Human Trafficking: Recent Empirical Research
By Sheldon Zhang and Ronald Weitzer
Since our special issue on human trafficking in The ANNALS (Volume 653 in 2014), we are happy to report that there has been significant growth in both the quantity and quality of research on human trafficking. For a very long time, varied estimates or unsubstantiated claims filled much of the policy discourse as well as the human trafficking literature. So much so that there were concerns over the legitimacy of a social movement that was fueled mostly by moral fumes over a social phenomenon yet to be documented and substantiated by empirical evidence. From 2000 to around 2015, practically all published review articles on human trafficking research pointed to the general lack of data and raised concerns over the prevailing policy initiatives, given poor empirical guidance.
Things are improving. Significant progress has been made in the realm of measurement and prevalence estimation techniques, as seen in a growing number of empirically based prevalence studies. Not surprisingly, much of the progress in methods and prevalence estimations is accomplished overseas, funded mostly by the US government and some by international organizations. For instance, currently a group of researchers is embarking on a collective effort to achieve greater consistency in measuring human trafficking in several parts of the world. Funded by the US State Department, a total of six teams from different universities and international organizations joined force to conduct prevalence-estimation studies in different countries involving different victim populations, including child sex trafficking in Brazil, forced labor in the agricultural sector in Brazil, forced labor in the fishing industry in Costa Rica, domestic servitude among migrant workers in Tanzania, domestic servitude in Morocco, forced labor in brick kilns in Pakistan, and domestic servitude in Tunisia.[1] Unfortunately, such a large scale or multitude of studies emphasizing empirical data collection have thus far not taken place inside the U.S itself. Funding for large scale data collection on human trafficking activities here has thus far remained anemic, probably due to low funding priority relative to other pressing social issues (e.g., gun violence and police brutality).
Still, the bottom line is: With much progress in empirical data collection and estimation methods, studies continue to affirm the widespread problem of human trafficking or forced labor of various forms.
One glaring problem remains: We know little about what works in preventing trafficking victimization or protecting victims. Most of the world’s anti-trafficking programs are aligned with the 3P model (prevention, protection, and prosecution) promoted by the US government. A few years ago, a study by the United Nations University calculated that more than $4 billion had gone into fighting human trafficking, of which the US accounted for 60%, and there are no shortage of programming guides and recommendations for effective anti-trafficking programming. However, the reality remains the same: there is practically no knowledge of an effective model that victim advocates and law enforcement authorities can learn from. We know so little about what works in the field. This is partly because of the inherent challenges of evaluating social engineering projects, and partly because few funding agencies demand rigorous evaluation of their funded programs. There is simply little accountability. Instead, evidence of program efficacy is typically based on anecdotal stories provided by the program administrators, or testimonials from supposedly grateful “victims.” Whenever subjected to the most rigorous evaluation strategies, such as randomized controlled trials, few (if any) such programs can withstand the impression that “nothing works”.
Things are gradually improving, however. For instance, the US State Department has in the past few years demanded an RCT or quasi-experimental design be considered whenever possible for proposing anti-trafficking intervention programs. We hope that it is just a matter of time before the trafficking research and programming fields become more professionalized, with the ultimate goal of identifying the places and populations where people are at most risk of victimization.
[1] Details about the PRIF can be found at: https://apries.uga.edu/prevalenceforum/