By HTC Staff: Catie Fowler, Jocelyn Iverson and Leanne McCallum
In the world of academia, it is important to remember that even educated guesses are just that– guesses. A recent report published by the University of Texas at Austin, entitled Human Trafficking by the Numbers: The Initial Benchmark of Prevalence and Economic Impact for Texas, opened its study with this quote:
“Harriet Tubman did not wait around for a proper measurement of how many slaves were in the South. Neither should we. The work needs to be done while we try to measure it.”
-Timothy McCarthy, University of Texas
Harriet Tubman was not a researcher- she was an activist. She based her decision to act solely on her understanding that the presence of any amount of slavery at all was unacceptable. She didn’t wait for accurate numbers to justify her mission to abolish slavery. She didn’t conduct literature reviews or try to generate prevalence data. It was Tubman’s job to react, but it is the job of academics to carefully apply academic rigor and integrity to their examination of pivotal social issues.
This report utilizes a broad spectrum of guesswork to estimate Texas’ trafficking prevalence and the net economic impact of trafficking in the state. The Human Trafficking Center commends any research that seeks to bring to light the real impact of human trafficking, however such attempts must be conducted not only with good hearts, but with sound methodology and reliable data. After all, how can good policy or services be crafted if they are based on faulty research?
This is the reason why we are providing a critique of this study. Not because we believe that this report is ill-intentioned or that it is purposefully misrepresenting its findings, but because we worry these numbers could be misinterpreted. Anti-trafficking advocates could utilize the numbers in this report without understanding some of the assumptions and misunderstandings that informed its conclusions. That is to say, if activism is to be based on numbers, those numbers should be founded on sound methodology and a clear explanation of those numbers’ limitations.
The University of Texas at Austin (UTA) may have chosen to present their preliminary estimates on human trafficking to gain greater media publicity around the issue of trafficking. Publicity is not inherently harmful, but providing inaccurate or misleading data can be. The beginning of the report includes an infographic, which tells its readers that there are an estimated 313,000 victims of trafficking in the state of Texas and details the victim breakdown and economic impacts of labor versus sex trafficking. Each number is preceded by the clarification that it is an estimate or approximation, but nowhere on the infographic does it state that these numbers are preliminary or that they exclude large portions of potential victims. Throughout the report, the researchers acknowledge that their data is yet to be finalized, however, in the abstract and data infographic, this is not made clear. Our concern is that the majority of non-academic readers will take the information presented in the outset as solid fact. These numbers become increasingly dangerous as they run the risk of being copied and circulated as truth—without the necessary disclaimer that they are only rough estimates. A primary example of this is the release of an article by KXAN, an Austin, Texas NBC affiliate with the title, “Study: Nearly 70,000 minors are sex trafficked in Texas.”
Defining terms and populations is a key component of research design. We were surprised to find that this research project does not narrowly define key terms like human trafficking, labor trafficking, or sex trafficking. The narrative does imply that it defines trafficking as a “broad umbrella term that encompasses trafficking in persons of any age, nationally or internationally, for sex or labor purposes” in its introduction, but the report does not indicate how this term was selected. The narrative subsequently suggests that it utilizes the definitions from the Trafficking Victims Protection Act (TVPA) but does not clarify how. Though the researchers do eventually define labor trafficking deeper in the report, they do not cite where they got the chosen definition from, if they crafted it, or how they crafted it.
Even if researchers do intend to use the TVPA definition, this becomes confusing because the research project strays away from the TVPA definition of domestic minor sex trafficking (DMST). The researchers acknowledge the criticism of the term, and the limitations and problems associated with it, but continue to use it anyway. The report includes people age 18-20 in their DMST data, even though the TVPA definition of minor sex trafficking only includes people age 18 and under. Because the report lacks definitions on forms of trafficking and trafficked populations, this gives the audience a muddy concept of who the victims are and what trafficking entails.
