Wage Theft and Work Safety: Immigrant Day Labor Jobs and the Potential for Worker Rights Training at Worker Centers

Immigrant day laborers routinely experience exploitative behaviors as part of their employment. These experiences are understood in the context of their immigration histories and in the context of their long-term goals for less precarious labor and living situations. Using mixed methods, over three data collection periods in 2016, 2019, and 2020, we analyze the work experiences of immigrant day labors in Houston and Austin, Texas. We report how workers judge precarious jobs and respond to labor exploitation in an informal labor market. We also discuss data pertaining to a worker rights training intervention conducted through a city-sponsored worker center. We discuss the potential for worker centers to be a convening and remediation space for workers and employers. Worker centers where immigrant day labors meet employers offer the potential for informal intervention into wage theft and work safety violations, by formalizing the context where laborers are hired.


Introduction
This essay examines day laborer experiences of precarious work in Texas through a series of three case studies. Following a participatory action research (PAR) [1] approach that uses an emergent design, [2] we report on the ways workers perceive precarity in their job choices and the way they respond to that precarity. We end with a discussion about the potential of worker centers as sites for intervention into wage theft and worker safety, through worker rights education.
We used three data collection events to examine the overarching social problem of labor exploitation, with wage theft and safety being the most common forms of exploitative abuse endured by low wage laborers. We started data collection in 2016 (by the second and fourth authors) seeking to improve policy makers' understanding of the frequency of labor tra cking and exploitative behaviors that day laborers endure in Texas. [3] The authors conducted a second study in 2019 (reported on in this essay) that examined day laborers' decision making processes in Houston, Texas (TX), when they were seeking work in post-Hurricane Harvey conditions and during increased presence of Immigration and Customs Enforcement (ICE).
The third case study (by the authors and reported on in this essay) was conducted in 2020 among workers who frequented a worker center in Austin, TX, to examine how workers utilize such a community resource when seeking precarious work and how, in the context of an experimental manipulation, an educational intervention might empower workers when evaluating options available to them when exploited during employment. This experimental manipulation also assessed the propensity for workers to share information among themselves as an example of a reciprocal form of group cooperation. It is important to note that the nal case study was conducted during COVID-19 and the methods used were approved by the second author's institutional review board. These modi ed protocols protected both day labors and the research team, but also constrained the amount and kinds of data collected. Although precarity seems to be an emerging character of work in recent decades, particularly in Western Europe and the United States, labor historians argue that precarious work has been the norm for human labor for most of the world's history. [1] Two de nitions of precarious work, include Standing's precariat social category, de ned by seven forms of labor security [2] and Cranford et al.'s, continuum of security along four criteria. [3] For this study, we use Kalleberg et al.'s de nition, "employment that is uncertain, unpredictable, and risky from the point of view of the worker." [4] Kalleberg and Hewison [5] conceptualize precarity as a process, instead of a continuum or a category, pointing to the relationships between labor, capital, and state which produce precarious work. This de nition of precarity is most appropriate for our study because it is de ned by the worker's perception of precarity and opens a discussion into the level of formality or informality of work according to government law and its enforcement, which is a critically important aspect for the immigrant workers we interviewed. Informal labor describes jobs that do not offer standard terms, conditions, and bene ts according to state law because the law does not pertain to these jobs or because the law is not enforced. [6] Labor scholars argue that precarious labor is, in general, distinct in the contexts of the Global North and the Global South, particularly due to variations of local economy and national political structure. Whereby the criteria used to de ne labor as precarious in the Global North could be perceived as standard labor in the Global South. [7] This dichotomous comparison is critical to understanding the production of workers for precarious work in the Global North. By using a de nition of precarity, from the worker's standpoint, [8] production of workers for precarious work could be explained by workers perceiving the precarity of work as less important than an alternative situation, like life in another country or inability to immediately provide for themselves or dependents. Campbell et al. [9] found that Italian temporary migrant workers to Australia did not report when they were paid lower than minimum wage requirements because underpayment of wages was less important to the workers than their long-term migration goals of working in a better labor market than they had experienced in Italy.
While formality of labor exploitation under the state varies in different contexts, labor exploitation has always existed and been economically justi ed through identifying and differentiating categories of people. For our purposes, exploitation refers to a situation whereby taking advantage of another entity, the actor doing the exploitation gains more than they deserve in the interaction and the exploited entity gets less than they deserve. [10] The distinction between the de nitions of precarity, (in)formality, and exploitation are important for this research because the day laborers we interviewed did not always perceive precarity in their informal work, even if the wage theft they experienced was exploitative because they viewed it in the context of their migration histories. Moreover, consistent with general strain theory, [11] anecdotally, some the employers who were doing the exploiting through wage theft, had also experienced exploitation through production in the supply chain. Conceivably, some speci c situations, like workplace injuries or workplace safety violations, would likely not always result in the employer getting more than they deserve from taking advantage of a worker.
All workers dependent on low-wage jobs are at a higher risk of labor exploitation than workers with job security and living wages [12]. However, undocumented workers (speci cally) and migrant workers working in short-term, low-wage jobs are even more likely to be exploited by employers. When antiimmigration laws are strengthened, wage theft increases. [13] Even in places where there are legal protections for low-wage workers, employers deter workers from using these protections by promising future payment and using confusing payment processes to obfuscate ongoing wage theft. [14] Our study re ects the labor market position of workers similar to Syed's "Market Migrants" in Canada. [15] Market migration is a result of intentional low-wage labor recruitment from migrant and racialized groups that have historically been used to create a labor surplus for exploitation and capitalist accumulation.
Racialized production and economic migration are the result of institutional state exploitation. [16] Capitalism requires exploitation of people's labor, and racial capitalism articulates the people that will be exploited to build the nation. To the extent that precarious labor continues to exist at different levels of formality speaks to the way that the nation state needs racialized capitalism to build and de ne itself. [17] After the 1990 Immigration Act, which drastically decreased the number of legal low-wage immigrants permitted into the U.S., [18] unauthorized immigration steadily increased until the Great Recession in 2008. Over the last decade, undocumented migration to the U.S. is reported to have decreased, [19] yet economic inequality and precarious labor has arguably increased during the same period. While day laborers, especially in Texas, tend to be undocumented immigrants from Latin American countries, [20] the day laborer pool includes immigrants of varying legal statuses, U.S. born citizens, and people from other racial backgrounds. [21] In Texas, the last two decades included the Great Recession in 2008, Hurricane Harvey in 2017, and increased Immigration and Customs Enforcement (ICE) activity in 2019, and most recently the COVID-19 Pandemic, which have all increased the precarity of day labor.
