The Use of Hotspots in the Identification of the Factors that Predict Human Trafficking

 

Lincoln J. Fry[1]

Office of the State Attorney, Florida 15th Judicial Circuit, USA.

 

Abstract

The purpose of this paper is to determine whether the use of "trafficking hotspots" is a useful approach in the quest of identifying factors that predict human trafficking at the country level. The paper builds upon earlier studies by Bales (2007) and Fry (2010, in press) that attempted to identify the factors which predicted trafficking based on the classification of countries on the basis of origin, transit, and destination sources of human trafficking. This study uses the database constructed by the Global Program Against Trafficking of Human Beings (GPAT), sponsored by the United Nations Office on Drugs and Crime to assess the trafficking hotspots question. The paper begins by discussing the definition of human trafficking, covers international agreements, data collection issues, reviews the literature, including the papers presented by Bales and Fry and then describes the present study. The results indicated that the notion of trafficking hotspots has some efficacy in the quest to identify the factors that predict human trafficking. The conclusion was that research in the area needs to be supplemented by additional data collection and that future research should be guided by the identification of hotspots at the country or regional level.

______________________________________________________________________________________________________________

Key Words: Trafficking; Victimization; Hotspots; Offenders; Victims. 

 

Introduction

    Human trafficking has been recognized as a global phenomenon and the international community has made it a high priority, devoting considerable resources to its prevention, prosecution and the care of its victims. Prior research suggests that approaches based on global estimates of human trafficking do not lead to policy relevant findings and this paper continues the quest to identify the factors that predict trafficking at the country level. Lindstrom (2006) argued that human trafficking policy is top down, and is in need of input from the bottom up, especially from victims. In his search for the factors that predict human trafficking at the country level, Bales (2007) took a global perspective but suggested that although every trafficking case (country) is unique, they all share certain characteristics. These comments suggest the need to develop human trafficking policy at the country level and this paper will attempt to do so by understanding the use of the trafficking hotspots concept.

 

Human Trafficking

    The most commonly used definition of human trafficking is drawn from a supplemental protocol to the United Nations Convention Against Transnational Crime, adopted in December 2000. The document defines human trafficking as, "Trafficking in persons shall mean the recruitment, transportation, transfer, harboring or receipt of persons, by the threat or use of force, by abduction, fraud, deception, Coercion or the abuse of power or by the giving or receiving of payments or benefits to achieve the consent of a person having control over another person... sexual exploitation, forced labor or services, slavery or practices similar to slavery" (UN, 2000). The UN protocol has helped to define the problem in many ways; this is not to say that there is universal acceptance or endorsement of its content. As Oxman-Martinez, Martinez and Hanley (2001) noted, the UN Protocol does not mention borders. They see human trafficking as part of a broader picture which includes the connections of human trafficking to economic globalization and transnational crime. This paper agrees with that position and is concerned only with international trafficking, which by definition, includes crossing at least one border.

 

International Agreements

    The UN General Assembly adopted two international treaties (protocols) to fight international trafficking which supplemented the United Nations Convention Against Organized Crime. One protocol dealt with trafficking and the other with smuggling. They are part of a package designed to address transnational organized crime. The Trafficking in Persons Global Patterns (TPGP) Report (2006) offered recommendations derived from those protocols in a number of areas, prevention, prosecution, protection and intervention. The trafficking protocol is the focus here and it attempts to create a global language as the basis for subsequent legislation to define trafficking, assist victims of trafficking, and prevent trafficking in persons. The protocol also attempts to establish judicial cooperation and information exchange among countries.

    As Gallagher (2002) indicated, the trafficking protocol requires countries to:

1)      Criminalize trafficking and related conduct as well as impose appropriate penalties.

2)      Facilitate and accept the return of their trafficked nationals and permanent residents with due regard for their safety.

3)       When returning trafficked persons, to ensure that this happens with due regard, both for the safety of the trafficked person and the status of any relevant legal proceedings.

4)      Exchange information aimed at identifying perpetrators or victims of trafficking, as well as methods and means employed by traffickers.

