The covert nature of human trafficking has long complicated efforts to tally trafficking victims and pinpoint their location. So disparate and unreliable were past estimates of trafficking victim totals, the U.S. Department of State stopped reporting those numbers altogether in 2016. The agency concluded, “Given the complex nature of human trafficking, it is difficult to amass reliable data to document local, regional and global prevalence.”
In the face of that inherent difficulty, a recent NIJ-funded study of migrant farmworkers in North Carolina used a novel combination of analytical techniques to pull back the curtain on the presence of labor trafficking in that state’s agricultural sector. To the researchers, the results were surprising. Among the key findings were:
- Overall, just over one quarter of interviewed migrant workers experienced some form of employment abuse; nearly 18% reported incidents that could constitute labor trafficking; and 22% reported lesser forms of labor abuse and exploitation.
- Only 17% of more than 400 interviewed farmworkers reported they were undocumented.
- For those without work visas or other legal documentation, however, undocumented status was a strong indicator of exposure to some forms of abusive treatment of those workers.
- Farmworkers forbidden to have visitors in their dwellings were more likely to experience some form of victimization.
The researchers qualified those findings, however, by noting study method limitations as well as the possible hesitancy of migrant workers to candidly report their documentation status to research staff.
The research team from San Diego State University and RTI International set out to generate reliable measures of the prevalence of labor trafficking among migrant farm laborers in North Carolina. The study was limited to one state and one labor sector — agriculture. The team noted that multijurisdiction, multisector studies often encounter barriers to precise prevalence estimation.
The team’s scientific method of estimating the migrant labor population improved upon prior research on hidden populations by innovatively combining analyses of census blocks and corresponding agricultural tract statistics. The researchers saw this approach as better suited to a geographically isolated population — discrete clusters of migrant workers spread across large swaths of a state.
More than 90% of the interviewed farmworkers were male and born in Mexico, and 99% were primarily working in agriculture or crop farming in North Carolina.
A High Percentage of Migrant Workers Held Visas
Notably, the high percentage of documented workers found in the study suggests that most of North Carolina’s migrant farm laborers could enjoy significant legal protections from abuse. Of the 327 migrant workers who told interviewers they possessed legal documentation upon entering the United States (83% of the total sample), 307 had temporary H-2A work visas.
The research report emphasized that, in North Carolina, a union contract with the state’s growers association protects visa-holding laborers from retaliation, ensures their right to appeal unjust firings, and gives them a grievance mechanism. “The importance of documentation status and temporary work visas cannot be overstated,” the researchers reported.
The study’s finding that 17% of the farmworkers sampled were undocumented sharply contrasted with the same team’s finding of a 42%-undocumented rate among workers sampled in 2012, but the newer report cautioned that the older study used a different sampling method.
The researchers offered two potential reasons for the apparent sharp drop in the undocumented status of interviewed migrant farmworkers in the later study:
- Conditions have improved over time, with more migrants receiving visas and more growers requiring them.
- Workers may have become less comfortable sharing their documentation status, and so may not have been forthcoming in their interviews.
Documentation Status a Strong Predictor of Abuse
The team found documentation status to be a strong predictor of abuse, both in this study and the prior study of trafficking prevalence. “[U]ndocumented workers are far more vulnerable to abuse and exploitation,” the North Carolina-focused report said.
The study’s finding of a correlation between a particular grower’s ban on visitors to a farmworker’s lodging and some form of elevated victimization was important, the team concluded, because it suggested that some factors could be early red flags of more serious violations. Thus, a ban on visitors to a migrant worker’s lodging could have implications for legal and health outreach professionals charged with aiding at-risk workers.
The finding that a statistically weighted 27.72% of all sampled farmworkers reported experiencing some form of abuse was a downward departure from most of the literature. For example, as the researchers noted, their 2014 study on migrant farmworkers in North Carolina had found that fully 45% of sampled workers indicated they had experienced some form of labor exploitation. But the team’s new findings were very close to another notable study of a migrant population in San Diego, the report noted. That study, also done by some of the same researchers, found that 27% of sampled migrant farmworkers reported experiencing abuse. (The study in San Diego also found that agriculture had the lowest reported incidence of trafficking violations, compared with other employment sectors.)
The newer, North Carolina-focused study used core items from the San Diego-focused study instrument, including much of its interview questionnaire. It sought to secure a better statistical handle on trafficking prevalence, however, by deploying a different sampling method better suited to estimating hidden populations spread out in socially isolated clusters. In the end, however, the North Carolina study methodology proved not to be as productive and efficient as planned, the researchers reported.
Research Challenges Particular to Labor Trafficking
As noted above, researchers have long encountered challenges in identifying and quantifying human trafficking in general. The challenges are even more pronounced in studies focused on labor trafficking, including farm labor trafficking. Those special challenges have included:
- Empirical Anemia: Empirical estimates of labor trafficking are scarce, as noted by such agencies as the United Nations Office on Drugs and Crime, the International Labor Organization, and the International Organization for Migration. Historically, labor trafficking gained scant attention relative to sex trafficking, the North Carolina migrant worker research team noted. Yet most human trafficking does not concern sex.