Wage Theft and Worksite Safety for Immigrant Laborers in the United States Workplace safety and wage theft issues, especially in low wage work, remains largely overlooked in addressing human rights and economic equality under capitalism and globalization. [22] In the United States, the minimum wage was established by the Fair Labor Standards Act of 1938 [23] -or the Wages and Hours Bill -and meant to provide a social safety net. Even so, wage theft, or the denial of rightful compensation of labor, is one of the most common crimes in the US and one of the most prevalent forms of labor exploitation worldwide. [24] When workers experience wage theft, they often receive payment, usually in cash, but the payment is a fraction of the amount originally agreed upon, typically verbally, by both parties. [25] In addition, employers often see immigrant labor as an opportunity to pay less than agreed upon wages [26] or less than the wages of non-immigrant workers for the same work. [27] Workers are generally unaware of how to address wage theft, have less equitable access to resources to address wage theft, and are also unaware of how effective attempts to address wage theft might be. [28] Workers navigate a cycle of agency and vulnerability, at times, making decisions that trade-off the chance of providing for basic needs against worse alternatives. Laborers will risk exploitation by an employer to have the chance of earning something. The alternative is often a day without pay. Inadequate workplace safety, and the resulting hazards and injury, is the second most common form of labor exploitation. [29] Both forms of exploitation emerge from a culture among employers of optimizing their returns and devaluing the contributions of their workers. Indeed, workers are assumed to be another resource consumed through work. Moreover, operational policies that focus on short term cost minimization lead to shortcuts in workplace management and a delegation of responsibility for, or an insu cient regard for safety. [30] Although there is not a universal de nition for low wage work, a job that pays less than two-thirds of the median wage for the industry, provides limited opportunity for career advancement, has inconsistent scheduling, and offers no employment bene ts is classi ed as 'low wage'. [31] Domestic work -including childcare, home aids and personal aids-make up a large part of the low-wage workforce, with women making up the majority of domestic laborers. [32] Construction is one of the largest industrial sectors in the global economy as it accounts for about eleven percent of global GDP. The sector is characterized by low barriers of entry, low-wages, hazardous work (the International Labor O ce de nes construction as one of the most hazardous sectors for workers), and a disproportionate share of migrant workers. The on-the-job risks are particularly alarming in the state of Texas where death rates of construction workers are the highest in the US. Nearly half of the excess mortality rates were in specialty construction services who are often hired as independent contractors, making them responsible for their own personal protective equipment. Many of these employees are undocumented immigrants, largely from Latin America. Consequently, construction is one of the main sectors in which workers are extremely vulnerable to exploitation and human tra cking. Indicatively, a lack of visa portability, withholding of passports, and high recruitment fees are some of the risks contributing to the increased vulnerability of migrant construction workers. [33] Speci cally, in the US, most construction rms are small to medium enterprises employing fewer than ten employees, while around three million construction workers are self-employed. In this employment landscape, the industry is characterized by informal, part-time, and temporary working relationships with production work often subcontracted. The subcontracted rms often employ temporary workers on a per project basis. The archetype temporary worker at this depth in the labor supply chain is the day laborer, the focus of this study. As these workers are lacking nancial security and are often socially stigmatized due to their migratory [34] or worker status, they can be easily intimidated, are more prone to accept dangerous working conditions and are more likely to be exploited. [35] In Texas, the Department of Labor and the Texas Workforce Commission both provide ways for workers to report wage theft and all contracted labor, including undocumented laborers, are subject to and protected by labor codes assuring minimum wage. [36] Law enforcement o cers investigating such cases are thus allowed to enforce workers' rights to wages without having to know the citizenship status of the workers. However, immigrant workers, and especially those who are undocumented, are unlikely to report stolen wages. Especially with the use and coverage of a policy like SB-4 in Texas that allows law enforcement o cers to ask anyone to show them their papers, employers can take for granted that undocumented workers won't pursue lost wages through formal means.

Worker Center Interventions
Worker centers are formal, community based, and community run organizations that provide support to day laborers through services, education, and advocacy [37]. Day labor worker centers have emerged as a response to the exploitation of day laborers in the United States and the marginalization workers experience in the communities where they look for work [38]. These centers are one mechanism to impose formal regulation of the informal market, providing a space and resources to the workers that can increase their security as the operate as 'entrepreneurs' of their own labor [39]. The success of these centers can vary based on a variety of factors, like location [40], funding, and worker engagement [41]. Frantz and Fernandes[42] argue that worker center funding model is critical to the organization's ability to serve day laborer needs. In the 1990s, philanthropic foundations adopted strategic funding practices from the nance industry to monitor their relationship with nonpro ts, meaning that funders saw grants as investments and nonpro ts as entrepreneurs to be audited, even though nonpro ts inherently are not supposed to prioritize pro ts. The neoliberal logic of these foundations leads to social and political programming meant to shape day laborers into morally acceptable economic citizens.
[43] These worker centers that are regularly funded by the same large external foundations, tend to have less politically contentious goals to maintain a good relationship with their funders. Instead, they develop programs to promote workforce development, nancial training, and entrepreneurship through employer relationships. Despite these concerns, worker centers are regarded as the generally acceptable response[44] to the most common problems that informal, low-wage workers face, most notably wage theft.
Most day laborers work outside the scope of formal labor regulations and laws that have been put in place to protect workers from hostile work environments[45] -abusive employers, wage withholding, unsafe environments and lack of training. Furthermore, formal workplace policies provide an avenue for reporting abuse that is inaccessible or unknown to day laborers, making them even more vulnerable to continued exploitation. Safety training, seminars, and equipment can ensure a safer worker place environment, but such resources are rarely provided by employers in the informal sectors. Worker centers can provide interventions to ensure such precautions are allocated to the workers.