Formulating a methodology for calculating human trafficking prevalence is a difficult, if not impossible, task. It has been done by many researchers, but usually under the premise that there are gaps and assumptions within their methodology. This does not necessarily mean that these data are invalid or shouldn’t be circulated within the anti-trafficking community, but it is vital that they are transparent about the complexities of the project. The general public, funders and government agencies all want to know “how big the problem is” and thus pressure researchers to come up with ways to give them numbers. It is up to the research community to push back as to why these numbers can be so misleading or fraught with uncertainty.
The UTA study uses a “victimization rate” to estimate prevalence. The DMST victimization rate is calculated using a survey distributed by the researchers, to which only 19 service providers responded. Among the pooled responses of the service providers, 1,877 of the total 7,484 clients were deemed to be trafficking victims, which is the equivalent of 25.1% of the population. This rate is then extrapolated to communities identified as at a high risk for DMST, in this case children and youth who have experienced abuse, homelessness or have been a part of the foster system. The prevalence of labor trafficking was calculated in a similar way, utilizing a victimization rate to assert numbers of victims within at-risk industry segments. However, since the pilot labor study’s rate was rather high, the researchers decided to use the lower victimization rates from two completely separate studies.
The first problem with such a methodology is that it only applies to certain visible populations and does not attempt to encompass all victims of human trafficking. There is also no clear path laid out as to how the study plans to expand its methodology to adult sex trafficking and other victim categories. At some points the study acknowledges that it is attempting to measure a hidden population, yet it utilizes population numbers published mostly by government agencies which do not account for the hidden portions of these populations. Second, the study seems to come to these rates in a superficial manner, either utilizing the information from service providers who happen to respond to their survey or utilizing prior research with varying methodologies and assumptions. The study then attempts to justify the rates used by asserting that they are conservative estimates. Using a conservative estimate does not necessarily lead to higher accuracy, and is not the most methodologically sound argument.
In addition to the victimization rate and prevalence assertion, the study endeavors to calculate the economic impact of both minor sex trafficking and labor trafficking on the State of Texas. Although the study puts forth big, flashy numbers for each, the two models measure dramatically different economic impact parameters. The economic impact of DMST encompasses costs for the state throughout victims’ lifetimes, while the economic impact of labor victims is presented as an annual calculation and references costs to the victims themselves. The problem is that these numbers are presented as though they are somehow parallel, or both speak to the same issue.
A “Net Present Value”(NPV) of $83,125 is calculated for the lifetime expenditures borne by society related to each victim of DMST under the prevalence portion of the study. It is not clear exactly how this number is reached, although a cost-benefit analysis from a previous study is referenced. It seems to include both the service and prosecution costs which may relate to each child victim, but it is never broken down in a line item fashion equaling the number asserted. In reality, many victims may not receive services or participate in the prosecution of their perpetrators. It is uncertain if this is taken into consideration when calculating the NPV as it should be. It is dubious to assume that the wide range of experiences of trafficking victims can be averaged out to come to a standard economic impact of each.
The economic impact of labor trafficking on victims in Texas is calculated in a way that relies on more assumptions than facts. The expected wages for victims of labor trafficking are estimated using the Department of Labor’s “Adverse Effect Wage Rate” of $11.15/hr for H2-A workers and a standard 2080-hour work year. The results of the pilot labor study are then used to assert that workers are denied 11% of their wages due, based solely on the perceptions and reporting of workers surveyed in the study. This methodology does not take into account the variation among labor trafficking victims in wages promised, hours worked, or degree and length of exploitation. To boil down the economic impacts of labor trafficking on victims to such generalities is highly misleading and problematic.
One of the biggest holes in the logic of this research project is the numerous acknowledged (and unacknowledged) conflicts in the narrative. We found many assertions that contradicted other information foundational to the research, or information that the report quietly acknowledged was problematic but continued to use anyway. We applaud that this narrative acknowledged some of the critiques and faults of the information they were using, but we find it problematic that these critiques were buried deep in the report so the drawbacks weren’t immediately visible to the reader.