5)      Provide or strengthen training for law enforcement, immigration and other relevant personnel aimed at preventing trafficking, prosecuting traffickers and protecting the rights of victims.

6)      Strengthen border controls as necessary to detect and prevent trafficking.

7)      Take legislative or other appropriate measures to prevent commercial transport being used in the trafficking process and to penalize such involvement.

8)      Take steps to ensure the integrity of travel documents issued on their behalf and to prevent their fraudulent use.

 

    Gallagher (2002) also notes that the protocol contains victim protection measures. Most of these recommendations are optional but countries are supposed to do the following:

1) Protect the privacy of trafficking victims and ensure they are given information on legal proceedings and facilities to present their views and concerns during criminal procedures against offenders.

2) Consider implementing a range of measures to provide for the physical and psychological recovery of victims of trafficking.

3) Endeavor to provide for the physical safety of trafficking victims within their territory.

4) Ensure that domestic laws provide victims with the possibility of obtaining compensation.

5) Consider adopting legislative or other measures permitting victims of trafficking to remain in their territories temporarily or permanently in appropriate cases with consideration being given to humanitarian and compassionate factors.

6) Endeavor to establish policies, programs and other measures aimed at preventing trafficking and protecting trafficked persons from re-victimization.

7) Endeavor to undertake additional measures including information campaigns and social and economic initiatives to prevent trafficking.

 

Global approaches to Studying Human Trafficking: The Literature

    Perhaps the most comprehensive review of the trafficking literature is the IOM's "Data and Research on Human Trafficking: A Global Survey" (Laczko & Gozdziak, 2005). As Laczko (2005) indicated in the introduction of the volume, the rapid rise in the number of publications on trafficking reflects the mounting national and international concern for human trafficking. They point out that since the mid-1990s, a great number of reports covering individual countries and regions have been produced. These studies have typically tried to examine the entire trafficking process, including the causes of trafficking. Studies also have attempted to describe the recruitment process, the transport of victims, as well as the exploitation of victims/survivors. These studies usually include a description of existing legal and policy frameworks in national case studies. Most reports end with a set of recommendations for further action with the objective to both inform and contribute to the development of counter-trafficking projects and strategies.

    The IOM volume (Laczko & Gozdiak, 2005) also includes chapters on specific regions of the world and issues related to trafficking research methods. The volume also includes a chapter which provides bibliographies for all of the human trafficking literature broken down by the regions of the world.

 

Data Collection Issues and Obstacles 

    There is a general agreement that there is a lack of reliable data regarding human trafficking (Kangaspunta, 2003, Kelly, 2005, Laczko, 2005). The GPAT Report (2006) suggests that this can be traced to a number of factors. Their list begins with the fact that many countries lack anti‑trafficking legislation. Even when countries have legislation in place, laws may only define trafficking for certain practices, like sexual exploitation. In some countries trafficking only applies to the exploitation of women and children. Laws are not always enforced and victims may not be seen as crime victims but rather as smuggled migrants. Countries lack centralized data collection systems. When data is collected, it is often provided by inter-governmental (IGO) or non-governmental agencies (NGO) that assist or repatriate victims. Those numbers only represent a small portion of trafficking victims in countries.

 

Global Estimates versus Trafficking Hotspots

    One major research focus has been the quest to determine the scope of human trafficking world-wide. Since 2002, the U.S. Department of State has started publishing global estimates (2002, 2003, and 2004).  A more recent U.S. Department of State Report (2006) estimated that the number of persons trafficked globally are between 600,000 and 800,000. Many are critical of that approach, (Kangaspunta, 2003, Kelly, 2005, Laczko, 2005) and both Kangaspunta and Laczko use the word "guestimates" when they comment on the efficacy of the global estimate approach. Kangaspunta also raised the question as to whether global estimates serve any serious policy purposes, suggesting that global estimates are vague and cannot serve as a reliable knowledge base for policy formation. She suggested that mapping trafficking "hotspots" can provide valuable information on the nature and context of trafficking. This includes identifying origin, transit, destination countries, the involvement of organized crime in different countries and the main routes used by traffickers. She also suggests the need to monitor the impact of interventions with data that is carefully collected and analyzed, and the use of national as well as regional data. According to her, this approach might yield profiles that can be used for developing regional cooperation in the fight against trafficking. As will be apparent below, this study has adopted some of Kangaspunta's suggestions regarding the need to study "hotspots".

