group of 15 independent experts (GRETA) elected by the State Parties. The country evaluation reports of GRETA cover all key treaty obligations of the Convention. An analysis of these reports revealed that national anti-trafficking policies are systematically assessed on their compliance with 34 main policy requirements. If GRETA judges that a country’s policies do not comply with the treaty obligations, it ‘urges’ changes. If the policies are roughly in compliance but need improvement, GRETA ‘considers’ that changes should be made. If no recommendations are formulated, the country’s existing policies are apparently deemed satisfactory. Building on these results, a quantitative scorecard was designed, assessing compliance with the 34 policy requirements applied in the GRETA reports, by Van Dijk and Klerx-Van Mierlo.23 This scorecard is subdivided into scores for requirements regarding the legal-institutional framework, prevention/protection of victims, law enforcement and prosecution and prevention, respectively. The maximum total score on the index is 68 (indicating full compliance with all treaty obligations).
The GRETA-based country scores were compared with the so-called 3P index scores mentioned above. The concurrence of the two ratings is far from perfect, yet relatively strong (r= .65). This statistical result suggests that, in the European region, the quantitative ratings of countries based on the US TIP reports are broadly in line with the more refined evaluations by GRETA.
Table 30.5 presents the total scores and sub-scores, on the GRETA-based index, of 18 destination countries of human trafficking in Europe, and in Canada.24
Table 30.5 CRETA-based scores and sub-scores, and ranks of 19 destination countries
Table 30.5 shows that the Netherlands, Austria and Belgium received the most favourable assessments from GRETA. Small countries like Malta, Cyprus and Luxembourg received relatively low scores, especially with regard to the legal-institutional framework. The insufficiency of their specialised THB frameworks seems related to a low number of THB cases. France stands out with a low ranking, too, which seems to reflect a lack of political will to address THB. Unlike the favourable rating in the 3P index, based on the TIP reports, Canada scored a relatively poor 33 points on the GRETA-based index, ranking 14th. Canada lacks, for example, a formal identification and referral system as well as specialised shelters for THB victims in large parts of the country.
Similar evaluations have been made of a sample of source countries of THB in Europe. On average, the source countries received the same scores as destination countries (a total of 37 points). Regarding enforcement, the scores are somewhat higher – possibly reflecting a strong commitment to arrest and prosecute recruiters in source areas. Of the ten source countries evaluated, Bosnia and Herzegovina, Latvia and Slovakia received the lowest scores.
In the Human Slavery Index reports of the Walk Free Foundation, the estimates of the numbers of victims are supplemented by an index of the quality of government responses to human trafficking.25 A preliminary analysis suggests that scores on this government responses index show fairly strong correlations with the scores based on the TIP reports just mentioned.26 This finding further supports the validity of these indices.
Discussion and recommendations
This assessment of the available descriptive statistics on THB victims begs the question: for what purposes can these statistics reliably be used? In our view, statistics on registered victims mainly reflect the efforts of governmental agencies to identify victims through formal mechanisms of identification. In countries with comprehensive identification systems, such as the Netherlands, Romania and the UK, the numbers of identified victims are relatively high. A multivariate analysis by the econometrist Young showed that numbers of identified victims of countries are correlated with the resources for law enforcement available per capita.27 The more resources available for enforcing THB laws, the higher the numbers of identified victims. The numbers of identified victims, then, provide no useful estimate of the real volumes. Since law enforcement efforts vary across countries, they do not provide a reliable ranking of countries in terms of the seriousness of the phenomenon of THB, either. In fact, as is the case with other complex crimes, such as corruption and fraud,28 the official statistics on registered THB cases may be inversely related to the prevalence of the criminal phenomenon under examination. High rates of official statistics on recorded cases point to a serious commitment of governments to address the phenomenon, perhaps resulting in lower actual prevalence. Countries with low numbers of identified victims, such as countries in the Middle East, may be in denial about the occurrence of THB, invest little in its detection and may consequently tolerate a comparatively high prevalence. Thus, official THB statistics on victims can definitely not be used as a measure of true prevalence. Neither can they be used to determine trends over time in the phenomenon, since improved anti-trafficking policies are likely to result in higher numbers of identified victims. For example, recent increases in identified victims of forced labour, observed by both UNODC29 and GRETA,30 are more likely to reflect extra efforts by law enforcement agencies and labour inspectorates to tackle this type of human trafficking, than a rise in real prevalence.