One example of this is the use of the term DMST. As the study mentions, “the literature promotes the idea that there is a possible difference between individuals who are involved in trafficking versus those who are involved in commercial sex. This is a conundrum to be explored because all youth who are involved in commercial sex are automatically deemed trafficking victims. Additionally, methodological limitations and lack of theoretical application in commercial sex literature involving youth prevent this information from providing a comprehensive picture of the risks and experiences of youth.” It then subsequently uses a different study, that is too dated to be considered a viable standalone source, to assert that most sex workers have experienced child sexual abuse and lack understanding of their trauma (and therefore are sex trafficking victims). This is just one example of a trend throughout the report of using conflicting information. It is important to acknowledge conflicting information in order to show the depth and breadth of literature available on a topic, but actively utilizing the conflicting information to support different portions of its various conclusions undermines the report’s credibility.
There are many misconceptions about labor and sex trafficking in the mainstream anti-trafficking narrative. This study makes some inherently problematic assumptions that are common, but not acceptable from a study of this magnitude. For example, labor trafficking victims are not generally trafficked for an entire lifetime, or even for a full year. According to Kevin Bales, a prolific expert on human trafficking, conditions of labor trafficking tend to happen for short periods of time periodically throughout a victim’s life, ranging drastically in duration. The victim will often return to exploitative labor conditions later in life, but not in a consistent manner. Although the researchers acknowledge that “…we can only estimate how many hours the average victim works under conditions of modern slavery (we know that most episodes of victimization last only a few days or weeks and not months or years)… and applied to a normal 2080-hours worked per year” this is problematic because it relies on an estimate that labor trafficking victims perceived they had not been paid 11% of their wages.
Even if there was a way to determine the average duration of a trafficking experience, this further de-legitimizes the study’s claim to have sound, methodological measures of how much traffickers have exploited from labor trafficking victims in Texas. There is no specific age of entry in trafficking, length of life, duration of exploitation at the hands of traffickers, or average salary trafficked people receive. Labor trafficking can include: wage garnishing, non-payment, or underpayment of a worker per their agreement upon entering that particular job. In creating a single number to represent the money taken from all victims based on an aggregate, this study fails to recognize the broad range of experiences victims of trafficking experience. This also highlights how problematic it is that the study never defines ‘labor trafficking’, because a definition of labor trafficking could shed light on the scope of trafficking victim experiences being encompassed in that term. Exacerbating this issue, there are only two paragraphs exploring the vulnerabilities associated with forced labor, but there are several pages worth of background specifically for DMST.
The UTA study could have broad implications for human trafficking data collection and future policy, despite several major flaws in its research design and methodology. We provide this critique for two reasons. The first is to caution well-intended anti-trafficking stakeholders against jumping to conclusions about this data without understanding the implications of its faulty methodology. The second is to help the University of Texas at Austin reform its study and methodological process, so that the data produced in Phase 2 can be a more accurate representation of what actual trafficking conditions may look like in Texas.
We applaud the researchers for their foray into the difficult venture of measuring the impacts of human trafficking in their state, and we hope they will find our critique both valuable and thought-provoking. We too want to support the modern Harriet Tubmans of the anti-trafficking movement. We believe the best way to do this is through rigorous, methodologically sound research that is transparent about the difficulties of collecting data on this crime and accurately represents the populations we aim to serve.
Photo by fritsdejong via Pixabay
The Human Trafficking Center, housed in the University of Denver’s Josef Korbel School of International Studies, is the only two-year, graduate-level, professional-training degree in human trafficking in the United States. One way graduate students contribute to the study of human trafficking is by publishing research-based blogs. The HTC was founded in 2002 to apply sound research and reliable methodology to the field of human trafficking research and advocacy.
Founded in 1964, the Josef Korbel School of International Studies is one of the world’s leading schools for the study of international relations. The School offers degree programs in international affairs and is named in honor of its founder and first dean, Josef Korbel.
Print This Post