 

Study Background:

The Identification of the Factors that Predict Human Trafficking

     Reacting to the lack of centralized data collection, Bales (2007) assembled a trafficking data set with information collected from around the world. Besides the United Nations World Statistics Pocketbook (1995) Bales used data from other sources, like Amnesty International. Bales also surveyed professionals working on trafficking field and asked them to help refine and correct his estimates for each country. Also included was Bales' own database of slavery and trafficking (see Bales, 2007, footnote 3, 279).

Bales' study (2007) attempted to answer two basic questions: 1) What are the strongest predictors of trafficking FROM a country on the global scale and 2) What are the strongest predictors of trafficking TO a country on the global scale? Bales used regression analysis to identify the factors which predicted the amount of trafficking FROM and TO a country. Bales found six factors predicting trafficking from a country. The most powerful predictor was government corruption, followed by the percent of population under 14 year’s old and infant mortality. In descending order, the other factors were food production, population pressure and conflict, and social unrest. Collectively, these factors explained 57 percent of the variance in trafficking from countries.

    The findings for trafficking TO countries were weaker, with four significant factors explaining 15.5 percent of the variance in trafficking to countries. In order, these factors were percent of male population aged 60 plus, government corruption, infant mortality, and food production.

    Fry (2010, in press) extended Bales' (2007) study by including transit countries to his analysis and he utilized the Global Program against Trafficking in Human Beings (GPATHB) database as his data source. The two studies had used data sets which included different sets of indicators, with some overlap, so it was not surprising that the results were mixed. Both studies identified governmental corruption and the percentage of the population under 14 as the two top predictors of trafficking from a country. Fry’s study identified the percentage of population under 14 and governmental corruption as the primary predictors of trafficking through transit countries, a category not included in the Bales study. Bales reported that the proportion of the countries' population over 60 and corruption were the two strongest predictors of trafficking to a destination country. By way of contrast, Fry found that the Human Poverty Index and the total population measure were the strongest predictors of trafficking to a destination country.

 

The Global Program Against Trafficking in Human Beings

    The most recent report of trafficking in persons: Global Patterns (TPGP, 2006), GPATHB was launched in 1999. The purpose was to help enable governments to respond to trafficking in human beings and smuggling of migrants. GPATHB aims to shed light on the causes and processes of migrant trafficking and smuggling, as well as the promotion of the development of effective responses to those problems. One strategic area is the collection and analysis of data in order to increase the global community's knowledge base, raising awareness to prevent human trafficking and migrant smuggling.

 

The Present Study

    The question which arose from the two studies described above was whether there was a more effective, policy relevant, way to examine the question of the factors that predict human trafficking at the country level. Fry’s study (2010, in press) pointed at the need to look at trafficking hotspots. The other implication supported one recommended in an earlier paper (Fry, 2009), which was the need to utilize the GPATHB database in future human trafficking research. That data base is described below.

 

The Database

    According to Kangaspunta (2003), the global database was designed to collect a wide range of open-source information on trafficking. Information may be qualitative or quantitative and come from official government statistics, reports from research institutes, or from newspaper articles and news agency bulletins. As the methodology section of the GPAT Report (2006) explained, the database includes publicly available information from 113 different source institutions that provides data on 161 countries. The 113 source institutions produce numerous publications, reports, books, articles, journals, and newspaper articles. Most of the relevant data provides details on human trafficking, routes, victim profiles, and the purposes of trafficking.

