Article

Regulating the Human Supply Chain

I.     Introduction

Over the past decade, the United States has experienced a 65% decline in undocumented immigration.1 While politicians largely ignore this change, firms that have relied on local undocumented workers as a low-wage labor force feel it acutely. Such companies have increasingly applied to sponsor so-called “guest workers” to fill low-wage jobs. In 2015, the number of temporary migrant workers entering the United States on such visas was nearly double that of undocumented arrivals—the inverse of just ten years earlier.2

U.S. employers’ shift to migrant worker visas has given a newly prominent place to intermediaries abroad in the country’s low-wage work structures. In moving toward greater reliance on legal temporary migrants as a low-wage labor force, the United States is joining a trend already dominant around much of the globe. Almost all low-wage temporary visa programs worldwide, including in the United States, require the employers to hire the worker while she is still in her home country, which makes an intermediary a practical necessity if the employer does not plan to travel there to contract workers.3 Recent estimates put the number of low-wage legal temporary migrant workers in the United States recruited via intermediaries at 80% of the total or higher,4 on par with figures in other migration corridors around the world.5 By contrast, U.S. firms rarely use recruitment entities located in other countries when hiring undocumented immigrants. Instead, they tend to hire among workers already in the country, using informal networks and word of mouth.6

This growing reliance on global labor intermediaries7 to obtain workers for low-paying jobs in the United States is deeply troubling. Around the world, the costs of temporary-labor migration are very high relative to the capital that most migrants have saved or can access through formal channels.8 Legal migration expenses usually include the price of the visa itself, and of travel to the destination country.9 These costs are compounded by the rent-seeking10 behavior of recruiters, who routinely charge high and often illegal fees for their services.11 To shoulder these expenses, migrants often resort to informal lenders charging usurious interest rates.12 Once they arrive at their destination and begin working, they must pay off the resulting load of high-interest debt before they can begin sending money home for the purposes that led them to migrate in the first place: supporting those left behind and pursuing longer-term education, business, or home-ownership goals.13 In the United States, as in most destination countries, a low-wage migrant’s visa is tied to her sponsoring employer, and ends when the job does.14 This means that no matter how dangerous or low-paid a migrant’s work turns out to be, she will be very reluctant to protest violations of her rights, or even to reveal them to advocates—much less government agencies—lest she be sent home.15 This, in turn, reduces the enforceability of workplace laws, and undercuts conditions for local workers.16 As a whole, this process shifts the costs of matching global labor with capital from employers onto migrants.17 Firms that employ recruited migrants receive a labor force far cheaper than the law purports to allow.18 In most countries, the firm benefits from this subordination of migrants without any accountability, because it is insulated from liability for the actions of its intermediaries that perpetuate it.19

This Article argues that these harms to migrant workers are not generated by anomalous bad actors, but instead are structural. They are the product of the global market for labor under current economic conditions, laws, and enforcement levels. Bad actors exist among labor recruiters, of course, as the copious evidence on fraud, threats, and violence in the recruitment industry around the world illustrates.20 But even if the worst actors were eliminated, most transnational migrants would still be paying amounts that require them to incur high-interest debt, both because of the cost of the visa itself, set by the destination country, and because functionally unregulated labor recruiters are able to charge fees for access to a scarce and desirable low-wage job abroad.21 Therefore, criminalizing bad apples among recruiters is not likely to correct the situation. A structural problem requires a structural solution.

The global industry that delivers migrants to temporary jobs abroad is a prime candidate for regulation to address market failures. Many countries have responded by legislating limits to recruitment fees, and they are banned completely by Mexico and by the United States for its low-wage migrant worker visa programs.22 Yet for all the concern that rent-seeking behavior of transnational recruitment intermediaries in many migration corridors, there is no effective governance regime regulating recruiters’ actions or their interactions with the firms that hire them.23 A number of migrant-origin countries have enacted innovative laws regarding recruitment, but lack the resources, political will, or market power to enforce them effectively.24 Most destination nations insist that recruitment is a matter over which they have no jurisdiction, because it occurs on foreign soil.25 No international entity has the authority to set rules for the terms on which migration occurs.

Legal scholarship has failed to take note of the increasing impact of recruitment intermediaries on the rule of law in the United States, and on the regulation of employment in U.S. low-wage labor markets in particular. Intermediaries appear in articles on human trafficking, but most trafficking scholarship focuses on bad actors rather than a structural analysis of the recruitment system as a whole.26 Likewise, although legal academics have explored the way employers of temporary migrant workers in the United States have violated workplace laws,27 they have not emphasized the routine economic interactions that occur at the employer’s behest before the migrant arrives on the job.28

This Article fills that gap. Its first contribution is a novel conceptual framing of migrant recruitment as a “human supply chain.” Global labor arbitrage,29 the process through which corporations seek to reduce labor costs by contracting workers from competing low-wage countries, brings human supply chains into being. It also gives rise to the more familiar product supply chains when brands and retailers move their manufacturing operations abroad.30 Where the work cannot move because it must be done in a particular location, corporations seek to bring in low-wage workers from less-wealthy countries.31 Human supply chains recruit and carry migrant workers to these jobs. It is not a coincidence that much of this migrant employment takes place at the lower levels of product supply chains.32 When multinational companies demand lower prices from their suppliers in product supply chains, tapping the human supply chain to bring in migrant labor is one way that those suppliers reduce costs to remain competitive.

This Article builds on this analysis to make and evaluate a set of proposals to address the subordination of workers and the negative effects on the rule of law caused by an unregulated human supply chain in low-wage labor markets. While this Article’s proposals draw on the considerable learning that has taken place over the past two decades in the context of product supply chains, they also reflect an analysis of the important differences between the two chains. These differences are generated by the political complications that result from moving human beings, rather than objects, across borders, but also by distinctions between the market structures and business models of the global supply of goods versus the global supply of workers. The proposals set out here are also fundamentally different from the recommendations that have emerged from the anti-trafficking framework, in that they do not emphasize the criminalization of bad actors. Instead, this Article argues for an effective means of changing the economic incentives of the entities and individuals in the human supply chain’s vast middle.

With this end goal in mind, this Article draws on my original research into emerging public, private, and hybrid regimes to govern human supply chains in order to evaluate the three most common approaches: (1) direct regulation of recruiters by origin-country governments alone, which to date has been the principal line of attack worldwide (with few examples of success); (2) the imposition by destination governments of joint and several liability on employers for their recruiters’ violations, which a few countries are currently piloting with some limited success through public and public-private initiatives; and, even rarer, (3) full chain liability that renders all actors in the supply network responsible for recruitment abuses, from the intermediary up to the multinational firm at the top of a product supply chain that relies on temporary migrant labor.33 It offers brief case studies of recent efforts that pursue the latter two routes, and concludes that the full-chain-liability approach is the most efficient and equitable approach to regulate the human supply chain, particularly when it arises as a transnational collaboration between government and civil-society actors in origin and destination countries. Chain liability for recruitment abuses will enlist the participation of the actors who create the market for global labor intermediaries and have the power to shift incentives in that market, in order to reinforce workplace protective laws and advance decent working conditions for both migrants and native workers.

Part II identifies the increasing presence of transnational labor intermediaries in the U.S. low-wage labor market and analyzes the reasons for their negative impact on workplace regulation enforcement. Part III theorizes the labor-recruitment process as a human supply chain, both roughly analogous to and often integrated with global product supply chains. Part IV proposes remedies to the governance deficit in the context of the human supply chain. It considers efforts to improve working conditions in product supply chains as a potential source of insight, but also identifies a number of critical differences between the two. It explores tort law as one model for regulation, proposing strict liability as an appropriate regime through which to govern the human supply chain. With those lessons in mind, it evaluates emerging approaches to the regulation of recruitment that involve some measure of supply-chain liability—illustrated with brief original case studies of recent government, private sector, and union initiatives—and argues that strict liability applied to the full chain will be most effective.

II.     The Recent Rise of Global Labor Intermediaries in U.S. Low-Wage Labor Markets

The movement of workers across borders is a pervasive aspect of globalized labor markets.34 An estimated 150 million transnational migrants labor around the world today.35 Agriculture, tourism, childcare, eldercare, and construction are paradigmatic migrant-heavy industries, all ones in which work cannot be offshored because it requires the worker’s presence in the wealthier country.36 There are significant commonalities, but also meaningful regional and industrial differences between different arenas of temporary labor migration. For the sake of brevity and accuracy, this Article grounds its discussion in one particular arena—low-wage, legal, temporary migration between Mexico and the United States—although it extends this analysis to the broader issue of temporary labor migration in general where appropriate.

This Article focuses on legally authorized, low-wage temporary migrants to the United States in important part because the country is in the midst of a pronounced shift away from a reliance on undocumented immigrants as a low-wage labor force. The U.S. undocumented population hit “a peak of 12.2 million in 2007,” following several years in which undocumented immigrants came to the United States at an average rate of 850,000 annually.37 Shortly thereafter, the global economic downturn, an increase in U.S. government enforcement against undocumented immigrants and their employers, and demographic factors in Mexico, led to a sharp drop in unauthorized immigration to the United States.38 Since 2009, the flow of undocumented immigrants into the country had fallen to an estimated 350,000 per year.39 The undocumented population fell to 10.9 million in 2014, following several years of net-negative unauthorized immigration.40 Mexican migration has been particularly affected: there are more than 1 million fewer Mexican immigrants in the United States now than approximately a decade ago.41

Over the past ten years, with the number of undocumented workers shrinking and the likelihood of penalties for their employment increasing, low-wage employers have increasingly sought to hire legal temporary migrants through U.S. low-wage migrant-worker visa programs.42 Most of these workers come from Mexico.43 Approximately 600,000 migrants and their families now enter the United States each year on temporary work visas,44 far outnumbering the current annual entries of undocumented people who were until so recently this country’s principal low-wage immigrant labor supply. The Chamber of Commerce and major corporations are seeking rapid expansion of migrant-worker visas, pressing Congress to allow hundreds of thousands of additional migrants per year.45 Their lobbying has already borne fruit,46 and seems likely to bear more. The immigration bill passed by the Senate in June 2013—this country’s most recent serious attempt at immigration reform—would eventually have permitted the creation of some 400,000 additional migrant-worker visas annually.47

Many temporary work visas are granted to professionals in technology and other areas, inter-company transferees, and sports figures.48 Perhaps a third of employers go to migrants who will work in two low-wage categories: farm work (“H-2A” visas, so called for their place in the Immigration and Nationality Act) and other “temporary service or labor” (H-2B visas).49

The H-2B visa is limited to work where the employer’s underlying need is one-time, seasonal, or intermittent,50 such as in landscaping, forestry, amusement parks, and food processing.51 Demand for H-2A and H-2B visas has increased steadily since 2009,52 even as employers have complained about the program’s process and worker protections.53

As U.S. employers have turned to temporary migrant workers, they have hired intermediaries abroad to help locate, recruit, and transport them.54 This is very different from the methods that firms use when hiring undocumented immigrants.55 Formal intermediaries are not necessary in that context, because undocumented immigrants are most often already present in the domestic labor market, allowing employers to use a range of local recruitment methods.56 Contracting with an intermediary to recruit undocumented workers also exposes a firm to serious legal sanctions.57 By contrast, the H-2 program only permits a migrant to come to the United States if a U.S. firm approved for such visas contracts her while she is still in her native country.58 Rather than traveling to Mexico or Thailand to hire workers, the majority of employers use recruitment firms in migrant countries of origin.59 Behind the scenes, these intermediaries are thus playing an increasing role in this country’s structures of low-wage work.