Worker centers can provide a wide range of services[46]-operating hiring centers, facilitating communication between employers and employees, educating workers on their rights, offering translation services for non-English speakers, and addressing the need for basic hygiene services, from showers to bathrooms. Worker centers can address more than the needs of the individual workers; the centers provide an avenue for the integration of the informal labor of day laborers into the local economy, promoting the standards set by Occupational Safety and Health Administration (OSHA) for any workspace[47] -minimum wage, safe work environment, lling claims and providing adequate training.

Methods
We integrate findings from three studies to examine day laborer experiences of precarious work in Texas. Studies 1 and 2 were conducted in Houston, Texas.
[1] Study 3 was conducted in Austin, Texas. Both cities, like all large metropolitan centers, have communities of day laborers. Germaine to this essay, both have day labor centers, but they differ considerably in their operational models, as will be described further below.
With about 2.3 million people, Houston is the fourth most populous U.S. and with 7.2 million people in the Houston metro area, the Houston metro is the fifth most populous in the U.S. 38 percent of the Houston metro population identifies as Hispanic/Latinx, followed by 35 percent non-Hispanic, white, 17 percent Black, 8 percent Asian, and 2 percent of another race or ethnicity. About 25 percent of Houston's population was born outside the U.S. and of that 25 percent born outside of the U.S., about 62 percent are from Latin America. Houston is the seventh largest metro economy in the U.S. Despite the Great Recession, economic dependence on the oil and gas industry, and Hurricanes Ike and Harvey, the Houston metro economy has steadily grown in GDP, jobs, and population, since the early 2000s. While the bulk of employment and job growth over the last decade occurred in the service industry, construction, and manufacturing-especially related to energy, medicine, and technology-represents 10 percent of industry job growth in the Houston metro and 6.7 percent of industry employment.
Compared to Houston, the City of Austin's population is only 961,000. Yet over the last decade Austin has grown by 27 percent, which is a faster rate of growth than Houston experienced during the same period. [2] Notably, while Houston seems to model what the national population will look like in four decades, Austin trended opposite of the current national trend, whereby 40 percent of the increase in Austin's population over the last decade has come from non-Hispanic whites. [3] Austin's metro area is about 2.3 million people and growing from technology companies moving to the area because of lower taxes, less business regulations, and lower cost of living. [4]

Study 1 -2016 Pilot Study of Day Laborer Exploitation and Labor Trafficking in Houston
In 2016, members of our team (second and fourth authors) conducted a study examining human trafficking in Texas, including a pilot study of labor exploitation and trafficking among day laborers. [5] One of the researchers (fourth author) went to street corners where day laborers went to be hired to recruit for interviews. The researcher had been volunteering with a well-known worker center, with the Fe y Justicia Worker Center in Houston, Texas, that did advocacy work with day laborers. The researcher accompanied an advocate from the Center to street corners and to organizing meetings where both provided information to day laborers about their labor rights. At the street corners, consistent with standard safety protocols for construction and standard procedures for the Center, they offered day laborers bottles of water and masks to cover the mouth and nose. At both sites, the advocate and researcher would offer information about labor rights and trafficking. The researcher asked day laborers if they would be willing to do an interview about their work experiences for 40 US dollars. Twenty-two men and twenty-two women were interviewed. Most of the workers were looking for domestic work or construction-related labor and most of them were undocumented immigrants. All information shared and all interviews were conducted in Spanish.
The survey adapted behaviorally specific questions from Zhang et al. [6] to capture labor trafficking and exploitation endured as part of their employment. Questions asked about abusive practices during transportation to the US, human trafficking during transportation, labor exploitation, threats to physical safety, restriction/deprivation at the workplace, and various forms of deception and lies. Supplemental Table 1 contains the specific items for each of these themes. Pertinent for this analysis, labor exploitation questions asked whether an employer had denied the day laborer pay for work, paid the day laborer less than what they had been promised, had disappeared before paying for work, or had given the day laborer a bad check. Labor exploitation asking about deception and lies asked if day laborers had done different work, a different amount of work, or worked in different environments than they were promised, or if they had been told to work in hazardous environments without proper protection. Day laborers were also asked if employers had restricted their movement during work hours, prevented them from eating or sleeping, kept identification papers from them, told them to lie to authorities, discouraged them from seeking help from authorities, physically or sexually abused them, or threatened physical or sexual abuse. A total of 44 interviews were conducted (22 women) among respondents ranging in age from 20 to 70 years old (mean of 40 years of age). Participants worked in a variety of industries in a day labor capacity, with men primarily working in construction and women primarily working in childcare and housekeeping/janitorial services. All participants were immigrants who had lived in the United States for several years (mean of 12 years in the US); 67% indicated that they were currently undocumented. Primarily these workers emigrated from Mexico (28); other origin countries included El Salvador (8), Guatemala (1), Honduras (3), and Nicaragua (3).

Study 2 -2019 Day Laborer Work History Interviews and Job Choice Experiment in Houston
In 2019, the same research team collaborated with the Fe y Justicia Worker Center to interview day laborers at street corners, again providing workers with information about their rights. Study 2 was designed to examine opportunities to identify and disrupt or remediate exploitation in labor operations as part of a research agenda elucidated by participants of a workshop led by the second author. [7] Additionally, these interviews were conducted in the aftermath of Hurricane Harvey, with goals that included improving what is known about labor in the context of post-disaster reconstruction efforts and used the same behaviorally specific measurement approach as Study 1, adapted from Zhang et al. [8] .
After receiving information about their labor rights, the day laborers were asked if they would be willing to do an interview about their work experiences for 40 dollars. Twenty-two men and twenty-two women were interviewed. All the workers were immigrants; documentation status was not asked. Most of the workers were looking for domestic labor or construction-related work. It is worth noting that the US Immigration and Customs Enforcement raids increased in Houston during this data collection period, and it was harder to recruit day laborers for interviews because they were more suspicious of strangers at levels that were qualitatively greater than seen in Study 1 during 2016.
Initially, the interview protocol focused on gathering employment histories of day laborers in the reconstruction after Hurricane Harvey, but after the first six interviews, researchers realized that day laborers were not able to recall work histories in detail due to the short length of most jobs and the number of employers with whom they interacted with in the months since the hurricane. Researchers adjusted the interview protocol for the remaining thirteen interviews to prioritize understanding how workers made decisions to work in the context of limited work options. The sample included nineteen-day laborers (seventeen men and two women).