 

The Dependent Variable

    There are three different sections in the database: country reports, profiles, and trafficking routes. The country profile section includes details about the reported trafficking in countries. These reports are broken down by origin, transit, and destination and the country is ranked from 'very low' (1) to 'very high' (5) on each dimension, origin, transit, and destination. This study's dependent variable, total trafficking was created by creating a summed index from those three dimensions, with a possible range of 3 to 15.

 

Independent Variables

    The GPAT report (2006) also included a range of data collected at the country level, especially the factors that are thought to be the root causes of trafficking. These measures constitute the studies independent variables.

 

The Measures

    The total population of the country or territory is reported according to the most recent United Nations estimation and population under the age of 14 is calculated from the percentage of total population. There are a number of indexes derived from the literature, starting with the net migration rate per 1,000. This is the total number of immigrants minus the total number of emigrants in the year 2000 divided by the total average population of the country in the same year. The rate is expressed as the number of migrants per every 1,000 population in the country. A positive or negative sign indicates whether the country experienced a gain or loss in total population. The Human Development Index (HDI) ranking is a summary measure of human development. It measures the average achievements of a country in three basic dimensions of human development: 1) a long and happy life as measured by life expectancy at birth; 2) knowledge as measured by the adult literacy rate; and 3) a decent standard of living , measured by gross domestic product (GDP) per capita (purchasing power parity) in U.S. dollars. While the HDI measures average achievement, the Gender-related Development Index (GDI) adjusts the average achievement to reflect inequalities between men and women on the following dimensions: 1) a long and healthy life as measured by life expectancy at birth; 2) knowledge as measured by the adult literacy rate and the combined primary, secondary and tertiary gross enrollment ratio; 3) a decent standard of living as estimated by estimated earned income.

    The Transparency International Corruption Perceptions Index (CPI) gathers data from sources that span over the previous three years. All sources provide a ranking of countries and measure the overall extent of corruption in the public and political sectors. Evaluation of the perceived extent of corruption in individual countries is done by non-resident business leaders from developing countries and resident business leaders evaluating their own country.

    The Organized Crime Index (OCI) of the World Economic Forum (WEF) is based on an assessment of the extent to which organized crime imposes costs on business. The index is measured on a scale of 1 (imposes significant costs) to 7 (does not impose significant costs). Since the file did not include a poverty index, the Human Poverty Index (HPI-1) was added. The HPI is a composite index measuring deprivation on three basic dimensions, a long and healthy life, knowledge, and a decent standard of living.

 

Results

    This was an exploratory study and stepwise multiple regression method was used as the first step in the analysis; as variables are brought into the equation in the order in which they contribute to prediction equations. The variable with the highest correlation with the criteria is selected first and then successive variables are added to the model until there is no appreciable difference in the R.

    The stepwise multiple regression procedure was run with total trafficking as the dependent variable. These results are presented in Table 1.

 

Table 1. Stepwise Multiple Regression of Total Trafficking Score.

Model Summary

 

			Adjusted R	Std. Error of
Model	R	R Square	Square	the Estimate
1	.308	.095	.083	2.992
2	.557	.310	.293	2.628

a)        Predictors: (Constant), population under 14

b)       Predictors (constant), population under 14, corruption index

 

ANOVA

 

Model	Sum of Squares	Df	Mean Square	F	Sig.
1              Regression	74.207	1	74.207	8.288	.005a
Residual	707.299	79	8.953
Total	781.506	80
2              Regression	242.646	2	121.323	17.561	.000 °
Residual	538.861	78	6.908
Total	781.506	80

a)        Predictors: (Constant), population under 14

b)       Predictors: (Constant), population under 14, corruption index

c)       Dependent Variable: Total Trafficking Index

 

Coefficients

 

Model	Unstandardized Coefficients		Standardized Coefficients	t	Sig.
	B	Std. Error	Beta
1             (Constant)	9.300	.909		10.231	.000
corruption index	-.089	.031	-.308	-2.879	.005
2             (Constant)	15.912	1.559		10.206	.000
corruption index	-.197	.031	-.681	-5.648	.000
Pop under 14	-..793	.161	-.596	-4.938	.000

a. Dependent Variable: Total Trafficking

 

     Table 1 show that two variables entered the regression equation. These were the percent of the population under 14 years of age and the Corruption Index. These two variables, the percent of the population and the Corruption Index produced an R of .557, R square of .310, and an adjusted R Square of .293. As will be discussed below, these results were almost identical to Fry's (2010, in press) findings on the factors that predict trafficking from origin countries. This suggests that the total trafficking score did not add much in the way of producing new information on human trafficking at the global level.