- Definitional Ambiguity: Defining trafficking is difficult, with a range of definitions adopted in the field. Historically, a number of researchers failed to distinguish between sex trafficking and labor trafficking. Conflation of terms such as modern-day slavery, forced labor, and human trafficking makes the discourse even less precise.
The research team cautioned that the North Carolina study encountered limitations influencing the results, including failure to realize certain study design goals. The study used an overlay of census tracts and agricultural data to identify potential migrant farmworker dwellings and then dispatched research staff to observe all dwellings and estimate the likelihood of their occupancy by migrant workers, using predetermined criteria. To reach the study target of 400 interviews, the research team sought to interview occupants of all “likely” farmworker dwellings and of all 717 “probable” farmworker dwellings, but it still fell short of the target. As a solution, the team added adjacent census blocks in order to achieve a final total of 404 interviews.
The addition of clusters outside of the original study design complicated the statistical analysis needed to generate trafficking prevalence estimates. In that respect, the team reported that it found the original sampling “frame,” or design, to be neither productive nor efficient.
Although the sampling strategy failed to generate sufficient leads for the interviewers, the team said, “we still feel that this method is more productive than alternative sampling techniques, including respondent-driven sampling.”
Small sample size was another limitation, the team reported.
About This Article
The research described in this article was funded by NIJ award 2013-IJ-CX-0047, awarded to San Diego State University. The article is based on the report “Labor Trafficking in North Carolina: A Statewide Survey Using Multistage Sampling” (2019), by Sheldon Zhang, Kelle Barrick, Brian Evans, Ryan Weber, Joe McMichael, Paul Mosquin, Kyle Vincent, and Derek Ramirez.
Sidebar: Research on Labor Trafficking in San Diego
The San Diego geographical study of migrant laborers was led by the researchers who would later direct the North Carolina-focused farm laborer study. NIJ supported both efforts. The goals of the San Diego study were to (1) provide sound estimates of the prevalence of trafficking victimization among unauthorized migrant laborers in the area and (2) investigates types of trafficking victimization — the study report observed that little was known at the time about how migrant workers were being victimized.
The San Diego-based study targeted the six largest labor sectors where unauthorized workers were most likely to find work: agriculture, construction, landscaping, janitorial/cleaning services, food processing, and manufacturing. Construction had the highest rate of reported trafficking (35%) and abusive labor practices (63%), with agriculture having the lowest reported rates for both trafficking (16%) and abusive labor practices (27%).
The North Carolina-focused research on migrant farm labor had a goal in common with the San Diego area study: to produce reliable estimates of migrant worker victimizations. The studies similarly sought insights on largely “hidden populations” of migrant labor, but they diverged in their methods of capturing the target population.
The San Diego geographical study used a sampling method known as respondent-driven sampling (RDS). The RDS method can work where random sampling of a target population is impossible because not all members of that population are accessible to researchers. RDS instead targets an initial group of recruited study participants who then recruit other participants through their social networks. New recruits receive marked coupons from their recruiters, and they redeem the coupons with the study managers. The method is designed to spread study participation through the social network of all participants, allowing researchers to better estimate the hidden population.
In the North Carolina study, however, the research team calculated that RDS would be ineffective because migrant farmworkers in the state, unlike a more closely networked and concentrated hidden population, are socially isolated as clusters in different farm labor camps. Respondent-driven sampling relying on social networks would be less effective in that circumstance, the team concluded.
[note 1] As of July 15, 2020, principal investigator Sheldon Zhang was on the faculty of the University of Massachusetts Lowell. Co-principal investigator Kelle Barrick was with RTI International, Inc.
[note 2] The prior research was reported here: Sheldon Zhang, “Looking for a Hidden Population: Trafficking of Migrant Laborers in San Diego County," Washington, DC: U.S. Department of Justice, Office of Justice Programs, National Criminal Justice Reference Service, November 2012, grant number 2009-IJ-CX-0011, NCJ 252521.
[note 3] Kelle Barrick et al., “Labor Trafficking Victimization Among Farmworkers in North Carolina: Role of Demographic Characteristics and Acculturation” International Journal of Rural Criminology 2 (2014): 225-243 quoted in Sheldon Zhang et al., “Summary Overview, Labor Trafficking in North Carolina: A Statewide Survey Using Multistage Sampling” Washington, DC: U.S. Department of Justice, Office of Justice Programs, grant number 2013-IJ-CX-0047, January 2019, NCJ 252521, p. 19.
[note 4] Sheldon Zhang et al., Estimating Labor Trafficking Among Unauthorized Migrant Workers in San Diego, Annals of American Academy of Political and Social Science, 2014, DOI 10.1177/0002716213519237
[note 5] Sheldon Zhang, “Looking for a Hidden Population: Trafficking of Migrant Laborers in San Diego County” Final report to the National Institute of Justice, grant number 2009-IJ-CX-0011, November 2012, NCJ 240223, San Diego, CA: San Diego State University, pp. 4-8, 26.
[note 6] Sheldon Zhang, “Looking for a Hidden Population: Trafficking of Migrant Laborers in San Diego County” Final report to the National Institute of Justice, grant number 2009-IJ-CX-0011, November 2012, NCJ 240223, San Diego, CA: San Diego State University, pp. 4-8, 26.