Participants' ages ranged from 23 to 65 years old with a mean of 45 years of age. Their years of experience in construction ranged from 2 to 30 years with a mean of 12 years of construction experience, and all but one laborer began working in post-Harvey reconstruction efforts immediately after Hurricane Harvey. These workers emigrated from Cuba (4), El Salvador (4), Guatemala (2), Honduras (3), and Mexico (2). (Four participants declined to provide their country of origin.) All of these workers immigrated to the US prior to Hurricane Harvey.
Interviews included a discrete choice experiment wherein participants were asked to make a series of choices with each choice being between two jobs chosen from an experimental plan involving the following attributes, see Table 1. The first six interviews as part of this round of data collection were used to develop this attributes space. The resulting choice experiment was completed by thirteen participants.
Work is commonly offered with a promise of a daily wage, with some expectation about the length of the workday. Most commonly the length of the job is day by day, but frequently an expectation is set that the work might require multiple days of effort.
Workers also at times have some information about those offering work, either through past personal experience or from information shared among workers. Other factors included the approximate location of the work and some understanding of the tasks involved. Although not included in this exercise, an additional criterion that also factors into worker decision making is whether the employer is looking for multiple workers. On those occasions, workers can at times make a group decision among friends or family members. Anecdotally, workers will attempt to manage perceived risk collectively by accepting work as a group rather than as individuals. These considerations led to a choice design consisting of eight paired choices, each containing two job options, A and B, drawn from the option in Table 1. Participants were asked to consider the two choices, deciding which one they might accept, if either. They had the option to decline both to wait to see if other options were more desirable. The entire set of choices in this experiment are available in Supplemental Table 2.

Study 3 -City of Austin First Workers Labor Center
During the spring and summer of 2020, researchers recorded the experiences of day laborers and gauged diffusion of information and its potential impact on future decision making. More specifically, message recall and willingness to share learned information were examined as critical dynamics that impact the effectiveness of educational interventions among day laborers. To improve understanding of these dynamics, a series of two-part interviews were conducted among day laborers in collaboration with First Worker's Center in Austin, Texas (Center). It is important to note that this data collection effort occurred during the COVID-19 pandemic. As such, the data collection protocol was adapted per IRB and Center policies to utilize telephone interviews rather than in-person interviews to enhance the safety of participants and the research team. First Workers is part of Austin Public Health (APH), a department of the City of Austin. APH operates First Workers to facilitate workers finding short-term employment.
First Workers allows employers, typically contractors of various types, business owners, and homeowners to request laborers by phone, online, or in person at the Center. Fulltime, bilingual staff manage and operate the Center. Customers and workers are provided with direct assistance to help facilitate the hiring process. There is no paperwork for employers to fill out and no fee charged to anyone for this service. Employers can also call ahead to request a specific worker with whom they have worked previously. APH granted our research team access to the First Workers center to conduct research activities as part of their current study. APH used its position in the community to promote the study to area employers and prospective workers in the laborer community. Promotion opportunities included its website and other communication vehicles (flyers, signage, etc.). The research team offered day laborers at the center the opportunity to participate in the study by telephone and email using a list provided by the APH from the center's database. Interviews were conducted in English or Spanish at the preference of the participant. All the participants in the study were males over the age of 18, and the majority spoke some level of English; 33% indicated the Spanish as the main language they were comfortable with. The vast majority of participants (97%) considered themselves to be the main breadwinners of the family, and had varying sources of income, primarily in odd/day jobs. The median income for this sample was $1050 per month. Immigration status was not part of the interview process.

Study 1 -2016 Pilot Study of Day Laborer Labor Exploitation and Labor Trafficking in Houston
This exploratory study [1] showed that day laborers endure high rates of various forms of exploitation, including behaviors characteristic of human trafficking, namely force, fraud, or coercion. The following quotes (translated from Spanish by members of the research team, the third and fourth authors and a paid student researcher, who are native Spanish speakers) from in-person surveys corroborated findings about the ways that day laborers experience labor exploitation, particularly through deception, partial payment, and threats of deportation. Our approach in this section is to support recurring themes with specific testimony from study participants.
Partial Payment of Wages "A month ago, they reopened work and they didn't pay me. Only $694 when it was $2,000. It was for selling water filters." -44-year-old, married woman. She came to the US two years previously to escape violent street gangs in Honduras. She lived with and was financially supporting one of her three children in the US. This quote illustrates the most common form of wage theft in our study, which was partial payment for services rendered.
In contrast to most of the day laborers in this study, this woman described a job requiring more social skills instead of manual skills, probably because she had worked as a nurse for 18 years when she was in Honduras. Although she had more formal training in a profession than most of the other participants in the pilot study and could understand more English than other participants in the pilot study that had been in the US for more years than she had, her level of education did not protect her from exploitation. With the intention of recognizing the agency of the pilot study's day laborers, it is important to note that the day laborer in the previous quote tells her employer that she is supposed to be paid more than three dollars an hour. In the next quote, a day laborer describes how he has tried to decrease the amount of wage theft he experiences. "What happened to me is that they [employers] deceived me a lot. So, I changed my charging system. I charge by the hour, and I ask for their information." -36-year-old, single man. He came to the US from Mexico twenty years ago to earn more money and for a better quality of life. He was financially supporting two children, but they did not live with him. Despite these individual day laborer efforts, like getting an employer's information, or requiring an hourly wage instead of a daily wage, many day laborers described employers like magicians, that would just disappear without paying.
In the pilot study, day laborers explained that delaying promised wages was another way to ensure partial payment, while also securing continued labor from day laborers. "I worked from 6 AM to 10 PM daily for two weeks in an auto shop.
[Then] I worked for 8 months and this time the employer refused to pay me because they owed me more this time.