    As a result, the next task in the analysis was to continue to attempt to separate trafficking hotspot countries from all others. The total trafficking variable is a summed index and when the frequency scores were studied in detail there was a clear break in the data. Twenty four countries scored 10 points or better compared to 137 countries that scored 9 points or less. The mean score for all countries was 6 points and the list of the countries that scored 10 points or more are listed in the Appendix. A new variable, high volume was created and broken down into two categories, those countries that qualified as hotspot countries compared to all others. The next step in the analysis was to examine the relationship between the scores on high volume distinction and the study's independent variables. The t-scores resulting from the comparisons of the countries with scores at 10 or more as opposed to all other countries are displayed in Table 2.

 

Table 2. t-Tests for all Study Independent Variables Broken Down by Traffic Hotspots (n=24) And All Other Countries (n=137)

 

Variable		Ns*	t-test for Equality of Means	df	Significance (2 tailed)
			t
Population	1	18	-3.943	157	.000
	2	79
%of Population under 14	1	23	3.773	155	.000
Under 14	2	133
Net Migration	1	24	-1.311	147	.192
	2	125
Human Development	1	23	-1.333	150	.185
	2	129
Gender Development	1	19	-1.037	23	.302
	2	111
Corruption Index	1	23	1.966	131	.051
	2	110
Organized Crime	1	18	2.084	95	.040
	2	79
HPI	1	23	.813	149	.417
	2	128

 

 

    Table 2 revealed that four of the eight comparisons reached statistical significance. These were total population, the percent of the population under 14 years of age, the Corruption Index and the Organized Crime Index. Total population and the percent of the population under 14 were highly significant, with both t scores over 3.7 (p=.000). The Corruption and the Organized crime indexes were less significant, at about the .05 level and .04 levels respectively.

    The last task was to use the high volume variable as the dependent variable in a stepwise discriminant analysis. This procedure allows the researcher to examine the differences between two or more groups with respect to several variables simultaneously. Those results are presented in Table 3.

 

Table 3. Stepwise Discriminant Analysis Comparing High Volume Trafficking Countries (n=24) with All other Countries (n=137)                                                                                  

Variables Entered

 

Step	Entered	Wilks' Lambda
		statistic	dfl	df2	df3	Exact F
						Statistic	df1	df2	Sig.
1 2 3	population population under 14 Corruption Index corind	.893 ....830 .653	1 2 3	1 1 1	79.000 79.000 79.000	9.429 7.964 13.649	1 2 3	79.000 78.000 77.000	.003 .001 .000

At each step, the variable that minimizes the overall Wilks' Lambda is entered.

 

                         Summary of Canonical Discriminant Functions

                                                Eigenvalues

 

Function	Eigenvalue	% of Variance	Cumulative %	Canonical Correlation
1	.532(a)	100.0	100.0	.589

a  First 1 canonical discriminant functions were used in the analysis.

 

 

Wilks' Lambda

 

	Wilks'
Test of Function(s)	Lambda	Chi-square	df	Sig.
1	.653	33.048	3	.000

 

    Table 3 shows that the discriminant analysis was consistent with the regression findings where three variables entered the stepwise equation: total population, percent of population under 14, and the corruption Index, and collectively these three variables created a canonical correlation of .589 and a Wilks' lambda of .653. The Organized Crime Index, significant when the bivariate relationships were examined in Table 2, did not enter the stepwise discriminant equation.