There has been no systematic study of the recruitment industry that feeds the H-2A and H-2B programs. In the absence of industry-wide data, advocate surveys of migrants provide the most comprehensive information available.60 These reveal a variety of approaches to recruitment.61 Large companies employing H-2 migrants may have human resources departments through which they manage recruitment internally. Smaller firms sometimes deputize a favored migrant employee to serve in this role. Many employers, however, work with one or more recruitment agencies that are at least nominally independent. These range from large H-2 specialist agencies like Más Labor or CSI Services62 to small firms to individuals working as brokers in migrants’ hometowns. There are often multiple links in the chain that brings would-be migrants to their employers. A U.S. employer may contract with a recruitment or staffing agency in the United States, which subcontracts to a Mexican nationwide recruitment firm, which in turn sends the order to a branch office, which contacts a local broker to find the requested workers.63

It is important to distinguish between recruitment-only and staffing agencies because the former do not remain the migrants’ employers of record after they deliver them to their destination countries. Recruitment-only agencies find workers to fill jobs for firms in other countries. They process migrants’ immigration papers and transport them to the location in the destination country where they will work. After they hand the workers over to the firm that will be their employer, recruitment-only agencies cease being part of the employment relationship. Staffing agencies also generally recruit workers, process their immigration papers, and transport them to the location in the destination country where they will work. However, staffing agencies remain the migrants’ employers of record on arrival, while leasing them as laborers to another U.S. firm or firms. This Article focuses on the more difficult case of the recruitment-only agencies that do not operate as the employers of migrants in the United States.64

From the perspective of both migrants and migrant employers, recruitment agencies offer useful services, including identifying, interviewing, and processing the visa documents of potential workers, matching them with jobs abroad, and helping them travel to their destination.65 In these ways, they decrease information costs and frictions in the global market for employment.

On the employer side, the use of recruiters has few downsides under the current legal regime. A firm can choose from a wide array of global labor intermediaries, pay an amount determined by a competitive market,66 and exercise as much or as little control over the hiring process as it wishes.67 If an employer recruits workers directly, it is liable for any violations of U.S. law occurring during that process. By contrast, using an intermediary almost always insulates it from liability for recruitment abuses. U.S. law governing the H-2A visa requires employers to contractually forbid their recruiters from charging fees to workers.68 Once the employer has done so, however, it does not bear responsibility for recruiter fees under the H-2A regulations unless it can be demonstrated that the employer “knew or reasonably should have known” of the violation by the recruiter.69 Advocates have argued in the alternative that the minimum-wage protections of the FLSA should take recruitment fees into account as deductions from wages when calculating whether H-2 migrants were paid in accordance with employment law.70 Courts have largely rejected these arguments, except where the petitioner demonstrates that the employer knew of and bore some direct responsibility for the charges.71

Migrants have much less choice and power in their interactions with recruiters than employers do. There may be just one recruiter where a would-be migrant lives and many others competing for the visas the recruiter offers to obtain. Migrants’ desire to enter the labor market abroad, combined with stiff competition for a limited number of visas available from a limited number of recruiters, generates large potential rents for the intermediaries who control access to temporary jobs abroad.72

The amount a migrant is willing to pay for access to the global labor market is determined by the difference in wages, the ratio of would-be migrants to visas, what migrants are told about the work on offer, and the (often inaccurate) information the migrant has about likely income and costs of living in the destination country.73

The gap between migrants’ potential earnings in their origin country as compared to their potential earnings in a destination country is one of the most significant factors affecting their willingness to pay for access to that country. In the United States, Mexican workers can earn four to nine times as much as they could at home, depending on the method of calculation;74 for Bulgarian workers migrating to the Netherlands, the ratio is roughly six to one;75 for a worker from Vietnam in South Korea, it is at least eleven to one.76 Policy makers sometimes refer to this as the “wage wedge.”77 It is calculated without reference to the expenses a migrant must undertake in order to get one of the coveted jobs abroad. From the perspective of firms in the destination country, the wage wedge is an opportunity to hire workers willing to labor for less than those in the domestic market. From the perspective of many in developing countries, a job abroad and the higher income the wage wedge brings is the only way to support a family, fund a small business, pay for children’s education, or allow—some distant day—for retirement back home.78

Recruiters often charge far more than permitted, more than they could demand in a fully competitive market, and indeed more than the average annual income in migrants’ origin country.79 Functionally, as migration scholar Manolo Abella explains, “[w]hat the recruiter gets is not a fee for recruitment services per se, but a bribe for the jobs that he or she offers,”80 except that “extorted payment” might be a more appropriate label.81 Indeed, every actor in the migration industry has its hand out to claim a piece of the wage wedge, from the broker at the village level to the bus driver who takes the migrant to the capital to the government official at the airport.82 These behaviors represent a common business model in the industry, not a deviant one.83

On top of recruitment fees, migrants may have to pay the destination-country visa fee and travel expenses.84 Where employers are required to repay these charges once workers arrive,85 migrants still must find the capital to cover them up front. The combination of authorized and illegal costs means that few would-be migrants can pay the amounts required for entry into the transnational migrant labor market without resorting to informal moneylenders offering loans at usurious rates.86 The high interest rates on such loans translate even small borrowed sums into substantial ones.87 Lenders, some of whom are also recruiters, may demand titles to land, houses, or cars as collateral.88 The need to repay lenders cuts deeply into a migrant’s effective wage rate abroad. Yet even if migrants have accurate information about the job they will receive (which many do not), they are likely to be unaware of how much the payments on a high-interest loan will reduce their expected income.89

The impact of this rent-seeking by recruiters is compounded by the fact that each H-2 visa only allows the migrant to remain for a single season, which means that she must find work in the United States anew each year. Even where she believes that her employer has requested her return, re-hire is far from certain. Recruiters are known for keeping blacklists of migrants who displease them.90 Because migrants often have little choice between recruiters in their home communities, they must stay in the particular recruiter’s favor so as not to be barred from future work opportunities abroad.

It has been less than a decade since advocates began bringing evidence of pervasive recruitment abuses in U.S. temporary-work-visa programs to the public’s attention. Since then, numerous policy reports, government complaints, and public-interest litigations have laid out story after story of unlawful recruitment fees leading to high levels of migrant debt.91 In some cases, companies with temporary migrant workers may call in recruiters to quiet unrest, if protest is brewing. Recruitment agents may threaten to repossess collateral on loans, bring in immigration officials to initiate deportation proceedings, and use violence against families back home, all to help the employer maintain control over its migrant workforce.92 For the purposes of this Article—focused less on the extremes than the mediocre middle of the recruitment continuum—it would be helpful to know how often these abuses occur in the H-2 program as a whole, in order to get a sense of average recruitment practices. Such information is unfortunately not available.

What limited studies have been done with regard to fees and debt in the context of H-2 visas have not produced generalizable results, but do give a sense that migrants are regularly required to pay unlawful charges. A survey of 382 Mexican migrants by a coalition of five advocacy organizations found that 42.6% of those questioned had been required to pay a recruiter for an H-2A job offer or even the hope of such an offer, in amounts ranging from $31 to $350 (in addition to passport and visa expenses).93 The Centro de los Derechos del Migrante, a transnational advocacy organization for Mexican migrant workers, reported in 2013 that of 220 Mexican migrants it surveyed, 58% had paid such fees, amounting on average to $590.94 For purposes of comparison, the Mexican daily minimum wage—which is more than many workers earn—is approximately $4 per day.95 Moreover, most migration corridors around the world require migrants to pay considerably higher fees and costs than those demanded in the Mexico–United States corridor.96

The amount that migrants pay to obtain work abroad subsidizes their employers’ recruitment costs. Firms that choose to use temporary migrant labor rather than hiring resident workers reduce expenses in other ways as well, even though U.S. law usually requires firms hiring migrants on H-2 visas to pay in excess of the U.S. minimum wage.97 H-2 employers all but eliminate turnover, as the migrants must stay on the job for the duration of the season.98 The employers also save on wages. In 2015, the Economic Policy Institute (“EPI”) released the results of the first empirical study of the magnitude of this effect. Laura Apgar’s analysis, based on data collected by the Mexican Migration Project Survey between 1987 and 2013, found that temporary migrant workers’ hourly earnings were roughly the same as undocumented workers, despite the former’s legally authorized status, and approximately 11% less than immigrants who were legal permanent residents.99 In 2016, EPI published a second paper focusing on wages in the H-2B program in particular. In that study, Daniel Costa used wage data from the Department of Labor in 2012,100 2013,101 and 2014,102 to determine that for the top two H-2B occupations, employers paid H-2B workers approximately $3.25 per hour less than the national average for those sectors. This amounts to a reduction in pay of almost 25% when employers hire H-2B migrants instead of their resident counterparts.103 This suggests that either H-2 employers are paying wages that violate the law, or the process for setting H-2 minimum wages significantly underestimates the actual prevailing wage in the region and occupation.

In the United States, heavy reliance on labor recruiters abroad to bring in low-wage workers is a relatively new phenomenon outside of agriculture.104 It has a longer history in many other migrant-destination countries, which have relied heavily on temporary labor migration initiatives. Popular after World War II, but out of favor by the 1970s, legal temporary labor migration programs came back in vogue around the globe in the 1990s, rebranded as “circular migration.”105 In Asia and the Gulf States, the vast majority of labor migration occurs through such legal temporary visas.106 Canada uses the same model, as does the European Union (“EU”) for migrant workers from non-EU countries.107 Major international organizations, including the World Bank,108 the International Organization for Migration,109 and the Organization for Economic Co-operation and Development (“OECD”),110 among others, tout temporary migrant visas as a win-win-win for origin and destination countries and migrants themselves.111

The global recruitment business is enormous, and growing rapidly.112 As one indicator of how central intermediaries are to the temporary migrant worker process in the Asia–Middle East corridor, almost all of the estimated 10 million migrants in the Gulf States are there on temporary work visas, and recruiting agencies place approximately 80% of them in jobs.113 Abuses in that context are endemic, similar in kind (that is, involving fees, fraud, and retaliation) but often considerably greater in degree than those noted in the U.S. program.114 As the United States increasingly comes to rely on temporary migrant workers as low-wage labor, it urgently needs to develop a better grasp of the practices of global labor intermediaries and the likely impact of granting them a newly prominent role in its labor market, in order to design effective policies to protect migrant workers from the outset. This Article is written for that purpose.

III.     Theorizing Global Labor Recruitment

A.     Recruitment as a Human Supply Chain

Since the United States is a relative newcomer to the modern world of transnational labor recruiters, it is perhaps not surprising that global labor intermediaries have received very little attention in U.S. scholarly literature generally, and almost none in the field of law.115 In the past five years, legal advocates, rather than academics, have brought the harms that accompany recruitment to the public’s eye.116

Scholarship on human trafficking, which can encompass incidents of severe recruitment abuse, is the principal exception.117 The trafficking lens has been successful in drawing the attention of the public and politicians to the worst problems of labor recruitment, but inadvertently reduces political incentive to reform the system as a whole. For the purposes of addressing the broader harms of the recruitment industry, it is troubling to focus exclusively on recruitment situations that fall within the trafficking ambit. Highlighting the super-exploiters who perpetuate “modern day slavery” has generated legislative responses that are individualized in that they seek to criminalize the worst actors, and are backward-looking rather than preventive.118 The implication of this focus is that the important problems are caused by aberrant behavior, and do not require a structural examination of systemic incentives embedded in the global economy under its current regulatory regime.119 Relatedly, the anti-trafficking framework has given rise to an industry of certification programs that encourage the public to take reassurance from the voluntary assurances of “good actor” brands that they do not tolerate such abusers in their supply chains.120 Such promises have proven extremely hard to verify and enforce.121

The time is ripe for a new theory of global labor recruitment, one that offers a comprehensive account of the forces driving the field. This Article provides that theory, contending that labor recruitment for work abroad is best conceptualized as a transnational human supply chain.122 The human-supply-chain framing insists on the recognition that labor migration is a fundamental aspect of production in the global economy—as fundamental as the outsourcing of manufacturing and services, which also takes place through a supply-chain structure. These supply chains often intersect. 123

While the human supply chain analysis shares with the anti-trafficking framework a primary concern with addressing negative impacts on workers, it does so with a new and different view of the scope of problems that merit attention, their causes, and how they should be addressed. The human-supply-chain lens sees migration intermediaries as operating on a continuum, rather than divided into bad and good actors, and its central interest is understanding the space that most intermediaries occupy between those extremes. It sees the violation of migrants’ rights as a structural phenomenon, the product of a political economy that incentivizes a large number of firms to use recruiters to hire temporary migrant workers and permits recruiters to capture substantial economic rents from those migrants. This opens the door to considering broader solutions that would shift the incentives in the market for recruitment as a whole.