The employer told me that I should not work so much. The employer told me that I need to wait." -29-year-old, married man. He came to the US from Mexico seven years ago to earn more money and for a better quality of life. He lived with and was financially supporting one child in the US In the previous quote, the employer tells the day laborer to limit the amount of work he does while he waits to get paid. However, day laborers that aren't working won't be able to support themselves and their families. If the day laborer were to work for another employer, that laborer might never get paid, since several day laborers reported employers disappearing without paying them. "I worked cleaning cars. In the summer they don't take care of us. I asked the supervisor to give us money and he mocked us, disappeared…One man fainted. I also got sick." -31-year-old woman living with a partner. Her parents had brought her to the US from El Salvador ten years ago. In the previous quote, when the day laborer asked for payment, the employer mocked her and disappeared without paying. In addition, the day laborer describes a common theme among the pilot study's day laborers, indicating intentional disregard for day laborer health and well-being. "A manager in a national Tex-Mex restaurant chain asked me to work on my day off. I said I would work for 4 hours. The manager took 8 hours from me and would not let me eat. If I complain, they tell me that they are going to deport me." -55-year-old, single woman. She came to the US from El Salvador twenty years ago to escape violence from the civil war in the 1980s. She had three children, two of which she was financially supporting and lived with her. The previous quote also shows another way that employers were able to exploit more work out of day laborers than the day laborers originally agreed to, by using deportation or delaying payment, thereby paying less money for more work. "They don't give us mealtimes and when they have parties we have to be there until we can't take it anymore. And they don't pay us to clean up for their friends." -26-year-old, single woman. She came to the US from Mexico 9 months previously to earn more money and for a better quality of life. She had two children she was supporting financially who did not live with her. In the previous quote, the day laborer described how her employer did not allow her to eat, or leave work, and that she had to clean up for her employer's friends without being paid. Another way that employers forced more labor out of day laborers than the amount originally agreed upon, included taking day laborers to a site to work, and not bringing them back from the site until the work was finished.
Health and Safety Workplace safety was a frequent concern among day laborers. Workplace safety included a myriad of issues, including basic needs like using the bathroom or drinking water, and more extreme concerns like sexual harassment, and medical care for work accident-related injuries. "The employer always yells at me, and I had to go to the bathroom in my truck. The employer told me that the driver does not get to rest." -48-year-old man, separated from his partner. He came to the US from Mexico ten years ago to earn more money and for a better quality of life. He did not live with any of his three children, but he was financially supporting one child. "In restaurants, we work without A/C. It's there, but they don't turn it on because it costs money."-47-year-old woman, separated from her partner. She came to the US from Mexico eleven years ago to escape domestic violence. She had four children, two of which were living with her and which she was also financially supporting. Ideally labor conditions would be a concern not only for workers but also for employers and customers. Bad labor conditions increase the likelihood of costly accidents and decrease the quality of work.
One way that employers avoid the penalty of costly accidents from bad labor conditions is by not paying for the cost of those accidents. "I worked and had an accident in the garden. I burned my hand and arm on the cutter. It swelled up and they didn't take me to the hospital. They took me home. They didn't give me anything. They called me in two weeks to see if I could work again. I told them I was going to go with a lawyer, and they fired me." -25-year-old, married man. He came to the US from Mexico seven years ago to earn more money and for a better quality of life. He had two children that he lived with and was financially supporting. Notably, researchers did not specifically ask about access to medical care as it related to day laborer exploitation and abuse. However, this finding was a prevalent concern throughout both 2016 and 2019 studies, especially because of the difficulty of paying for health care in the US. "They have deceived us. They lie to us so that we are afraid. They tell us, 'I brought you and I pay you what I want.' When we are sick, they don't take care of us…they say that the doctor charges a lot. They are inconsiderate with us. They say that being sick here is a luxury. We don't have access to the doctors." -26-year-old, single woman. She came to the US from Mexico two years ago to earn more money and for a better quality of life. Other day laborers described accidents where employers did not seek medical attention for their workers. One day laborer said their employer would not let them go to a doctor.
Several day laborers described sexual harassment on the job. "The man always touched me and told me I shouldn't be single." -36-year-old, single woman. She came to the US from El Salvador fifteen years ago to earn more money and for a better quality of life.
She had one child that lived with her and who she financially supported. "Many times, people do not report out of fear and there is a lot of sexual harassment. It happens a lot in restaurants to the (female) workers. They remain in fear." -45-year-old, married woman. She came to the US from Mexico seventeen years ago. She had one child that did not live with her. While Crenshaw uses a basement metaphor to theorize about the social location of Black women at the bottom of a race and gender hierarchy, [2] a basement metaphor can also be used to understand the better-known intersectional matrix of domination [3] that racial minority immigrants face. While women day laborers described sexual harassment being a particular problem for women, racial minority immigrant men also experience sexual harassment. "An employer was harassing the workers…Since now it is legal for gays to marry, the employer told us that he can do us the favor of fixing papers if we have relations(hips) with him." -49-year-old, single man. He came to the US from Mexico eleven years ago to escape drug gang violence.
Using the basement metaphor to understand the experiences of day laborers shows how current worker protection policies reinforce existing race and gender hierarchies, especially when immigrant workers compete to successfully use these policy doors in the basement ceiling. One female day laborer said, "It happens a lot. People abuse immigrants. My son and his friend haven't been paid for their work." Another female day laborer said, "In general, the discrimination against Latinos-it doesn't matter if you have papers…they discriminate against us. I haven't met many Americans who understand this experience." In the previous two quotes, one day laborer thinks immigrant status predisposes a person to abuse. The second day laborer thinks it is racism against Latinos that predisposes a person to racial discrimination, regardless of whether that Latino is legally allowed to stay in the US. Another third female day laborer said, "Only a whip is missing-it's slavery. And the worst are Hispanics." This third quote demonstrates the relevance of the basement metaphor to understand how immigrants with and without legal permission, Latinos/as, and Latino/a immigrants, are all in the basement with varying levels of advantage that they can use to stand on top of each other to reach the ceiling.
Day laborers explained that the competition to be hired and not deported made them unlikely to report employers for wage theft or work safety issues. "There are many of us illegals working, and they keep very quiet because we can't say anything, and they don't believe that they are going to get another job." -43-year-old woman who was living with a partner. She came to the US from Nicaragua twenty-one years ago to reunite with family.
She had five children. She lived with and financially supported four of those children.