 

Discussion

    The question was whether the trafficking hotspot designation was a useful way to assess the factors that predict human trafficking at the country level. There is no clear answer to that question in the context of the results presented in this paper. However, there is some support for the efficacy of the trafficking hotspot designation, but it is clear that much work needs to be done, specifically in the area of data collection.

    Above designed Tables 2 and 3 show that there are significant differences when countries with the high volume designation are compared to all other countries. The bivariate statistics revealed that four of the eight comparisons reached significance and three of them, total population, percent of the population under 14 years of age, and the Corruption Index were identified as variables that discriminated between high and low volume trafficking countries.

    From the analysis it is also apparent that there is the need to improve data collection and use that additional information to supplement the GPAT database (Note the number of missing cases in Tables 1, 2, and 3). Some of the missing information is due to the fact that several new countries have emerged over the last several years, with Kosovo a clear example. Kosovo issued its unilateral declaration of independence on February l7, 2008, and it is not surprising that data is missing on all of the study’s independent variables. There are other countries in similar situations but some countries are not included on some of the indicators because they do not appear to be relevant to that country

    The implications of the missing data situation for future research are twofold:  1) there is the need to update the GPAT database where possible; and 2) the need to generate better comparisons for high volume countries, perhaps via a matching process where high volume and other countries are matched on other indicators, especially this study’s independent variables.

    Several areas of future research are suggested by the findings presented here. The first is the issue of governmental corruption.  Kangaspunta (2003) pointed out the need to monitor the impact of interventions with data that is carefully collected and analyzed, and the use of national as well as regional data. The UN Office on Drugs and Crime has provided that kind of opportunity with its Global Program against Corruption. The Program's website (http://www.undoc.org/undoc/en/corrotion/links.html)  lists 25 links to national anticorruption agencies, including Thailand, and the Philippines: countries that appear in the Appendix as high volume trafficking countries. The underlying question here is whether these national efforts have any effect on human trafficking within their borders. The website indicates that a data base is being updated and will be available by the end of 2008. The availability of data that supplements the Corruption Index used in this study would be a welcome addition to studies which attempt to examine the effects of corruption on human trafficking.

The list of high volume trafficking countries provides another instance where a better approach to future research is required. The list of high volume trafficking countries identifies the Balkans as the region of the world with the highest concentration of trafficking hotspots. Of the countries commonly identified as the Balkans, only Croatia and Greece are not on the high volume list, and both of them had scores of 9 on the Total Trafficking Index. All of the other countries in the Balkans are included as high volume countries, Albania, Bulgaria, Bosnia-Herzegovina, Kosovo, Macedonia, and Serbia-Montenegro and two of its neighbors were also found to be high volume trafficking countries, Turkey and the Ukraine.

    The fact that the Balkans was identified as the epicenter of hotspot countries came as no surprise. This region has received a great deal of attention from international agencies attempting to track human trafficking activity and enumerate the number of victims of trafficking located in the Balkans Region. Laczko and Gremegna (2003) described the IOM database on Trafficking in Southeastern Europe that was implemented in the Balkans in 2002 and provided the list of countries and provinces where data was collected, all located within the Balkans region. UNICEF also produced an inventory of the situation and responses to human trafficking in the region (Gronow, 2000). Lindstrom (2006) provided an overview of anti-trafficking policy in the Balkans describing it as “a Transnational policy developed by, diffused and implemented with the direct involvement of global policy and actors and coalitions at or across the international, national, or local levels of governance”. The paper briefly discusses the process of diffusion and adoption of anti-trafficking in the Balkans and concludes with an examination of some unintended consequences of implementing anti-trafficking initiatives. The paper concludes with some recommendations to improve the overall anti-trafficking system in the Balkans. The point to be made here is that numerous agencies are involved in the Balkans and most have developed the type of interventions that Kangaspunta (2003) indicated are in need to be monitored to assess their impacts on human trafficking. The list of possibilities for future research appears to be endless.