In its most basic form, the recruiter’s function in linking migrants to an employer is set out in Figure 1.124

 

Figure 1: The Basic Recruiter-Employer Relationship

Figure 1 Gordon2

       

Although the core action of recruitment is captured in this simple diagram, the process of delivering people in one country to work in another often involves a chain of subcontracting arrangements.125 As shown in Figure 2, a principal recruitment firm—which may be located in the destination or origin country, or both—may manage a network of sub-agents who stretch its reach into far-flung rural areas and offer ancillary services like moneylending, transportation, and a place to stay along the journey.126

 

Figure 2: Example of a Human Supply Chain

Figure 2 Gordon2

 

In turn, human supply chains often (although not always) recruit workers into jobs for firms that are positioned at the bottom of product or service supply chains.127 Figure 3 illustrates this integration.128

 

Figure 3: Example of Intersecting Labor Recruitment and Product Supply Chains

Figure 3 Gordon

 

When a human supply chain brings migrants to a firm at the bottom of a product or service supply chain, those workers carry the burdens that grow from labor recruitment into subcontracted jobs that are already likely to be poorly paid and unstable.129

B.     Where Human and Product Supply Chains Meet: An Example

Apple Fresh is a (fictitious) apple cider maker in Washington State. Apple Fresh owns an orchard and, in its first decade, controls the apple harvest and all aspects of cider production, from hiring the pickers to pressing the juice to storing the finished beverage and selling it at farmers’ markets and local stores. Like all employers, Apple Fresh is responsible for ensuring that its employees’ wages, benefits, and working conditions comport with legal and contractual minimums. It must also pay social-security premiums on its employees’ behalf and cover their unemployment and workers’ compensation insurance.130 In its second decade, Apple Fresh begins to expand beyond the local market, selling its cider to large grocery chains such as Safeway and Wal–Mart. It is under continual pressure from those retailers to make its cider supply more predictable and to reduce its prices. As part of its effort to meet those demands, Apple Fresh decides to outsource its apple pressing to one of several food processors in the market, Presser Inc., which can produce the cider more cheaply and efficiently. Once it signs a contract with Presser, Apple Fresh is released from responsibility for the social insurance and many of the working conditions of the workers who press its apples, because it is no longer their employer. Presser now bears those obligations. Of course, the contract price should reflect Presser’s costs for fulfilling them. But since, like most businesses in production chains, Apple Fresh sought bids for its pressing contract and favored low bidders, Presser had an incentive to cut corners on wages and compliance with workplace laws in order to get the job. Presser’s employees protest the way they are paid and treated, and Presser has a high turnover rate, but it is able to meet its commitment to Apple Fresh in the first year of the contract and still earn a slim profit. Meanwhile, Apple Fresh has reduced its costs, and continues to reap income as before from the sale of its brand-name cider.

In year two of the contract, Presser decides to try to decrease turnover and increase its profit margin by using temporary migrant workers to staff its plant. Its owner had been contacted not long before by the U.S. agent of a labor-recruitment firm in Mexico City to discuss the advantages of contracting the firm to hire H-2B migrants, including lower costs and a steady supply of uncomplaining workers.131 Presser’s human-resources department calls the agent and asks him to begin the recruitment process. The agent in turn contacts the firm in Mexico City, which has sub-agents in a Mexican state capital, who work with brokers in rural areas to sign up would-be migrants. As the migrants travel up this chain, each of the agents demands a side payment from them, on top of the cost of the visa and travel expenses. By the time the workers arrive at Presser’s plant, they owe high-interest lenders over three months’ salary to pay back the loans they took out to cover the charges. These fees are all in violation of Mexican law, but none of Presser’s recruiters has been penalized, both because the law is rarely enforced, and because the principal recruitment firm blames “unauthorized individuals” for the violations.

The migrants are well aware that to make good on their loans and begin to earn the money that their families back home are expecting they must not displease their supervisors at Presser. If the migrant workers do complain, and are fired, they are immediately subject to deportation, because their visas are valid exclusively to work for Presser. In that sense, Presser can rely on U.S. government enforcement of immigration law as an additional mechanism of control over its labor force, whether it or the recruiter makes the call to the government. With their debt and the fear of deportation foremost in their minds, the migrants at Presser work hard and make no demands, despite their concerns about safety and treatment. Presser’s productivity and profits rise with this new staff of subservient workers.

When the law releases Presser from liability for the actions of labor recruiters that provide it with indebted workers, liability that Presser would have borne if it recruited the workers directly, it allows Presser—and Apple Fresh and the retailers that purchase its cider—to shed costs and (potentially) increase profits132 without paying the price for the means through which these benefits come to them. Their lack of responsibility is a legal fiction. In fact, both Presser and Apple Fresh have actively chosen to subcontract aspects of their business because the combination of private contracting arrangements and public laws about immigration control and liability for the treatment of migrants and workers allows them to benefit from the decision to outsource a firm function without bearing the true cost. Neither, however, is likely to benefit much from this decision, beyond the ability to stay in business. It is the retailers higher in the supply chain that, by pressuring suppliers to reduce prices, both encourage subcontracted production and reap the bulk of its profits.

IV.     Regulating the Human Supply Chain

This Part sets out and considers ways to address the global-governance gap in the context of the human supply chain. It summarizes key lessons from the scholarship, exploring the lessons the product supply chain offers as to how to regulate labor intermediaries and their implications for working-conditions regulation. However, contrary to an economic analysis that sees product and labor mobility as functionally identical, this Part identifies important differences between the two that impede the neat transfer of approaches developed in one context to the other. Particularly important are the combination of factors that make it possible for intermediaries in the human supply chain to capture substantial rents, while evading inspection and limiting the returns to a “good recruiter” business model. With these distinctions in mind, this Part considers three possible approaches to regulating the global human supply chain: (1) direct regulation of recruiters; (2) joint liability between employers and recruiters; and (3) full chain liability for all actors. It presents brief case studies of the latter two approaches, and argues that the full-chain-liability approach will be the most effective way to protect temporary migrant workers in the modern economy.

A.     The Governance Deficit in Global Labor Recruitment

The problems that workers face in the human supply chain occur in a vacuum of effective regulation. A decade ago, Professors Gary Gereffi and Frederick Mayer identified such a “governance deficit” in the context of the rights of workers in global product supply chains.133 They traced the gap to “a mismatch between the global economy and the institutions of market governance”134 on three levels. Government institutions in wealthier nations have a diminished capacity to regulate economic activity when it occurs beyond their borders; governments in less-developed countries generally have very limited resources and institutions to monitor and enforce workplace standards (as well as a lack of political will); and international standards in the context of labor rights only bind countries that have adopted them and are notoriously difficult to enforce.135 Although Gereffi and Mayer made their observations in the context of global manufacturing supply chains, they also apply—perhaps even more acutely—to the regulation of human supply chains. Labor recruitment has earned a reputation as not only ungoverned but ungovernable.

This reputation is derived largely from observation of the failure of unilateral efforts to effectively establish standards for recruiters in migrant-sending nations. Recruitment takes place in origin countries, and therefore chiefly falls within the origin governments’ legal jurisdiction.136 What origin-country laws say on paper about how recruitment services should operate, however, is rarely reflected in how the industry actually works. A number of origin nations limit or ban recruitment fees and costs, and require recruitment agencies to be inspected, bonded, obtain a license, and/‌or register, yet lack the capacity and political will for effective enforcement against principal recruitment companies, much less their many sub-agents and brokers.137 Perhaps the most powerful impediment to effective regulation of recruitment from the origin-country government perspective is a fundamental reluctance to imperil the income that migrant remittances generate.138 Income from migrants is essential to keeping many origin-country economies afloat. According to the World Bank, worldwide migrant remittance flows to developing countries reached an all-time high of $441 billion in 2015.139 In El Salvador, remittances make up 18% of GDP;140 in Nepal the number is 29.8%, and in Tajikistan it is 28.8%.141 Origin-country governments do not want to create conditions that might make their citizens more expensive to hire, leading employers in destination countries to turn instead to other nations whose migrants will work for less.142 Acting alone, they are unable to address the factors that drive the recruitment market—the imbalance in the global distribution of wealth and the limited number of destination country jobs available to authorized migrants—without harming their own economies.143

In response to the difficulty of unilaterally regulating a transnational system, many origin countries have sought bilateral agreements with destination governments that set terms for temporary labor migration.144 While a number of agreements have been signed, and some of those include recruitment and employment standards, they are largely intended to open markets to new migration flows rather than to protect workers. Most are drafted and negotiated with no representation of trade unions, non-governmental organizations, or employer associations.145 The language about rights in these agreements is vague, with little or no discussion of enforcement mechanisms. Even when they do detail a complaint process, it often relies on diplomatic channels rather than a labor-standards agency for enforcement, and an origin government faces the same incentives not to pursue it on behalf of its citizens. Not surprisingly, then, such agreements have rarely been the springboard for actual coordination of enforcement, much less measurable improvement in conditions.146 Some major destination countries like the United States and many Gulf nations refuse to negotiate such accords at all.147

For their part, destination countries have until recently shown little interest in addressing recruitment abuses, which—most destination governments insist—largely occur outside their jurisdiction.148 The exception is where the violations meet the high threshold of the legal standard for human trafficking. Unlike other transnational systems such as trade, labor migration is not governed by internationally adopted standards.149 The International Labor Organization (“ILO”) has promulgated standards on global recruitment, but they have significant holes, and most major origin and destination nations have not ratified them.150 Taken together, these factors have created a governance gap of tremendous proportions.151 Because of that gap, intermediaries are able to engage in rent-seeking strategies—and worse—with impunity.152

B.     Lessons from the Product Supply Chain

Most manufacturing today—and an increasing amount of service provision—takes place through transnational supply networks.153 Beginning in the 1980s, as changes in geopolitics, financial markets, trade accords, and technology made production across borders increasingly feasible, brands, retailers, and other multinational corporations began to shift various aspects of their operations into less-developed countries.154 The incentives for this shift were varied, and they continue to evolve today. One consistent driver, however, has been the pursuit of increased profit through reduced labor costs.155 In a world characterized by tremendous income inequality, the differences in wages between countries allow multinational companies to bring their products to market for much less by producing abroad, often by contracting with logistics firms and suppliers in the low-wage country.156 In so doing, these firms move production outside the reach of the domestic legal regimes that would otherwise bind them, not coincidentally to contexts where wages are much lower and enforcement of domestic employment law is generally weak.

The past few decades have seen a groundswell of media attention highlighting the thousands of workers for such suppliers who have been injured and killed in factory fires, building collapses, or chemical disasters, and the much larger number paid a pittance for working to the point of exhaustion.157 Until the mid-1990s, most multinational corporations responded to efforts calling on them to take responsibility for these conditions by deflecting responsibility for workers they did not directly employ. A Nike spokeswoman summed up this sentiment perfectly when she stated, in response to unwelcome attention to the poor conditions in Indonesian factories producing its sneakers: “We don’t make shoes.”158 The message was clear: since Nike didn’t directly employ those workers, it should bear no responsibility for what happened to them.

In response, advocates and scholars have outlined the mechanisms through which brands and retailers control production in their transnational supply networks, demonstrated the relationship of multinational companies’ contractual requirements to the harms that workers face, and critiqued the legal landscape that allows large firms to avoid liability for poor treatment of those who labor to create their products.159 As consumers took up activists’ demands that major brands act to address problems in their supply chains, the brands began adopting codes of conduct, making commitments to baseline workplace standards. It is important to understand these interventions in the product supply chain as a first step in considering possible approaches to the human supply chain.