Another day laborer said, "Many times, a person stays silent because they don't know the language or because they're afraid of deportation." Not reporting wage theft or work safety violations seems like passivity or lack of awareness on the part of day laborers. However, the two previous quotes and the next few quotes would suggest that day laborers are quite aware of their precarious situations and actively choose not to report employers with the intention of securing their individual futures. Yet, given different work environments, they might act differently.
One 54-year-old, divorced woman who came to the US from Mexico thirty-one years ago said, "There is a lack of education…Manipulation and deception come in many forms." She had three children, one of whom she was financially supporting. A second day laborer said, "A lot of training is lacking in the field. The laws-they have to help the worker and not only the employer…The talks from the center need to be more widely shared." A third day laborer said, "I would like to help others." These answers were in response to the interview questions, "Is there anything else you think it is important for us to know or understand about abusive or exploitative work environments for immigrant workers?" and "Is there anything else you would like to add?" These individual stories and anecdotes combine in aggregate to alarming levels of exploitation across the sample of workers in our study. Table 2 shows the percentage of participants who had experienced labor trafficking and exploitation at some point in their past working experiences in the United States, with approximately two thirds of day laborers having experienced some behavior that meets the criteria for human trafficking and nearly nine out of ten having experienced other forms of labor exploitation.
Additionally, these day laborers indicated that they were, on average, paid about $20,000 per year and that, on average, they had also earned an additional $2400, or 12%, that was unpaid. Women and men endure these forms of exploitation at similarly high rates, with abusive labor practices being the most common. No substantive differences between women and men were seen in these summary measures.
More specifically to the focus of the present study, exploitative behaviors like wage theft and safety violations are included in abusive labor practices (See Table 2). At some point in their lives, 61% of participants were denied pay (with 77% and 45% for men and women respectively), 66% received less pay than promised (with 68% and 63% for men and women respectively), and 34% were told to work in hazardous environments (with unknown chemicals) without proper protection (with 32% and 36% for men and women respectively). Other frequent acts of deception included changing the type of work agreed upon, the working conditions promised, and the amount of work offered. A review of a subset of the presented job choices helps elucidate qualitatively how the factors influenced how participants navigate the trade-offs between accepting work under conditions of uncertainty versus declining work to wait for a better or more acceptable option. As shown in Table 3a, choice 1 provides an example of how participants navigate decisions in the context of available choices. All participants chose Job B, the lower paying job, based on how they valued the various attribute levels. One participant provided the following rationale, "[Job] B-more secure pay. Less daily pay but get paid to work.
Working without knowing if you'll get paid is the worst." This choice in the experiment illustrates the willingness of laborers to accept less money for an increased likelihood of receiving the wage that was agreed upon.  Table 3a shows the combination of attribute levels in Job Choice #1 from the set of 8 choices shown to participants.
Choice 8 included the two jobs as shown in Table 3b. Job A in this choice paid more, required working a longer day, and was perceived to involve a safer job site. Both jobs had poor perceptions about the employer's reputation for respecting and paying workers as agreed. When navigating these two opportunities, eight of the 13 participants indicated that they would wait for another opportunity rather than take either option. Comments captured during interview clarify the thinking of participants, including as a representative quote (translated to English) for this particular choice, "Neither pays and [there] will just be problems." To generalize from these two examples, a repeated measures logistic regression model was fit to the choice data when a job was selected to more systematically investigate how job attributes influenced participants' navigation of job opportunities. The 13 participants contributed 208 job choice responses. The attributes in Table 1 were entered as predictors in this model to estimate the effect that the various options had on the likelihood of a participant to accept a job. The model was fit in R using the glmmTMB package with a specification for repeated measures and a random intercept to account for variability across participants. The addition of the job attributes significantly improved the fit of the baseline unconditional model (Δ loglikelihood = 140.3, p = 0.006).
Reputation of the employer for paying the worker as agreed and for providing a save work site have the most substantial impact on participant decisions to choose a job. For example, see Table 4, the likelihood of a job being accepted by a worker when the reputation of an employer for paying as agreed is good is substantially higher than when the reputation is poor (AOR = 27.0, p < 0.001). Similarly, the likelihood of a job being accepted by a worker when the safety condition is perceived as completely safe is substantially higher than when the site is perceived to have little to no safety precautions (AOR = 14.9, p = 0.005). While these results are based on a small number of interviews, the model provides a means of characterizing the importance of worker perceptions about the employer and the jobsite, with implications on benefits to workers of having more reliable information about employers.  This study also provided qualitative insight about worker decision making. At the end of the 2019 interviews, researchers asked day laborers what they would want others to know about their work experiences, what services they needed help with, and what would most improve their work experiences. "Healthcare is always needed. We have to work and can't wait in line for a flu shot. One time I got in a poisonous plant and couldn't pay for the allergy shot they said I needed. I got pretty sick." -48-year-old man. Limited access to doctors was an emergent finding from the 2016 pilot survey interview data which was not specifically asked about in the survey questions regarding labor exploitation and workplace safety. In the previous job choice exercise, worksite safety was an important factor for taking a job, especially if day laborers knew they would not have access to medical care and that they needed to be healthy to earn a living. "English classes, medical assistance, obtaining documents. My wife and I had the gold card, but they removed it because she made a bit more than the threshold, but it all goes to bills…medical help and English classes -that's important. It opens doors." -64-year-old man. The day laborer refers to the gold card, which is the local county's healthcare finance assistance for low-income residents. Day laborers also sought English language classes, transportation, and identification that would allow them to legally work. One 57-year-old man said, "Financial help for transportation for some jobs, for insurance and my car to get to work." English classes, transportation, and identification would also be resources that could keep day laborers safe, since employers often took advantage of day laborers' inability to speak English, transport themselves, or provide identification to authorities in order to exploit them for more labor than they had agreed to.
Building on safety-related quotes from Study 1, where day laborers talked about wanting to help other day laborers, and the need for more worker center education outreach among day laborers, day laborers in Study 2 also wanted to help other day laborers and specifically spoke about how to avoid bad employers. One 64-year-old man said, "Communicating with my fellow day laborers to be aware of the people that don't pay us and take advantage of us -that's important… counting on the worker's center -that helps. [the worker center] supports me in different ways." Similarly, one 57-year-old man said, "Teaching others to be careful -that you think you'll be fine cause you do it daily, but we've all gotten injured. Don't rush, even if they [employers] tell you [to rush]-be careful." People who had been day laborers for decades had developed relationships, skills, and knowledge that newer day laborers did not have. One 48-year-old man said: I haven't had a lot of problems because I learned to ask before it was done. I will ask the contractor throughout the job "is this okay?" and then they know I'm working the way they want to and I interact with them. Since I started doing that, I've had little problems. I'd say 1 out of 10 hasn't paid me or pays me less. It's usually a protest to my work and I protest back so I'm paid at least something. But I got to know a lot of my contractors and they got to know my work.