 

Conclusion

    The conclusion of this study is that the search for and identification of trafficking hotspots is an effective way to begin future research on human trafficking. Certain factors do discriminate between high volume trafficking countries and other countries. Future research needs to focus on interventions designed to attack those factors, with corruption the clear front runner. There is a need for cooperation among agencies working in the trafficking area in terms of their data collection, intervention strategies and efforts designed to address the plight of victims. Policy perspectives appear to be based on ideologies instead of sound empirical findings and that needs to change in order for the international community and its individual states to effectively address the human trafficking plague at any level.

 

References

Bales, K. (2007). What predicts human trafficking? International Journal of Comparative and                                      
Applied Criminal Justice, 31(2), 269-279.

Fry, L. (2010, in press) Continuities in the prediction of human trafficking: A research note. International Journal of Comparative and Applied Criminal Justice. (in press)

______ (2009). Global human trafficking: Patterns, informational needs and prospectus.  In K. Jaishankar (Ed.), International Perspectives on Crime and Justice (pp. 61-80). New Castle, UK: Cambridge Scholars Publishing.    

Gallagher, A. (2002). Trafficking, smuggling and human rights: Tricks and  treaties. Forced Migration Review, 12, 25-29. (Electronic Version).

Gronow, J. (2000). Trafficking in human beings in southern Europe. Area office for the Balkans, UNICEF.

Kangaspunta, K. (2003). Mapping the inhuman trade: Preliminary findings of the database on trafficking in human beings.  Forum on Crime and Society, 3(1&2) 81-10.

Kelly, L. (2005). You can find anything you want: A critical reflection on research on trafficking in Europe.  In F. Laczko & E. Gozdziak (Eds.),  Data and research on human trafficking: A global survey (pp. 235-265). Geneva, Switzerland: International Organization for Migration. 

Laczko, F. (2005). Introduction. In F. Laczko and E. Gozdziak (Eds.), Data and Research on Human Trafficking: A Global Survey. (pp. 5-16). Geneva, Switzerland. International Organization for Migration.

________ & Gozdiak  (Eds.), (2005). Data and research on human trafficking: A global survey. Geneva, Switzerland. International Organization for Migration.

_______ & M. Gramegna (2003). Developing better indicators of human trafficking. Brown Journal of World Affairs. X(1), 179-194.

Oxman-Martinez, A., Martinez, A., & Hanley, J. (2001). Human Trafficking: Canadian Government Policy and Practice. Refuge, 19(4), 14-23.

United Nations (1995) World Statistics Pocketbook. New York.

  _________ (2000) Protocol to Prevent, Suppress and Punish Trafficking in Persons, especially Women and Children, supplementing the United Nations Convention Against Transnational Organized Crime. United Nations.

United Nations Office on Drugs and Crime (UNODC) (2006). Trafficking in Persons: Global Patterns.

United States State Department (2002).  Trafficking in Persons Report. Washington, DC.

_________ (2003)  Trafficking in Persons Report. Washington, DC.

_________(2004)   Trafficking in Persons Report. Washington, DC.

_________(2006)  Trafficking in Persons Report. Washington, DC

 

APPENDIX

 .Figure 1. Countries Identified as Trafficking Hotspots and Their Total Trafficking Index  Score (n=24)

 

Country______________________________________Score____________________

Thailand                                                                                  15

Albania                                                                                    13

Bulgaria                                                                                   13

Philippines                                                                               13

Algeria                                                                                      12

Czech Republic                                                                         12

           Hungary                                                                                   12

           Turkey                                                                                      12

          Ukraine                                                                                     12

          Bosnia-Herzegovina                                                                  11

             Myanmar                                                                                11

             India                                                                                       11

             Romania                                                                                 11

             Russian Federation                                                                11

                Kosovo                                                                                 11

             Benin                                                                                     10

             Hong Kong                                                                             10

             China                                                                                     10

             Italy                                                                                        10

             Kazakhstan                                                                            10

             Macedonia                                                                             10

             Mexico                                                                                   10

             Nigeria                                                                                   10

             Serbia-Montenegro                                                                10