1.      Lessons from the Regulation of the Product Supply Chain

The mid-1990s through the 2000s were times of optimism about the potential of self-regulation to fill in these gaps in the law.160 Evidence has mounted, however, that multinationals’ voluntary codes of conduct have had little impact on poor working conditions overseas.161 With few exceptions, global firms have not been willing to invest the money and time in their suppliers necessary to achieve better conditions, and courts have held that almost none of their commitments are enforceable.162 Multinationals continue to reward subcontractors for delivering fast turnaround, on short notice, at low cost, rather than for incurring extra expenditures to achieve quality output while increasing safety and pay rates.163 As political scientist Richard Locke writes in his recent book, summing up his and others’ extensive work on the study of private regulation in supply chains, “[a]fter more than a decade of concerted efforts by global brands and labor rights NGOs alike, private compliance programs appear largely unable to deliver on their promise of sustained improvements in labor standards in the new centers of global production.”164 In light of these results, the pendulum is swinging back to a call for public intervention to supplant or complement private regimes.165

A number of advocates and scholars have argued over the past decade that where a firm decides to subcontract production and other functions of its business, while structuring and funding its contracts in a way that functionally sets terms and conditions for its suppliers’ workers, some form of liability for the payment and treatment of those workers should follow.166 Where both the lead firm and its subcontractors are located in the same country, and where there is a strong culture of rule of law and a functional labor inspectorate and court system, there is the possibility of translating these approaches into enforceable legal reforms.167 Within the United States, for example, some academics have suggested—and some litigators and policymakers have advanced—broadening the definition of “employer” in various labor and employment laws to encompass a wider range of work relationships, including an expansion of the “joint employer” doctrine that currently allows claims to proceed against two or more enterprises under limited circumstances.168 Although the standard varies across U.S. employment laws, a finding of joint employment has often required a determination that the employer directly controlled the terms of employment of its contractor’s workers.169 Recently, however, the U.S. Department of Labor (“DOL”) and National Labor Relations Board (“NLRB”) have both moved toward a more expansive standard that permits a finding of joint employment on a demonstration of indirect control. For example, in 2016 the Administrator of the DOL Wage and Hour Division affirmed that indirect rather than direct control could be used to demonstrate joint employment for the purposes of wage and hour laws.170 In the 2015 Browning–Ferris case, the NLRB expanded the joint-employment standard applicable in cases under the NLRA to include indirect control as well.171

Others have called for wider use of statutory tools that do not require a showing of joint employment, such as section 215(a)(1) (“Hot Goods” provision) of the FLSA of 1938, which permits the DOL to seize any goods produced in violation of the protections of the Act.172 This seizure has an immediate impact on the brands or retailers waiting for delivery of the products in order to sell them. As the DOL delicately notes in its public fact sheet: “[f]or minimum wage and overtime violations, a manufacturer or retailer, which may be under no legal obligation to pay back wages, may choose to pay the back wages owed by the employer to achieve compliance and allow shipment of the goods.”173

With few exceptions, U.S. labor and employment law does not apply transnationally.174 In its absence, the scholarly literature on labor standards in transnational product supply chains offers several insights worth highlighting.

First, the lead firm must bear the primary responsibility for making its suppliers’ compliance with its private code of conduct feasible and their noncompliance a meaningful impediment to an ongoing business relationship. Under many circumstances, lead firms in product and service supply chains have the power to dictate and monitor their subcontractors’ processes down to the minute.175 This capacity translates to the power and ability to monitor compliance with labor standards as well. To translate its workplace guarantees into functional protections, the lead firm must put enough additional money into the supply chain to cover the cost of improvements in pay and conditions that its code requires, so suppliers who make those changes are not economically disadvantaged in their ability to bid for future contracts.176 Lead firms rarely bother to do this.177

Even if they did, money alone would not be enough. The production demands the lead firm makes of its subcontractors (like rapid turnaround time for orders, last minute changes, and availability of workers for surge production) often undermine the possibility of compliance with provisions in the code regarding safety, hours worked, and so on.178 Lead firms must modify these demands if suppliers are to comply with the code. Ultimately, suppliers will hesitate to make costly shifts in their practices unless the lead firm demonstrates its long-term commitment to the supplier, which will often require practical support and guidance from the lead firm as well. Lead firms that have spurred meaningful changes in the practices of their suppliers often put a great deal of time into building relationships, helping subcontractors to increase both productivity and compliance over time.179

Second, goodwill has proven to be an insufficient motivator for lead firms to do what is necessary to translate self-regulation into actual improvement. This should not come as a surprise, given that the demands on the lead firm are time-consuming and expensive, and require changes in consumer expectations as well. The presence of an external incentive, however, can shift the firm’s economic calculus. Clear and enforceable commitments in its code of conduct can be one such incentive, but they are vanishingly rare.180 In the context of the global production chain, pressure is more likely to come from credible threats to brand reputation generated by workers, activists, and consumers. So far, mere public awareness has not been enough. For the firm to be persuaded to act in response to the incentive, consequences for noncompliance must be market based, consistently applied, and significant enough to change both lead firm and subcontractor behavior.181 An effective monitoring and enforcement system is an essential component of such a system.182 But consumer pressure is only a viable strategy where there is a large, reputation-sensitive brand at the apex of the supply chain, with fans inclined to care about labor rights.183 It has a better chance of success if the brand is a centralized purchaser with a large market share, in an industry with a high level of concentration.184 In the absence of these characteristics, self-regulation alone rarely produces meaningful and sustainable results.

The final insight is more hopeful. Scholars have begun to question the assumption on which an entire generation of work on labor standards in the global economy is founded: that the governments in countries to which brand or large retailers move production are unwilling or unable to play a meaningful role in addressing workplace violations in the global export sector. In recent studies of workplace standards at suppliers for brands and retailers located in Jordan, Indonesia, Cambodia, Brazil, and the Czech Republic, among others, political scientists have often found improvement in working conditions in settings where domestic regulators and private enforcement entities are both active, to a degree that has been difficult to achieve when either is present alone.185 This appears to hold true even when the domestic labor inspectorate is underfunded and the two efforts are not coordinated.186

2.      Are These Insights Applicable to the Regulation of Human Supply Chains?

How much can we learn about regulating labor recruitment from insights developed in the context of the product supply chain? Both involve subcontracting across borders to facilitate economic activity. On the other hand, there are significant differences between globally mobile people and goods, reflected in the very different regulatory regimes governing trade and migration.

At a distance, an economic analysis views the human supply chain as fundamentally the same as the product supply chain. Both move a factor of production (capital and raw materials, as well as finished goods, in the case of product supply chains; labor in the case of human supply chains) from one country to another where it can be used more efficiently. Both are made up of a network of contracting relationships that stretch across borders. The incentives that have given rise to human and product supply chains both grow from comparative advantage in the global context—the idea that each country will benefit from specializing in what it can offer most cheaply to the global economy, be that coal or workers, and that global growth will be maximized when firms are permitted to draw resources and labor from wherever they are cheapest.187 In a global economy with low barriers to trade and migration, it would make little difference in terms of wages whether a lead firm in the United States moved production to Mexico (paying workers there a little more than the Mexican wage), or hires Mexican workers to come to the United States (paying them Mexican wages plus the minimum extra amount necessary to incentivize them to spend part of the year doing hard work away from home). Either way, the classical economic story goes, unregulated product and labor mobility would allow the U.S. firm and the Mexican worker both to benefit from the comparative advantage offered by lower wages in Mexico.

Yet the political economy of trade (the movement of products or services across borders) diverges from the political economy of migration (the movement of people across borders).188 Trade has generally received a warm welcome from governments and business leaders in both more- and less-developed nations, on the belief that it increases the global economic efficiency and the wealth it generates, while raising the income levels in poorer countries. Its broad acceptance is institutionally reflected in the global power of the World Trade Organization (“WTO”) and a wealth of binational and multinational agreements.

Labor migration is far more contested. Almost all lower-income countries would prefer that higher-income countries adopt more open labor migration policies, and development economists argue that the global returns on labor migration and its potential to diminish inequality between nations tower over those found through trade.189 Wealthier nations have resisted, preferring to unilaterally impose more restrictive regimes that protect their domestic labor markets and restrict access to citizenship and the benefits that go with it. This has limited transnational coordination of migration, particularly as to immigrants likely to work in low-wage occupations.190 Such laws have rendered almost all legal migration opportunities for low-wage workers temporary and tightly controlled. In the United States, for example, Congress has resolved the conflict between demands from business lobbyists to admit more temporary migrants and populist calls to limit immigration by creating temporary-work visa programs that forbid low-wage temporary migrants from changing jobs (even if they are abused) and offer them no path to permanent residence, much less citizenship.191 This disincentivizes migrant workers from reporting violations of wages and working conditions, because the worker knows that if she is fired, her visa terminates and she must return home or face deportation. These differences mean that a firm’s decision to produce goods in a lower-income country is encouraged by international trade law, while a migrant’s decision to sell her labor in a higher-income country is governed by a restrictive immigration regime unilaterally established by the wealthier country.

In both settings, law is a critical factor incentivizing decisions by lead firms to bring their products or services to market through global product and human supply chains. Labor and employment law in the United States, for example, applies almost exclusively to workers in jobs located in the United States,192 generally regulates the direct employer only,193 and covers only the period when the employment relationship is active. By moving production abroad and contracting with a supplier for its completion, a firm can avoid liability for abuses it would have been responsible for if it directly had employed the worker within the United States. Likewise, by choosing to hire a labor intermediary for recruitment, the firm escapes the liability for violations during the recruitment process that would rest squarely on its shoulders had its own human resources department done the work. In a context of global inequality, the combination of private contracting arrangements and public laws allow firms to benefit from the decision to outsource a firm function, without bearing the true cost.

Classical economists would take issue with the assertion that this aspect of subcontracting is a source of savings for the lead firm, because it assumes that intermediaries will include the cost of compliance and the risk of penalties for noncompliance in their bids to the lead firm.194 What this fails to take into account, however, is how competitive the market for such contracts is, and how very low many suppliers at the bottom of product or human supply chains estimate the risk of being penalized for falling below baseline standards imposed by either public law or private codes of conduct.195 Limited legal protections, lax enforcement, and undercapitalized firms all contribute to the accuracy of this assessment. In the product supply chain, this allows suppliers to seek a competitive advantage by bidding on terms too low to cover the cost of safety measures or minimum-wage payments.196 In the labor supply chain, it allows recruiters to charge migrants whatever fees and bribes the market will bear, without concern for the limits set by law. In these ways, the end user is not required to bear the risk of violations.

That risk does not disappear into the ether, but instead is passed down the chain until it lands on the workers at the bottom.197 The U.S. workers who once held these jobs, often at higher wages, become unemployed. The only people who will accept the work under these conditions are from a lower-income country who measure the value of the job in the United States against one at home, and see in that equation a route to getting ahead in the long term—a goal only possible when dangerous work and a low salary by U.S. standards pays off in a place where the dollars go much further.198 Thus, where transnational human and product supply chains intersect, brands and other lead firms have access to a labor force structured to be even cheaper than global wage differentials would indicate.199

Many of the justifications for holding lead firms responsible for the actions of their subcontractors in the human supply chain do indeed track arguments made in the product supply context. As the Apple Fresh example illustrates, when an employer decides to outsource its recruitment function, it reduces its costs while retaining functional control.200 Presser has the power to correct the problems in its recruitment chain by changing recruiters or demanding more of its current one, and by paying more to cover the actual cost of recruitment, in order to replace the fees recruiters currently receive from workers.201 Likewise, Apple Fresh has the power to address the working conditions at Presser and the recruitment issues in Mexico, because it retains the ability to switch processors or demand that Presser use a different recruitment firm.202 The lead recruiter in Mexico has the power to decide which, if any, agents and brokers it will work with to locate migrants. In each case, it is both fair and effective to align that power with legal responsibility.

Nonetheless, human supply chains differ from product supply chains in ways that impact approaches to their regulation. The U.S. employers that hire their low-wage labor force through transnational recruiters may be further removed from a recognizable brand name than are global manufacturers.203 Recruiters have even lower barriers to entry than garment contractors, since they require almost no capital or equipment to go into business.204 Recruitment firms may not have a fixed location, and interactions between recruiters and workers usually occur out of sight, making the recruitment industry considerably harder to inspect for violations than global manufacturing.205

The most significant distinctions, however, lie in the different market dynamics and business models of recruiters and product suppliers. One of the biggest differences between the two is that recruiters have fewer options for generating revenue than their counterparts in global manufacturing. Suppliers in product chains may be able to increase efficiency and productivity, rather than raise prices. Recruitment, which is essentially priced per head, does not lend itself to such gains. If lead recruitment agencies are required to reduce what they charge migrants, one option might be to charge employers more. But this is only possible if all recruiters do the same; otherwise, employers will turn to cheaper competitors. The remaining options repackage, but do not lower, what workers ultimately pay. One is for the recruitment agency to increase volume and distance itself from violations by contracting work out to brokers, whose only source of income may be what they charge workers. Alternatively, the lead recruiter may just label what was once the recruitment fee as a charge for specific services (e.g., copying documents), or simply demand a cash payment off the books.206 Either way, migrants’ financial burden is not mitigated, and may indeed increase with added subcontracting and informality.