So why, one could ask, continue with the collection of statistics on identified victims at all? One compelling reason to do so is to monitor the efforts
of governments and NGOs to officially identify victims of different types of THB. Also important are comparisons between the numbers of identified victims and the numbers of victims who have received assistance, residence permits or compensation, and the numbers of convicted traffickers. Although such comparisons are far from straightforward and require detailed knowledge of the specific arrangements in each country, they can serve to identify failures in implementing treaty obligations. In many country evaluation reports of GRETA, such comparisons are made for this purpose; for example, by criticising wide gaps between high numbers of identified victims and low numbers of convicted traffickers in some countries, including the UK, Poland and Ireland.31
The time seems ripe for concerted efforts to estimate the true prevalence of victimisation by human trafficking. Plans have been made by the ILO and the Walk Free Foundation to conduct a new series of population surveys on experiences with THB in order to produce better, and country-specific, estimates of the real numbers of victims.32 Interestingly, an alternative method to estimate the true numbers of THB victims was recently piloted in the UK and the Netherlands, using multi-source data on identified victims. In countries with National Referral Mechanisms, possible victims of THB are often identified by several governmental agencies, such as the police, immigration and labour inspectorates, as well as by NGOs. While the possibility of the same victims being recorded by two or more organisations was initially regarded as problematic, this double counting of victims on different lists offers opportunities for estimating the true numbers through statistical modelling (Capture Recapture analysis or Multiple Systems Estimation/MSE). In the UK, the true volume of victims was estimated to be four times the number of identified victims.33 In the Netherlands, the true number was estimated as 17,000.34 If, as planned by UNODC in conjunction with the Walk Free Foundation, more countries are assisted to apply MSE to their multi-agency databases on identified victims, an additional method of estimating the prevalence of victimisation by THB will become available. Future global estimates of the numbers of THB victims are likely to be based on the mixed mode of survey research and Multiple Systems Estimation.
Tackling human trafficking and forced labour seems set to remain a political priority in many parts of the world. Action against these crimes has been incorporated in the newly adopted Sustainable Development Goals (SDGs) of the UN.35 Agreement has also been reached on “the numbers of victims of human trafficking disaggregated by sex, age, and type of exploitation”, as indicators to measure progress in achieving this SDG. Deeper investments by governments and International Organisations, as well as by foundations, in the production of reliable and comparable statistics on THB victims are therefore to be expected. When improved international statistics on victims of THB, based on extensive and repeated survey research as well as on MSE, are made available, the global action against human trafficking will be put on a much stronger evidential footing.
Notes
1 In the framework of the Global Plan of Action, the General Assembly mandated UNODC to collect information and publish a Global Report on Trafficking in Persons every two years (General Assembly Resolution 64/293).
2 The Group of Experts on Action against Trafficking in Human Beings (GRETA) considers: “that for the purpose of preparing, monitoring and evaluating anti-trafficking policies and measures, the authorities should develop and maintain a comprehensive and coherent statistical system on trafficking in human beings by compiling reliable statistical data on identified or otherwise registered victims, measures to protect and promote the rights of victims as well as on the investigation, prosecution and adjudication of human trafficking cases”.
3 The Global Report on Trafficking in Persons is a biannual publication. See UNODC, Global Report on Trafficking in Persons 2014. At the time of writing, the 2016 report had not yet been released.
4 Eurostat, Trafficking in Human Beings, 2015 Edition (Eurostat Statistical Working Papers, 2015).
5 UNODC, Trafficking in Persons: Global Patterns (Vienna: UNODC, April 2006).
6 International Labour Organization, ILO Minimum Estimate of Forced Labour in the World (Geneva: ILO, 2005); International Labour Organization, Global Estimate of Forced Labour: Results and Methodology (Geneva: ILO, 2012).
7 Walk Free Foundation, The Global Slavery Index 2016 (Australia, the United Kingdom, and India: The Minderoo Foundation Pty Ltd., 2016).
8 UNODC, Global Report on Trafficking in Persons 2014 (Vienna: UNODC, 2014).
9 Organisation for Security and Economic Cooperation in Europe, National Referral Mechanisms – Joining Efforts to Protect the Rights of Trafficked Persons: A Practical Handbook (Warsaw: OSCE/ODHIR, 2004).
10 UNODC, Model Law Against Trafficking in Persons (Vienna: UNODC, 2009).
11 Van Dijk, J., van der Knaap, L.M., Aebi, M.F., and Campistol, C., Counting What Counts: Tools for the Validation and Utilization of EU Statistics on Human Trafficking (Tilburg: PrismaPrint, 2014).