Lack of information about employers and job details was a prominent issue for day labor decision making. The previous day laborer said that by making sure to interact with employers on the job and to have built a relationship with them meant that he will usually be paid, at least partially for the work he does. Although day laborers might informally share information to help each other avoid bad employers or dangerous worksites, an intervention that could leverage these existing informal practices would be to formalize information about employers and to regulate access to the places where employers and day laborers find each other. Moreover, a formal site for employers and day laborers to meet each other would allow access to other kinds of resources, like skill-development through regular connections with the same laborers.

Study 3 -First Workers Labor Center in Austin, TX
The first round of interviews showed that almost half of day laborers (42%) experienced some kind of wage theft. Among of the participants that had experienced some sort of wage theft, approximately half (57%) chose not to report the incident. The remaining laborers reported to either First Workers or the homeowner/company owner of where the work was done.
In the repeated measures experiment, workers were asked at three different times, how likely they would be to report a wage theft incident using a 10-point scale where 10 meant extremely likely. Immediately after receiving the worker rights information, when asked how likely workers would be to informally report a wage theft incident to an organization like First Workers, there was an overall average score increase of 2.7 from pre to post education. This sentiment continued to increase in the second interview, 48 hours later, to 9.8, a total increase that almost doubled the initial likelihood. Those who had experienced wage theft previously and reported as well as those who had never experienced wage theft expressed a high likelihood of reporting for all three measurements.
In essence, these two groups started with a high estimate for reporting and retained that high estimate across the measurements. Figure 2 displays the mean likelihood to report for the three groups. The results are displayed for 3 points in time: before the workers' right information was provided in the first interview (Pre), after the workers' rights information was provided in the first interview (Post), and in the second interview, 48 hours after the first interview (Post + 48). Sample sizes for all groups and for all three measures are included. Although this was a small experiment, these artefactual findings suggest that education among those who have experienced wage theft has the potential to increase their likelihood of informal reporting if they have never reported before.
Participants were asked again how likely they would be to report, however this report would be formal, to a government agency or organization like OSHA. Figure 3 displays the results for the same three groups for each measure, pre-education (Pre), posteducation (Post), and 48 hours after education (Post + 48). As with informal reporting, those who have experienced wage theft and had reported as well as those who had never experienced wage theft expressed high likelihoods of reporting at all three measurements.
Those who had not reported wage theft that they had experienced doubled their likelihood of reporting, from 5 to 10, after receiving education about their rights and options for reporting. However, 48 hours later, this likelihood had reduced to by 3 points to 7. The net shift in likelihood was from 5 to 7, an increase of 2 across the 48 hours. Taken together, these results suggest that informal reporting may be a potent option for workers who experience wage theft.
Collectively, these data illustrate a learning effect associated with training delivered at worker center. Similar initial effects were seen for formal reporting, but that effect had attenuated by the third measurement point, 24 hours later. The learning effect was observed only among participants who had previously experienced wage theft but had not reported it. However, intuitively, participants who had already reported previous wage theft were already likely to report a future wage theft experience and, thus, showed little increase in their already high likelihood to report. Interestingly, workers who had not previously experienced wage theft also expressed a high likelihood to report at all three measurement points.

Discussion
Global efforts against worker exploitation primarily focus on the labor tra cking of immigrants. These efforts focus on the operations of supply chains as businesses, increasingly demanding cheap labor, while publicly declaring non-binding commitments to living wages for their workers. [1] The dearth of literature on prioritizing workers' rights-despite nationality-illustrates the limited understanding of and attention given to economic and social justice for the working poor. Wage theft and lack of workplace safety are far more common and normalized forms of worker exploitation and are legally not considered to be as severe as labor tra cking. These abuses are especially important to study in a legal and political context like Texas because of its weak labor unions and politics regarding immigration and the border with Mexico. The regional culture makes it arguably more di cult for day laborers to address wage theft and health safety through worker centers, in comparison to states that tend to have stronger labor unions, worker protections, and resources for immigrant workers. [2] The legal environment that normalizes these forms of worker exploitation for some workers and not for others reveals the hierarchical persistence of Crenshaw's "basement." [3] Whereby workers who only manage a single axis of disadvantage are most able to access a policy door to escape the oppressions of the metaphorical basement, and others who experience multiple axes of disadvantage are unlikely to get close enough to the policy door to ever use it successfully.
While Crenshaw refers to Black women at the bottom of the basement and anti-discrimination law as the door in the ceiling of the basement, for this analysis we use the metaphor of the basement to understand the hierarchy of oppressions immigrant workers manage at the bottom of the basement. Crenshaw argues that if Black women were able to enter the house where people without axes of oppression live, then the people who stand on top of Black women in the basement with singular axes of oppression would have already been liberated. For immigrant workers, intersections of gender, sexuality, cultural inequality, housing insecurity, documentation status, and mental health issues from surviving traumatic events would suggest that if these workers could reach the door in the basement ceiling, all the other people on top of them would have also been liberated. At the very least, the intersection of having undocumented status with race and immigrant experience would suggest that if these workers could reach the basement ceiling, immigrant workers dealing with racism and nativism would have been able to successfully use the door of worker protection laws.
Although not covered in this analysis, information on the immigration history of these day laborers indicates a need for mental and physical health services and English language services. A worker center would also help researchers, practitioners, and policy makers to improve interventions more e ciently and effectively over time. Instead of having to go to street corners to nd workers, a worker center could be a central site to provide services to workers, to collect data on workers and employers, and to implement and evaluate interventions, ideally without the involvement of law enforcement. Especially because day laborers almost never successfully use the policy door in the basement ceiling, a worker center is an informal way for a mediator to get the employer to pay a day laborer that has only received partial payment of previously agreed upon wages. If the employer becomes a repeat offender, they would not be allowed to use the worker center to hire day laborers. Employers would be motivated to use the center to nd a regular supply of labor and to also work with reliable day laborers that they had worked with previously. At the very least, meeting inside of a center is more comfortable than meeting on a street corner during the Texas summer.