Another major distinction is that the central argument for liability in the product supply context—that lead firms control what their suppliers pay workers and how they treat them—does not translate well to the context of recruitment.207 In the human supply chain, the amount the lead firm pays the recruiter only partially determines how much the recruiter charges workers.208 As noted earlier, limited competition among recruiters in some areas, migrants’ lack of accurate information, and the absence of effective regulation allow recruiters to extract rents from migrants, who know that without the assistance of an intermediary they have little chance of getting access to a work visa and job abroad. It is easy for the U.S. employer to plead ignorance (or actually remain ignorant) of the fees and bribes that its migrant workers have paid its recruiters and their agents. Indeed, the structure of U.S. regulations of the H-2 programs, which impose liability on employers for recruitment fees only if they knew or should have known of recruitment violations, and requires firms to contractually forbid their recruiters to charge fees, encourages firms to look the other way and gives them a legal out for doing so.209 This flaw has been fatal to most cases in the United States seeking to hold employers liable under these provisions.

Nonetheless, whether or not employers know that recruiters charge fees to the migrants they hire, they benefit significantly when such fees are part of the arrangement. Most obviously, migrants’ fees subsidize employers’ recruitment costs.210 In addition, the high-interest loans that migrants take on to pay fees and costs, and the fear of being fired before being able to repay the lender, much less send money home or meet their target for savings, makes them reluctant to refuse an employer’s demands for harder work or to protest violations of the law.211 From the perspective of an employer seeking a compliant workforce, the subservience that results from recruitment under these conditions makes temporary migrants even cheaper and more attractive. From the perspective of both migrants and native workers in overlapping labor markets, it is a threat to the possibility of decent work.

C.     Regulating the Human Supply Chain: Direct, Joint, and Chain Responsibility

As this Article notes in Part IV.A, the predominant approach to governing the human supply chain to date has been direct regulation of recruiters by the origin-country government. This has not been successful. The market for recruitment services is driven on the demand side by the firms that hire temporary migrant workers and the terms set by the restrictive immigration laws where those firms are located, both outside the origin government’s control. On the supply side, factors already noted allow recruiters around the world to continue demanding high fees for access to work, rendering the limits set by law all but irrelevant. In addition, when lead recruitment firms use layers of subcontractors to reach down into remote communities, they recreate the human supply chain within origin countries. Lead recruitment firms are not systematically held accountable for their subcontractors’ abuses in any origin country. It is time for a different intervention.

1.      Applying Shared Liability to the Human Supply Chain

Shared liability along the supply chain should act as a complement to direct regulation of recruiters, not a replacement. One variation of supply-chain liability addresses only the immediate dyad—for example, the U.S. employer and the Mexican recruiter with which it has a contract. This is generally referred to as joint and several liability.212 The other variation is full-chain liability. Full-chain liability considers all of the actors above the renegade contractor—including the brand or end user at the very top of the chain—potentially liable for the actions of that contractor and its subcontractors.213 Although it may be more politically expedient to push for narrow liability, this Article argues for strict liability applied to the full chain, on the grounds of benefit, foreseeability of harm, and ability to incentivize change while spreading the cost.

Full-chain liability for recruitment abuses can be enforced at three points: against the brand or other major firm at the top of an integrated product or service supply chain, against the temporary migrant worker’s employer, and against the lead recruitment firm that hired her. For each actor, the liability would be for all violations below that actor in the chain. This Article’s focus is on the middle level: holding employers responsible for harms arising in the course of recruitment through a chain of intermediaries. The strict-liability approach that this Article ultimately recommends can also be applied to the lead recruiter for all of the violations committed by agents and brokers below it, and to the brand or retailer for all recruitment abuses in its entire supply chain.

The frameworks proposed by labor and employment scholars are relevant to the human supply chain context, but they must be adapted in order to be effective. For example, in employment law, joint liability requires a showing that the lead firm and its contractor are joint employers.214 This may be difficult to demonstrate in the setting where the lead firm has hired a recruiter (rather than a staffing agency) to supply it with migrant workers. The core element of any joint-employer inquiry is a demonstration of the lead firm’s direct or indirect control over the employees of its supplier, or of those workers’ economic dependency on the lead firm. But the relationship is reversed in the human supply chain. In the recruitment context addressed here, the “lead firm” is the legal employer of the workers. The question is whether the intermediary abroad that recruits the workers for the lead firm should be considered a joint employer with that firm, such that the lead firm could be held liable for the recruiter’s violations.

To hold that recruiters are joint employers with the firms that contract them would demand a substantial restructuring of the usual employment law framework. This restructuring would require a broader understanding of the time frame and range of places and actors that are relevant to the employment relationship. For recruited migrants, the terms of work (particularly their inability to protest wages and working conditions) are set during their interactions with the village-level recruitment agent and lender, long before the employer formally hires the employee on his or her arrival. Conceptually, it is not difficult to see how recruiting workers is an employment function.215 Indeed, whether the recruiter or the employer itself does the recruitment—and the recruitment activities are located within the United States—it clearly is.216 But unless the recruiter violated the law at the employer’s express direction, courts have rejected claims against the employer where the recruiter was a separate entity in another country.217

This situation illustrates the impediments posed by the fact that with regard to place and actors, the employment law of the country or state in which the workplace is located does not ordinarily extend to entities beyond those borders.218 Economic interactions elsewhere, between recruiters and would-be migrants abroad, and the workers’ resulting debt and fear of retaliation, sharply limit their ability to resist abusive conditions once they arrive on the job, and thus lower their price as a factor of production. These interactions continue when, at the demand of the employer, recruiters or their agents appear in the destination country—virtually and sometimes in person—to threaten migrants who protest labor conditions. Expanding employment law to encompass the actions of recruitment entities abroad, however, would require an understanding of the globalization of labor markets and a willingness to extend employment law transnationally that courts in the United States have so far refused to adopt. But the nature of the human supply chain illustrates the impossibility of regulating a transnational system through a unilateral law with only domestic reach.219

On the surface, agency law is an appealing alternative approach to the joint employer framework. After all, recruiters work for and carry out the instructions of the firms that hire them to contract migrants, consistent with the lay meaning of “agent.” Yet agency theories have not prevailed as to recruitment fees.220 Demonstrating an agency relationship between the recruiter and the principal employer requires proof that the recruiter agreed to act “on the principal’s behalf and subject to the principal’s control.”221 Furthermore, courts have required a demonstration that the employer authorized the recruitment fees.222 These requirements raise the same difficulty as the joint-employment case, because it would be nearly impossible to prove that the employer actually controlled or even was aware of the recruiter’s actions at issue.

There is more that could be done to flesh out the argument that recruiters in origin countries contracted by firms in destination countries should be considered joint employers with, or agents of, the firms for which they recruit. But moving these arguments forward is likely to be an uphill battle, due to the difficulty of demonstrating control and the obstacles to applying U.S. law transnationally. Even if they succeed, a finding of joint employment or agency only creates responsibility between two of the links in what may be a very long supply chain.

2.      Strict Liability as a Possible Approach

A statutory tort regime may offer a better way to achieve shared responsibility for recruitment abuses in the human supply chain. In the product-supply-chain context, a few employment and labor scholars, including Alan Hyde and Brishen Rogers, have recently explored the concept of third-party liability in tort as an approach to holding lead firms liable for working conditions at their supplier factories.223 Both scholars, however, emphasize the lead firm’s functional control over the wages and working conditions at the bottom of the product or service chain as a precursor for holding it responsible.224 In the absence of a clear parallel to that control in the human supply chain, it is necessary to explore other arguments for tort liability in the recruitment context.

The word “tort,” from the French, means “wrong.” Most tort law seeks to compensate victims for the wrongs of those who have harmed them. The strict-liability branch of tort law, however, allocates responsibility to actors without regard to fault, even to those that have carried out an action that society on the whole encourages, in order to address outcomes of that activity that are seen as socially unfair or undesirable.225 Strict liability still relies on a notion of control.226 Yet as Stephen Perry has defined control in advancing what he terms an “avoidability-based [theory] . . . of outcome-responsibility,”227 it is capacious enough to include the decision of an employer to contract with a recruiter: “[I]f an agent knew of the possibility that his action would or might cause a certain outcome, and if he had the capacity to act otherwise . . . then he could have avoided bringing the outcome about.” 228 The actor should also reasonably be able to foresee the harm. This is close to the formulation advanced by scholars who argue for “enabling torts,” which assign liability to a party where its actions help create the conditions for another actor to harm a third party.229

Requiring a firm to bear the cost of its recruiter’s actions, even if the recruiter was not acting at the firm’s direction or even with its tacit consent, is appropriate for three reasons. When a firm chooses to use a recruiter:
(1) the firm benefits from costs savings beyond those permitted by law;
(2) there is a substantial and foreseeable risk of harm; and (3) the firm is in a unique position to incentivize recruiters to change their behavior.

i.     Benefit from Costs Savings Beyond Those Permitted by Law

The mere existence of low-wage migrant worker visa programs allows firms to go outside the domestic labor market to realize savings from the wage differentials between countries. This is the aspect of the program that neoclassical economists would characterize as efficient.230 Beyond this, all firms that hire H-2 workers benefit financially from the requirement in the
H-2 programs that the migrant worker is bound to her sponsoring employer and so cannot seek a better job elsewhere. This provision, by contrast, is antithetical to the concept of a free market.231 These aspects of the U.S. migrant visa program are normatively problematic from the perspective of resident workers and migrants themselves—but this Article simply labels them “cost savings authorized by law.”

It is the firm’s decision to contract with a labor intermediary, rather than to handle recruitment internally, that brings about the benefit. This decision should be subject to strict liability. It is independent from the choice to hire workers from the H-2 program. The rents that recruiters extract from migrants reduce what the employer must pay for hiring and retaining labor.232 They depress the migrants’ effective wages and reduce their ability to demand that the employer comply with labor and employment laws, imposing subordination through the fear of inability to repay debt and of losing collateral to foreclosure. The firm does not have to use a recruiter to find workers; it could send its own employees to Mexico, as a significant minority appear to do,233 thus internalizing the costs of recruitment and the liability for any violations that occur during the process.234

Indeed, even when a firm chooses to contract with a recruiter, all of its savings should not necessarily be redistributed to workers. The fact that a firm might choose to contract with an intermediary located in Mexico in order to reap the benefits of the intermediary’s proximity to would-be migrants, Spanish-language fluency, familiarity with the visa process, and access to reliable transportation, is not in itself a negative from the worker’s perspective, and is beneficial to the firm and intermediary both. Several other employment-law scholars also expressly seek only to regulate the aspects of subcontracted employment that have a subordinating impact and/‌or negative distributional effects. As Simon Deakin has argued, requiring employers to eschew all labor intermediaries “rules out certain uses of the agency form which might well be legitimate in the sense of being beneficial to all the parties concerned.”235 A strict-liability regime of chain responsibility would not prohibit employers from benefitting from these efficiencies. It would merely require that they compensate workers when their decision to contract out migrant-worker recruitment resulted in violations of the migrants’ rights under Mexican and U.S. law.236

 ii.     Substantial and Foreseeable Risk of Harm

When a firm chooses to engage a third party to recruit migrant workers, it brings about a substantial and foreseeable risk of harm to the migrants. High fees and interest rates, false promises, and threats are pervasive in the global recruitment industry.237 When a recruiter working in the employer’s service demands fees or bribes, the workers who are delivered to the employer come bearing significant debt coupled with a fear of retaliation—burdens that further suppress their willingness to report violations in their wages and treatment. Given how pervasive these abuses are, it should be possible to demonstrate tacit, industry-wide knowledge of the risks that come with contracting a recruiter. This reliance on risk-spreading rather than knowledge or control as a justification for strict liability finds support in Judy Fudge’s call to “move[] beyond a fault-based model of [allocating] responsibility” in global supply chains, toward “enterprise responsibility based on the commitment to risk absorption and spreading inherent in the concept of social solidarity.”238 She continues, “[t]he idea is that enterprises should share the risks inherent in socially useful activity . . . based on an analysis of the risks created by the enterprise’s activities.”239 This is the theory behind other strict-liability provisions in employment law, for example, the Hot Goods provision of the FLSA.240

 iii.     Incentives for Recruiters to Discontinue Their Rent-Seeking Behavior

Although origin-country governments may be well-positioned to regulate recruitment, they have not been successful in doing so. In the absence of effective regulation, recruiters respond to two things: the economic rents that wage differentials make possible, and the demands of firms abroad. Although employers do not control the fees that recruiters and their agents charge, they do control the supply of jobs on which recruiters depend. If employers refuse to contract with intermediaries that violate the law, it will create economic incentives for recruiters to avoid such behaviors.241 Meanwhile, firms at the top of the supply chain are also in the best position to redistribute the costs of compliance.242 They can decide whether to absorb these costs or pass them along to consumers.