12 UNODC, Global Report on Trafficking in Persons 2014 (Vienna: UNODC, 2014).
13 De Vries, I. and Dettmeijer-Vermeulen, C., “Extremely Wanted: Human Trafficking Statistics – What to Do With the Hodgepodge of Numbers?”, in Kangaspunta, K. (ed.), Special Issue: Researching Hidden Populations: Approaches to and Methodology for Generating Data on Trafficking in Persons (2015) 8 Forum on Crime and Society (New York: United Nations), pp. 15–37.
14 Kangaspunta, K., “Mapping the Inhuman Trade: Preliminary Findings of the Database on Trafficking in Human Beings” (2003) 3 Forum on Crime and Society 81–105. See, also, UNODC, Trafficking in Persons: Global Patterns (2006).
15 ILO, A Global Alliance Against Forced Labour: Global Report Under the Follow-up to the ILO Declaration on Fundamental Principles and Rights at Work (International Labour Conference, 93rd Session 2005, Report I (B) 2005).
16 ILO, Global Estimate of Forced Labour: Results and Methodology (Geneva: ILO, 2012).
17 ILO, Global Estimate of Forced Labour – Regional Factsheet: European Union (Geneva: ILO, 2012). For a review of the ILO studies see Van der Heijden, de Vries, Boehing, and Cruyff, “Estimating the Size of Hard-to-reach Populations Using Capture-recapture Methodology, With a Discussion of the International Labour Organization’s Global Estimate of Forced Labour”, in Kangaspunta, K. (ed.), Special Issue: Researching Hidden Populations: Approaches to and Methodology for Generating Data on Trafficking in Persons (2015) 8 Forum on Crime and Society (New York: United Nations), pp. 109–137.
18 Zhang, S.X. and Cai, L., “Counting Labour Trafficking Activities: An Empirical Attempt at Standardized Measurement”, in Kangaspunta, K. (ed.), Special Issue: Researching Hidden Populations: Approaches to and Methodology for Generating Data on Trafficking in Persons (2015) 8 Forum on Crime and Society (New York: United Nations), pp. 37–63.
19 Walk Free Foundation, The Global Slavery Index 2016 (Australia, the United Kingdom, and India: The Minderoo Foundation Pty Ltd., 2016).
20 The GRETA evaluation report on France observes that in France, in recent years, just one or two convictions for THB have been obtained. See Report Concerning the Implementation of the Council of Europe Convention on Action Against Trafficking in Human Beings by France (28 January 2013), p. 57.
21 According to the US Government Accounting Office, the rating procedures and criteria used in the TIP reports are unclear, making the final ratings vulnerable to political subjectivity. See USGOA, Human Trafficking, Better Data, Strategy, and Reporting Needed to Enhance U.S. Anti-trafficking Efforts Abroad (Washington: US GOA, July 2006), pp. 26–36.
22 Young Cho, S., Dreher, A., and Neumayer, The Spread of Anti-trafficking Policies – Evidence From a New Index (CESinfo Working Paper No. 3376, Munich, 2011).
23 Van Dijk, J.J.M. and Klerx-Van Mierlo, F., “Quantitative Indices for Anti-THB Policies Based on Reports of the State Department of the USA and the Council of Europe” (2013) 61(2) Crime, Law and Social Change 229–250. If no recommendations are made, the country receives a score o
f 2; if changes are considered, a score of 1; and if changes are urged, a score of 0.
24 Van Dijk, J.J.M., Mécanisme de suivi anti-traite du Conseil de l’Europe (GRETA), Presentation at Montreal University, on GRETA, at the Occasion of the Week of the Victim (24 April 2015), www.youtube.com/watch?v=NkwQUU9R7WI.
25 Walk Free Foundation, The Global Slavery Index 2016 (Australia, the United Kingdom, and India: The Minderoo Foundation Pty Ltd., 2016).
26 Personal communication of the first author, based on unpublished findings of Katharine Bryant of Walk Free.
27 Young Cho, S., “Towards a Comprehensive Index on Anti-trafficking Policy – An Assessment of the 3P Index, GRETA-based Scorecard and EuroStat”, in Van Dijk, J., Van der Knaap, L.M., Aebi, M.F., and Campistol, C. (2014), Counting What Counts: Tools for the Validation and Utilization of EU Statistics on Human Trafficking (Tilburg: PrismaPrint, 2013), pp. 154–198.
28 Van Dijk, J.J.M., The World of Crime: Breaking the Silence on Problems of Security, Justice and Development Across the World (Thousand Oaks: Sage, 2008).
Routledge Handbook of Human Trafficking Page 78