Notably, worker centers that want to address wage theft have to navigate advocacy practices that would legally categorize them as labor organizations under the National Labor Relations Act, and thereby subject them to a set of regulations speci c to labor organizations. To avoid being labeled a labor organization, the worker center would have to avoid "dealing with employers concerning grievances, labor disputes, wages, rates of pay, hours of employment, or conditions of work." [4] Frantz and Fernandes [5] nd that worker centers that seem to participate in more politically confrontational grassroots organizing and activism focus on local efforts and are structured by day laborer leadership, involving day laborers in decision-making, and using funding sources like membership dues, instead of relying on external philanthropic foundations.
An important next step in this supply chain research would be to nd out more about the social context in which employers of day laborers operate. From our time in the eld, we think that many employers of day laborers may only be one or two levels above day laborers themselves, in the basement metaphor, and some were once day laborers themselves. Knowing more information about employers would inform a more effective intervention, especially if employers are operating based on informal social norms for payment and might be more willing to change their payment practices if they were made aware of laws regulating worker rights.
As depicted in Figure 4, while a worker can discuss pay, inquire about required duties and site conditions, and the expected duration of the work, the actual conditions they experience may turn out to be different than they perceive when they accept the offer of work. That shift from what is expected when the job is accepted to what is experienced is almost entirely controlled by the employer, including decisions that are made after the work has concluded. It is also worth noting that within this informal system, workers will on occasion accept work when they know the conditions will be unfair, either deterministically or with high likelihood. The juncture where agency by day laborers is highest is at the time the job is offered. There is a secondary opportunity after the fact, where a worker can seek justice for exploitation that has been endured. Workers can formally disclose and seek justice through government agencies, like the TWC or law enforcement, or seek informal support from nongovernmental agencies and advocates, like Fe y Justicia, and municipal public health organizations, like First Workers. Figure 4 summarizes the daily journey of a day laborer, depicting how decisions are made with only partial information about the employment situation. Much of what is uncertain to the worker is so because it is controlled by the employer. This journey map also provides a means to identify targets for interventional strategies that improve the agency of works, reduce the opportunities of employers to unilaterally make choices that are harmful to workers, and provide workers with recourse options when exploitive behaviors are experienced. Anecdotally, day laborers want to work, they want to work in fair conditions, and they value opportunities to learn additional trades and improve their skills. Primary prevention programs that combine education about the rights of workers with skill training would appear to be powerful combination that would help workers navigate their hazardous work environment while increasing their employment options. Additionally, similar training among employers might help disrupt some abusive practices. Although constructive interventions that enhance worker conditions are preferred by day laborers, other interventions are needed, too. Enforcement of worker rights and providing easier and more e cient opportunities for workers to report exploitation also have potential to curb abuses by employers. Law enforcement actions coordinated with primary prevention programs provide an additional mechanism to remediate abuse.
The most common type of interventions [6] provided by worker centers is information directly to the workers themselves-trainings, talks, pamphlets, advocates and so on. These sorts of interventions have varying levels of success, most often dependent on how the local community accepts day laborers. [7] Worker centers that are run by community members, with consideration for the language and cultural makeup [8] of the day laborers themselves have high rates of engagement [9] and most importantly, provide an open line of communication between the day laborers and center workers. The day laborers feel secure enough to share concerns and negative experiences not only within themselves, but with the center employees [10], who can then adjust or introduce speci c interventions. Furthermore, the interventions themselves become more effective when the population being targeted is willing, or even better, enthusiastic about the perceived impact that such interventions could provide [11].
Speci c characteristics of the worker center could determine its effectiveness toward the goal of advocating for day laborer needs. For example, the Houston worker center known to the day laborers in this analysis was relatively small and hard to get to. Although the worker center held trainings and sent advocates to informal hiring sites, the center was not a formal convening space for day laborers and employers like the Austin center. Instead, the worker center was a site for advocates to organize themselves before going to the informal spaces where day laborers would be hired. Additionally, the Austin center is part of a public health department, further expanding its potential to serve precarious workers.
Worker centers that are inaccessible, or intimidating, to day laborers, or cater to employers, may have a formal structure but severely lack engagement [12]. These centers have reformed their strategy into the labor market intervention space, rather than working with the day laborers themselves. Through public policy reform, some worker centers have begun advocating for the legal protection of low wage workers by demanding harsher punishments for exploitative employers [13]. These sort of interventions most often involve high level political work in a space that is, ironically, inaccessible to the worker population they aim to help. Accomplishing legal change that is to the bene t of workers is positive but ensuring that these protections are followed through to effective intervention is the real challenge that worker centers face. <p>Mean Likelihood to Informally Report Future Wage Theft to Labor Center. Caption:&nbsp; Figure 2 shows the likelihood of informally reporting a future incident of wage theft to a labor center on a 10-point scale where 10 = extremely likely.&nbsp;The results are displayed for 3 points in time:&nbsp;before the workers' right information was provided in the rst interview (Pre), after the workers' rights information was provided in the rst interview (Post), and in the second interview, 48 hours after the rst interview.&nbsp;The results are strati ed to show the differences between those who had previously experienced wage theft but had not reported it (Did not report), those who had previously experienced and reported wage theft (Reported), and those who had not previously experiences wage theft (None).</p><p> <br></p> Figure 3 <p>Mean Likelihood to Formally Report Future Wage Theft to Workforce Commission.
Caption:&nbsp; Figure 3 shows the likelihood of informally reporting (1-10) a future incident of wage theft to the Texas Work Force Commission.&nbsp;The results are displayed for 3 points in time:&nbsp;before the workers' right information was provided in the rst interview (Pre), after the workers' rights information was provided in the rst interview (Post), and in the second interview, 48 hours after the rst interview.&nbsp;The results are strati ed to show the differences between those who had previously experienced wage theft but had not reported it (Did not report), those who had previously experienced and reported wage theft (Reported), and those who had not previously experiences wage theft (None).</p><p> <br></p>