From a policy perspective, a strict-liability regime holding U.S. employers liable for recruitment abuses is consistent with fundamental commitments to basic workplace standards reflected in U.S. labor, employment, and immigration laws, and in the policies of the agencies designated to enforce them.243 Excusing employers from responsibility for illegal fees and bribes charged by their recruiters runs counter to the stated policy purpose of the H-2 programs, which is to permit U.S. employers to hire migrant workers from abroad only when there are genuine labor-market shortages in the industry or occupation in question, not simply when they would prefer to obtain workers who will work for less than the market wage.244

Legislation imposing strict liability on U.S. employers for the violations of their recruiters abroad might also open the door to the complementary effects observed in the product-supply context where there is overlap between public regulation in the country of production and private regulatory efforts driven by the lead firm.245 A strict-liability regime should spur serious efforts by U.S. firms to demand compliance by the intermediaries in their human supply chains, with the potential to increase the Mexican government’s incentive to enforce its own laws regulating recruitment practices. Likewise, if the Mexican government demanded compliance with its regulations on recruitment—which among other things require it to inspect and register recruiters and for recruiters to post a bond246—it would be much easier for U.S. firms to identify trustworthy recruiters.

In the near term, features of the human supply chain may actually make shared liability more feasible than it has been in the product-supply context, because of the location where the actual work is done. In the transnational product-supply-chain context, the legal employer and the place of employment are in the lower-wage country, while the lead firm is in the higher-wage country. But in the human supply chain, the firm that employs the worker, and the worker herself, are both in the higher-wage country. This means that the hard domestic law of the wealthier country governs the terms and conditions of the workers. The wealthier country can, if it wishes, make employers within its jurisdiction liable for the actions of their recruiters operating in other countries, if the firm’s choice to contract out the recruitment function interferes with the enforcement of domestic employment regulation. In destination countries with existing labor protections and a functioning enforcement system, this opens up the possibility of statutory solutions that would be difficult to apply to the transnational product-supply chain.

3.      Practice: Public, Private, and Hybrid Approaches

In 2013, I launched a new project researching efforts to regulate transnational labor recruiters. Since little had been written in the field, the endeavor required site visits and interviews with government, union, and private sector actors in multiple countries. Through this process, I ultimately identified new initiatives by governments and nongovernmental organizations in the Philippines, the Netherlands, the United Kingdom, and Canada as having the greatest potential for success.247 I also sought out bottom-up approaches to regulating recruitment developed by workers’ organizations rather than governments, and found three of note. I developed case studies of each of these public, private, and hybrid efforts, and offered an analysis of factors conducive to their effectiveness; the resulting working paper was published by the International Labor Organization in 2015.248 I draw on that research here to offer early indications of how increased employer liability is beginning to take shape in places around the world—though only centered on joint liability at present.

In the United States, advocates have begun to call for new regulation that would hold H-2 employers liable for recruitment abuses. The International Labor Recruitment Working Group has proposed incorporating penalties against employers for some recruitment abuses into federal immigration reform bills, and in state legislation as well, as have many of its member organizations in their individual work.249 Lawyers for temporary migrant workers have asked courts to use the FLSA to impose more responsibility on employers for paying recruitment costs, arguing that the recruiter was acting as the employer’s agent and/‌or that the charges were for the benefit of the employer, rather than the worker.250 The National Guestworker Alliance (“NGA”), in particular, has put a joint-and-several‑liability approach in the forefront of its legal and organizing efforts to address recruitment violations. Working with migrants on a range of visas, from the H-2B to the J visa (which is ostensibly for cultural exchange visitors but increasingly used as a source of low-wage labor),251 NGA has carried out high-profile campaigns targeting Hershey Foods, Signal Shipbuilders, and Wal–Mart, among others, demanding that the legislature hold them accountable for recruitment abuses and exploitation of migrants by their subcontractors.252

To date, however, it is only outside the United States that a few legal regimes systematically impose liability of any kind on employers for recruiter violations. The Philippines is one of a handful of origin countries that require recruiters and employers to share liability for each other’s actions.253 In practice, however, its lack of jurisdiction over employers has meant that the law has only been enforced against recruiters, and there only sporadically.254 Two efforts in destination countries have had more success in influencing the behavior of employers.

In Manitoba, Canada, a public regime run by the provincial Employment Standards Branch pursuant to the 2008 Worker Recruitment and Protection Act has created a stringent licensing scheme for recruiters operating in Manitoba and a registration mandate for Manitoba employers that wish to recruit foreign workers.255 All recruitment fees are banned. Employers must undergo a labor inspection before being permitted to hire a worker from abroad, and they are only allowed to use licensed recruiters in good standing to do so. If they employ an unlicensed recruiter, migrants can obtain from the employer any fees they were charged by the recruiter through a simple administrative process for recovering unpaid wages. This scheme is effective in part because it is run by the Employment Standards Branch, whose primary focus is enforcing worker rights.256 But it gains teeth through its intersection with Canadian federal regulations. Since 2011, unless the Manitoba Employment Standards Branch certifies an employer as fully in compliance with this law, the Canadian government will not grant the employer’s application for foreign worker visas. While statistical studies remain to be done, the head of this program reports that these penalties—and particularly the denial of visas—have moved employers toward in-house recruitment or the exclusive use of licensed recruiters.257

A quite different approach has been taken by a private regulatory effort in the Netherlands, which operates in interaction with that country’s public law. The default rule in the Netherlands is that employers are jointly liable for their subcontractors’ failure to contribute to payroll taxes, obey immigration laws, and, most recently, comply with minimum wage law.258 Recruitment and staffing firms are considered subcontractors under these provisions.259 Meanwhile, a foundation called Stichting Normering Arbeid (“SNA”), together with trade union and industry partners, has launched a completely voluntary private certification scheme for recruitment and staffing agencies.260 Compared to most private schemes, SNA’s approach to certification is fairly rigorous, with inspections twice a year that regularly result in probation and suspension.261 All results are posted on the web and sent to the lead firms that contract with such intermediaries. Meanwhile, Dutch law releases the lead firm from lawsuits based on the joint liability requirement if it hires workers through a labor intermediary certified by SNA.262 Lead firms, eager to avoid responsibility for their contractors’ violations, have increasingly demanded that their recruitment and staffing agencies participate in the SNA certification program.263 As a result, the certification is becoming mandatory as a function of the market, even though not required by law.264

All of these public or hybrid examples have joint liability at their core. For its part, full chain liability for recruitment abuses has only been achieved through worker collective action, rather than public or corporate-driven regulation. Returning to the United States, at least three union/‌civil-society regimes have used different organizing strategies to require end-user firms to take responsibility for conditions of recruitment.265 These efforts are all located in agriculture, an industry in which the human supply chain commonly delivers H-2 migrants to U.S. growers, who in turn are at the bottom of a product supply chain that moves the cucumbers, strawberries, and tomatoes that the migrants pick up to the brand(s) or retailer(s) at the top. Each of the three initiatives began by seeking to improve the wages and conditions of a subset of farm workers, by enlisting the support of consumers to demand that both the direct employer and the brand(s) come to the table.

As each campaign negotiated new working standards, however, it also demanded rules to govern recruitment in the human supply chain. Interestingly, the three efforts have taken completely different approaches to regulating labor supply, focusing on union contracts, recruitment training organizations, and union codes, respectively. In 2004, the Farm Labor Organizing Committee negotiated the first-ever U.S. collective bargaining agreement for H-2 workers with the North Carolina Growers Association.266 The union contract sets out detailed rules about the recruitment of H-2A workers in seniority order, eliminating the blacklist.267 In 2013, the United Farm Workers union helped to launch the Equitable Food Initiative, which currently applies to lettuce and strawberry suppliers for Costco.268 Its standards require no-fee recruitment, and it has founded its own fair recruitment and worker training organization located in Mexico, replacing private recruiters in the human supply chain bringing workers to jobs governed by the Initiative. Finally, since 2005, the Coalition of Immokalee Workers, in Florida’s tomato industry, has succeeded in persuading brands such as McDonalds, Whole Foods, Sodexo and Wal–Mart to join its Fair Food Program.269 Among many other provisions, the Program’s code bans labor intermediaries, requiring direct hire for all growers who sell tomatoes to those buyers.270

These last examples in particular offer a challenge to the notion that private regulation in the supply chain context must be brand-led. The Coalition of Immokalee Workers has termed its approach “Worker-Driven Social Responsibility,”271 in contrast with the usual corporate social responsibility framework made up of codes voluntarily adopted by multinational firms. Worker-Driven Social Responsibility initiatives emphasize collective action by workers and consumers in the process of reaching an agreement; meaningful enforcement of the terms of that agreement through intensive and independent inspection regimes; the imposition of market consequences for noncompliance; and worker engagement in all aspects of the process, from setting the standards to educating peers to monitoring compliance.272 Widespread implementation of this approach is critical if private regulation is truly to benefit workers, rather than act as window-dressing for brands.

All of the initiatives presented here address recruitment as an integral part of the structures of globalized work. They are based on an analysis of the economic dynamics of the supply chain relationships at issue, and seek to create negative market consequences for a brand’s or employer’s previously profitable behavior of disavowing the actions of the recruitment intermediaries that supply it with workers. The goal is to change the business calculations of the end user, employer, and recruiters all along the chain, re-shaping the recruitment market as a result. Nothing here implies minimizing the responsibility of both origin and destination states to regulate recruitment effectively because imposing liability works best as a complement to domestically enforcing labor standards. In addition, there is potential for constructive interaction between private and public initiatives at work in the same space, whether or not they are formally coordinated.273 It is too early to present more than anecdotal evidence regarding the outcomes of these initiatives. But their common perspective and early successes offer hope that a supply chain approach to regulating recruitment may one day be able to fill the damaging governance gap in this critically important arena of the global economy.

Finally, these case studies make clear that there is no one blueprint for success. An effective approach to regulation of recruitment will depend on the geography and political economy of each country in the corridor, and the presence or absence of civil society organizations that can support collective action by migrants and advance their rights as workers. It will be crafted around the distinct firm-level and labor-market structures in each country and industry.

V.     Conclusion

A network of actors in different places around the world produce consumer goods and services, and a similar network operates to move workers from one country to another to carry out that production. Ironically, what has kept most low-wage employers in the United States from engagement with this human supply chain has been the ready availability of undocumented workers in the local labor market. Today, however, as the number of undocumented immigrants falls and enforcement against employers who hire them increases, U.S. firms are turning to labor intermediaries abroad to deliver more legally authorized temporary migrant workers each year.

This Article identifies the link between a shift to dependence on migrant worker visas and a new reliance on the human supply chain. It notes the frequency with which human supply chains feed global product supply chains, and identifies key points of commonality and difference between them. The greatest concern with this shift is the threat to decent work for migrants and native workers alike that grows from these functionally unregulated relationships of labor supply. In response, the Article makes the case for a new strict liability regime to assign responsibility for recruitment violations to the most powerful actors in the human and product supply chains that feed today’s global economy.


 

  1. [1]. See infra notes 37–41 and accompanying text.

  2. [2]. See infra note 44 and accompanying text.

  3. [3]. Centro de los Derechos del Migrante, Monitoring International Labor Recruitment: A Cross-Visa Exploration of Regulatory Challenges 10­­–13 (2011), https://www.
    macfound.org/media/files/Monitoring_International_Labor_Recruitment.pdf.

  4. [4]. Megan Twohey et al., Wanted: Foreign Workers—and the Labor Brokers Accused of Illegally Profiting from Them, Reuters Investigates (Feb. 19, 2016, 3:20 PM), http://www.reuters.com/investigates/
    special-report/workers-brokers (“Since the 2007 fiscal year, for instance, Reuters found that intermediaries were involved in helping secure visas for 80 percent of the 2 million foreign workers approved for agricultural and other low-skill jobs.”). A survey of 382 Mexican H-2A migrants across seven states in Mexico by a coalition of five nongovernmental organizations puts the percentage at 94.5%. Jornaleros SAFE, Executive Summary: Mexican H2A Farmworkers in the U.S.: The Invisible Workforce 7–8 (2013), https://www.splcenter.org/sites/default/files/d6_legacy_files/do
    wnloads/resource/executive_summary_jornaleros_safe.pdf (discussing the number of interviewees and percentage using intermediaries). Studies that rely on government documents rather than worker interviews reach lower numbers. For example, a Centro de los Derechos del Migrante report based on Department of Homeland Security (“DHS”) applications noted that almost 62% of workers applying for H-2A and 84% of those applying for H-2B visas in 2011 reported having used a recruitment agent. Centro de los Derechos del Migrante, Recruitment Revealed: Fundamental Flaws in the H-2 Temporary Worker Program and Recommendations for Change 11 n.12, http://www.
    cdmigrante.org/wp-content/uploads/2013/01/Recruitment-Revealed_Fundamental-Flaws-in-the-H-2-Temporary-Worker-Program-and-Recommendations-for-Change.pdf (last visited Nov. 4, 2016). Likewise, the Government Accountability Office (“GAO”) states that approximately 44% of U.S. employers use formal intermediaries for H-2 visa recruitment. See U.S. Gov’t Accountability Office, H-2A and H-2B Visa Programs: Increased Protections Needed for Foreign Workers 26 (2015), http://www.gao.gov/assets/670/668875.pdf. The 44% number is based on the GAO’s review of H-2 petitions filed by employers with DHS. The GAO reports that, in addition to the 44%, “some employer industry groups” state that a significant number of employers deputize a foreman or returning migrant as their recruiter. Id. at 27.

  5. [5]. See Dovelyn Rannveig Mendoza, Migration Policy Inst., Regulating Private Recruitment in the Asia–Middle East Labour Migration Corridor 2 (2012), http://www.
    migrationpolicy.org/research/regulating-private-recruitment-asia-middle-east (finding that 80% of migrant workers in the Asia–Middle East corridor are recruited by intermediaries, rather than by employers directly).

  6. [6]. Undocumented migrants do often enlist the help of intermediaries at some point in their journey. However, because the intermediaries in these cases are facilitating entry and/or employment outside the situations permitted by immigration law, they are much less likely to have formal contractual arrangements with employers. In Mexico, at least, those intermediaries (often referred to as coyotes) commonly take the migrants across the border, but do not find them work. For an interesting economic analysis of the coyote business model, see generally Zachary Gochenour, The Political Economy of Immigration (Summer 2014) (unpublished Ph.D. dissertation, George Mason University), http://ebot.gmu.edu/bitstream/handle/1920/8962/
    Gochenour_gmu_0883E_10698.pdf.

  7. [7]. I refer to labor recruiters as “intermediaries,” “recruiters,” or “agencies,” and to the businesses that contract with them to obtain a migrant workforce as “employers” or “firms.”

  8. [8]. See infra notes 93–95 and accompanying text.

  9. [9]. In some settings, for example the H-2A visa for agricultural workers, U.S. courts have held that the migrants’ employers are legally obligated to bear these authorized charges. See Arriaga v. Fla. Pac. Farms, L.L.C., 305 F.3d 1228, 1232, 1245 n.26 (11th Cir. 2002) (finding
    H-2A workers can recover travel and visa costs under the Fair Labor Standards Act (“FLSA”) because they were incurred for the benefit of employers, but not recruitment fees because the workers could not show that the employer consented to have the recruitment fees demanded on their behalf). In other settings, for example the H-2B visa for other seasonal workers, U.S. courts have required the migrants to cover the visa and travel expenses. See Castellanos–Contreras v. Decatur Hotels, LLC, 622 F.3d 393, 400–04 (5th Cir. 2010) (finding H-2B workers cannot recover travel, visa, or recruitment fees from employer under FLSA). The U.S. Department of Labor’s long-litigated proposed H-2B rules would equalize this situation. See generally Temporary Non-Agricultural Employment of H-2B Aliens in the United States, 80 Fed. Reg. 24042 (Apr. 29, 2015) (to be codified at 8 C.F.R. pt. 214). Employers and H-2B advocates have been engaged in a protracted battle over regulations proposed by the U.S. Department of Labor that would increase the protections for H-2B workers, including equalizing this situation. Muzaffar Chishti et al., Recent Court Decisions Put a Sharp Spotlight on U.S. H-2B Temporary Worker Visa Program, Migration Pol’y Inst. (Apr. 23, 2015), http://www.migrationpolicy.org/article/recent-court-decisions-put-sharp-spotlight-us-h-2b-temporary-worker-visa-program.

  10. [10]. As employed in this Article, “rent” means the return to an economic actor that goes beyond the amount that the actor could charge if the market were perfectly competitive. See David Ricardo, On the Principles of Political Economy and Taxation 35–56 (1819).

  11. [11]. There is no agreed-upon measure of what amount migrants should pay for recruitment services. Migrant and human rights advocates argue that the fee should be zero. See, e.g., Int’l Labor Recruitment Working Grp., The American Dream Up for Sale: A Blueprint for Ending International Labor Recruitment Abuse 6 (2013), https://fairlaborrecruitment.files.
    wordpress.com/2013/01/the-american-dream-up-for-sale-a-blueprint-for-ending-international-labor-recruitment-abuse1.pdf; United Nations Office on Drugs & Crime, The Role of Recruitment Fees and Abusive and Fraudulent Recruitment Practices of Recruitment Agencies in Trafficking in Persons 76 (2015), https://www.unodc.org/documents/human-trafficking/2015/Recruitment_
    Fees_Report-Final-22_June_2015_AG_Final.pdf. The International Labor Organization of the United Nations would permit fees under some circumstances. See Int’l Labour Org., Private Employment Agencies Convention art. 7.1–7.2 (June 19, 1997), http://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:
    12100:0::NO::P12100_INSTRUMENT_ID:312326. A neoclassical economist would see the fee established by the market as efficient and therefore appropriate. Other approaches might set fees as a percentage of what a migrant in a particular migration corridor can earn by working abroad, or seek to rectify the problems of a usurious informal lending market rather than (or in addition to) focusing on the fee itself.

  12. [12]. United Nations Office on Drugs & Crime, supra note 11, at 9.

  13. [13]. Id.

  14. [14]. There is no explicit statement of this in the law governing the U.S. temporary migrant worker programs. It is derived from the fact that the worker’s visa is only valid for the term of his or her employer’s petition. “[A]n alien’s stay as an H-2A nonimmigrant is limited by the term of an approved petition.” 8 C.F.R. § 214.2(h)(5)(viii)(C) (2016). Petitions are approved for the petitioning firm only; the approval cannot be transferred to other employers. From the perspective of migrants themselves, there are pros and cons to accepting the terms of legal migration as structured by the U.S. temporary worker visa program. I recall in particular a set of conversations I had in 2006 with several men in Monterrey, Mexico, who were waiting to receive H-2A visas from the U.S. consulate there in order to travel to North Carolina to work in agriculture. All had been both H-2 workers and undocumented migrants in the past and suggested that they might migrate again without authorization in the future. To them, the H-2A program’s advantages included its promise of a season’s worth of steady work (or, in its absence, a guarantee of 75% of their salary) and the possibility of return the following year, its legality, and the certainty that they would be returning home at the season’s close to attend to family and business needs in Mexico. They were also clear-eyed about the perils of undocumented immigration: concern about whether they could cross the border without being caught, fear of deportation and its future impact on their ability to get a visa to the United States, concern that they might be separated from their families for a long period of time. But they noted that undocumented migrants were able to move between jobs and regions as H-2 workers were not, staying in an area when demand for their work was high and moving on as hours and wages began to fall, and therefore at times were more economically successful. Groups of undocumented workers shared cars, and had the independence to seek out lower prices for food and goods that H-2A workers—all but captive on their employers’ farms, living without transportation in employer-provided housing—did not. For that reason, they joked that H-2 workers sometimes referred to undocumented workers as “Wal–Marts.” Their astute analysis is an important reminder that migrants make choices in the process of migration—among a restrictive set of options, to be sure, but choices nonetheless—and that the labels that the law of immigration places on them often fail to capture their complex realities.

  15. [15]. Int’l Labor Recruitment Working Grp., supra note 11, at 14.

  16. [16]. See infra notes 99–101 and accompanying text.

  17. [17]. Int’l Labor Recruitment Working Grp., supra note 11, at 34.

  18. [18]. United Nations Office on Drugs & Crime, supra note 11, at 5–9. For examples of these practices in various countries, see id. at 10–14.

  19. [19]. Int’l Labor Recruitment Working Grp., supra note 11, at 50 n.3.

  20. [20]. See infra note 79 and accompanying text.

  21. [21]. See infra notes 80–87 and accompanying text.

  22. [22]. Regarding the ban on fees in México, see Ley Federal del Trabajo [LFT], art. 14, Diario Oficial de la Federación [DOF] 01-04-1970, últimas reformas DOF 12-06-2015 (Mex.). Regarding the United States ban on fees, see infra note 68.

  23. [23]. See infra note 148 and accompanying text.

  24. [24]. See infra notes 137–43 and accompanying text.

  25. [25]. See infra note 147 and accompanying text.

  26. [26]. See generally, e.g., Naomi Jiyoung Bang, Justice for Victims of Human Trafficking and Forced Labor: Why Current Theories of Corporate Liability Do Not Work, 43 U. Mem. L. Rev. 1047 (2013); Mariana C. Minaya, American Dreams, Trafficking Nightmares, 2 Tenn. J. Race, Gender, & Soc. Just. 64 (2014). See also the various policy papers advancing the bad actor framework in the migration context, such as Demetrios G. Papademetriou, Migration Policy Inst., Curbing the Influence of “Bad Actors” in International Migration (Transatlantic Council Statement) (2014), http://www.migration
    policy.org/research/curbing-influence-bad-actors-international-migration. For critiques of this perspective, see generally Janie A. Chuang, Exploitation Creep and the Unmaking of Human Trafficking Law, 108 Am. J. Int’l L. 609 (2014); Maria L. Ontiveros, Is Modern Day Slavery a Private Act or a Public System of Oppression?, 39 Seattle U. L. Rev. 665 (2016); and Hila Shamir, A Labor Paradigm for Human Trafficking, 60 UCLA L. Rev. 76 (2012).

  27. [27]. See generally, e.g., Andrew J. Elmore, Egalitarianism and Exclusion: U.S. Guest Worker Programs and a Non-Subordination Approach to the Labor-Based Admission of Nonprofessional Foreign Nationals, 21 Geo. Immigr. L.J. 521 (2007); Michael Holley, Disadvantaged by Design: How the Law Inhibits Agricultural Guest Workers from Enforcing Their Rights, 18 Hofstra Lab. & Emp. L.J. 575 (2001); Jennifer J. Lee, Private Civil Remedies: A Viable Tool for Guest Worker Empowerment, 46 Loy. L.A. L. Rev. 31 (2012). Interestingly, student notes have paid more attention to recruitment, although their focus has been on doctrinal fixes. See generally, e.g., Bryce W. Ashby, Note, Indentured Guests—How the H-2A and H-2B Temporary Guest Worker Programs Create the Conditions for Indentured Servitude and Why Upfront Reimbursement for Guest Workers’ Transportation, Visa, and Recruitment Costs is the Solution, 38 U. Mem. L. Rev. 893 (2008); Eleanor G. Carr, Note, Search for a Round Peg: Seeking a Remedy for Recruitment Abuses in the U.S. Guest Worker Program, 43 Colum. J.L. & Soc. Probs. 399 (2010); Michael Prasad, Note, Immigration and Labor Law­­—We Need Your Help! But It’s Gonna Cost You: Arriaga, Castellanos–Contreras, and Why Point of Hire Fees Should Be Paid by the Employer, 33 W. New Eng. L. Rev. 817 (2011).

  28. [28]. For one exception, albeit asking different questions than those raised here, see generally Kati L. Griffith, Globalizing U.S. Employment Statutes Through Foreign Law Influence: Mexico’s Foreign Employer Provision and Recruited Mexican Workers, 29 Comp. Lab. L. & Pol’y J. 383 (2008).

  29. [29]. See infra note 155 and accompanying text.

  30. [30]. See generally Constantinos C. Markides & Norman Berg, Manufacturing Offshore is Bad Business, Harv. Bus. Rev. (Sept. 1988), https://hbr.org/1988/09/manufacturing-offshore-is-bad-business. Drivers of transnational production other than reduced labor costs include the promise of escape from the regulatory environment of the lead firm’s home country, and savings growing from clusters of production. On the advantages of production clusters abroad, see Charles Duhigg & Keith Bradsher, How the U.S. Lost Out on iPhone Work, N.Y. Times (Jan. 21, 2012), http://www.nytimes.com/2012/01/22/business/apple-america-and-a-squeezed-middle-class.html. Drivers of transnational or domestic subcontracting and other fragmented or “fissured” structures of production, other than reduced labor costs, include the rise of the business philosophy that firms should focus on “core competencies” and shed all other functions in pursuit of improved efficiency, and the fact that technology has made it possible for the lead firm to coordinate and control its fissured production. David Weil, The Fissured Workplace: Why Work Became So Bad for So Many and What Can Be Done to Improve It 49–61 (2014). For a summary of the literature, see generally Annette Bernhardt, Labor Standards and the Reorganization of Work: Gaps in Data and Research (Inst. for Research on Labor and Emp’t, Working Paper No. 100-14, 2014), http://www.irle.berkeley.edu/workingpapers/100-14.pdf. This is, of necessity, a simplified account of the structures and institutions through which global production operates today. For example, it uses the term “supply chain” when a more accurate description of how most brands today structure production is “global production network,” a label that better captures the complexity of relationships between firms in different sectors. Gary Gereffi, The Global Economy: Organization, Governance, and Development, in The Handbook of Economic Sociology 160, 166, 170 (Neil J. Smelser & Richard Swedberg eds., 2005). Another missing element in this description is the critically important role of logistics firms in coordinating elements of transnational production and domestic subcontracting. Weil, supra, at 57–58, 160–67. For the “logistics revolution” from a labor-rights perspective, see Edna Bonacich & Jake B. Wilson, Getting the Goods: Ports, Labor, and the Logistics Revolution 3–22 (2008).

  31. [31]. Jennifer Gordon, Suburban Sweatshops: The Fight for Immigrant rights 43–44 (2005).

  32. [32]. See infra notes 124–28 and accompanying text.

  33. [33]. To my knowledge, this third approach has only been implemented through private regimes involving brands, suppliers, and unions or other workers’ rights organizations.

  34. [34]. Social scientists and legal scholars have offered a range of conceptual frameworks through which to understand global labor migration. Perhaps most relevant for the purposes of this Article is the idea of “global care chains.” For an overview and sympathetic critiques of the global care chain literature, see Ann Stewart, Gender, Law and Justice in a Global Market 30, 200–08 (2011); and Nicola Yeates, Global Care Chains: A State-of-the-Art Review and Future Directions in Care Transnationalization Research, 12 Global Networks 135, 136–41 (2012). In addition to these sources, see generally Judy Fudge, Global Care Chains, Employment Agencies, and the Conundrum of Jurisdiction: Decent Work for Domestic Workers in Canada, 23 Can. J. Women & L. 235 (2011).

  35. [35]. Dep’t of Statistics, Int’l Labour Org., ILO Global Estimates on Migrant Workers: Results and Methodology: Special Focus on Migrant Domestic Workers 5 (2015), http://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/documents/publication/wcms_
    436343.pdf.

  36. [36]. Id. at 25–34.

  37. [37]. Jeffrey S. Passel & D’Vera Cohn, Unauthorized Immigrant Population Stable for Half a Decade, Pew Res. Ctr. (Sept. 21, 2016), http://www.pewresearch.org/fact-tank/2016/09/21/unauthorized-immigrant-population-stable-for-half-a-decade [hereinafter Unauthorized Immigrant Population]; Jeffrey S. Passel & D’Vera Cohn, U.S. Unauthorized Immigration Flows Are Down Sharply Since Mid-Decade, Pew Res. Ctr. (Sept. 1, 2010), http://www.pewhispanic.org/2010/09/01/us-unauthorized-immigration-flows-are-down-sharply-since-mid-decade. At this point, undocumented workers made up 5.4% of the U.S. labor force. Jeffrey S. Passel et al., Pew Research Ctr., Share of Unauthorized Immigrant Workers in Production, Construction Jobs Falls Since 2007: In States, Hospitality, Manufacturing and Construction Are Top Industries 5 (2015), http://www.
    pewhispanic.org/files/2015/03/2015-03-26_unauthorized-immigrants-passel-testimony_REPORT.
    pdf. Undocumented immigrants have been a part of the U.S. workforce ever since immigration law defined such a group into existence. For an excellent history of undocumented immigrants in the United States, see generally Mae M. Ngai, Impossible Subjects: Illegal Aliens and the Making of Modern America (2004).

  38. [38]. Ana Gonzalez–Barrera, More Mexicans Leaving than Coming to the U.S., Pew Res. Ctr. (Nov. 4, 2015), http://www.pewhispanic.org/2015/11/19/more-mexicans-leaving-than-coming-to-the-u-s.

  39. [39]. Unauthorized Immigrant Population, supra note 37. These numbers include the recent surge in undocumented immigration to the United States by Central Americans. That increase masks the striking decline in Mexican undocumented immigration over the same time period. See also Robert Warren, US Undocumented Population Drops Below 11 Million in 2014, with Continued Declines in the Mexican Undocumented Population, 4 J. Migration & Hum. Security 1, 2, 9 tbl.4 (2016).

  40. [40]. Warren, supra note 39, at 2, 4 tbl.1.

  41. [41]. Gonzalez–Barrera, supra note 38.

  42. [42]. U.S. Dep’t of State, Table XVI(B): Nonimmigrant Visas Issued by Classification (Including Crewlist Visas and Border Crossing Cards): Fiscal Years 2010–2014, https://
    travel.state.gov/content/dam/visas/Statistics/AnnualReports/FY2014AnnualReport/FY14AnnualReport-TableXVIB.pdf (documenting annual increases in the H-2A and H-2B programs between 2010 and 2014); U.S. Gov’t Accountability Office, supra note 4, at 17; see also generally H-2A; H-2B, 22 Rural Migration News (Jan. 2016), https://migration.ucdavis.edu/rmn/more.php?id=1945.

  43. [43]. U.S. Gov’t Accountability Office, supra note 4, at 17.

  44. [44]. These figures include all legal temporary migrant workers, not just those in seasonal occupations. Numbers are hotly contested, and the data on visas kept by the Departments of State, Homeland Security, and Labor differ significantly. Ashwini Sukthankar, Glob. Workers Justice All., Visas, Inc.: Corporate Control and Policy Incoherence in the U.S. Temporary Foreign Labor System 29–31 (2012), http://www.globalworkers.org/sites/default/files/VisasInc_FINAL.pdf (noting the lack of coordination between departments); see also U.S. Dep’t of State, supra note 42; Jill H. Wilson, Immigration Facts: Temporary Foreign Workers, Brookings (June 18, 2013), http://www.
    brookings.edu/research/reports/2013/06/18-temporary-workers-wilson.

  45. [45]. Daniel Costa, Econ. Policy Inst., The H-2B Temporary Foreign Worker Program: For Labor Shortages or Cheap, Temporary Labor? 3 (Jan. 19, 2016), http://www.epi.org/
    files/pdf/97394.pdf.

  46. [46]. In response to employer demand, Congress effectively raised the cap on H-2B workers by 50,000 in a 2015 omnibus spending bill. H-2A; H-2B, supra note 42; see also Costa, supra note 45.

  47. [47]. The bill would have created a new low-wage visa category, the “W-2 Visa,” with up to 200,000 visas per year. See generally Border Security, Economic Opportunity, and Immigration Modernization Act, S. 744, 113th Cong. (2013) (as passed by the Senate on June 27, 2013). There is no consensus on how many additional migrant worker visas the bill would have created in other categories. For a summary of S. 744’s changes to H visa provisions, see generally Daniel Costa, Econ. Policy Inst., Future Flows and Worker Rights in S. 744: A Guide to How the Senate Immigration Bill Would Modify Current Law (2013), http://www.epi.org/files/2013/Future-Flows-and-Worker-Rights-Immigration-Bill.pdf (estimating that the bill’s increases in the H-1B program, and its changes in the H-2A and H-2B programs, would have eventually made at least an additional 200,000 new temporary labor migration visas available per year). The bill was defeated in the House. Employers also (and increasingly) use visas outside the H category to bring in temporary migrant workers. Sukthankar, supra note 44, at 18–23.

  48. [48]. Although higher-skilled temporary migrant workers are beyond the scope of this Article, it is worth noting that they suffer from serious recruitment and employment abuses as well. See Matt Smith et al., Job Brokers Steal Wages and Entrap Indian Tech Workers in US, Guardian (Oct. 28, 2014), https://www.theguardian.com/us-news/2014/oct/28/-sp-jobs-brokers-entrap-indian-tech-workers. Of late, most of the news coverage of H1-B visas has related to U.S. companies that have fired resident employees and replaced them with workers on H-1B visas. See Julia Preston, Large Companies Game H-1B Visa Program, Costing the U.S. Jobs, N.Y. Times (Nov. 10, 2015), http://www.nytimes.com
    /2015/11/11/us/large-companies-game-h-1b-visa-program-leaving-smaller-ones-in-the-cold.html.

  49. [49]. 8 U.S.C. § 1101(a)(15)(H)(ii)(a)–(b) (2012).

  50. [50]. 20 C.F.R. § 655.6(a) (2016) (indicating that the “employer... must establish that its need for non-agricultural services or labor is temporary, regardless of whether the underlying job is permanent or temporary” (emphasis added)); see also 8 C.F.R. § 214.2(h)(6)(ii)(B) (2016) (indicating the employer’s need is considered temporary if justified to the Secretary as either a one-time occurrence, a seasonal need, a peak-load need, or an intermittent need, as defined by the DHS).

  51. [51]. See Emp’t & Training Admin., U.S. Dep’t of Labor, Office of Foreign Labor Certification: H-2B Temporary Non-Agricultural Labor Certification Program—Selected Statistics, FY 2015 (2015), https://www.foreignlaborcert.doleta.gov/pdf/H-2B_Selected_Statistics_
    FY_2015_Q4.pdf (including a chart illustrating that the landscaping and grounds keeping industry accounted for 41% of H-2B labor certifications granted in 2015, while forestry workers, amusement park staff, and hotel housekeepers each received seven to eight percent of labor certifications during that period, and the bulk of the remaining visas were granted for construction workers, food packers and processors, cooks, wait staff, and animal caretakers outside of farms).

  52. [52]. U.S. Gov’t Accountability Office, supra note 4 at 17; see also generally Costa, supra note 45 (indicating that over the past decade, employer lobbying groups and the Chamber of Commerce have demanded a substantial increase in the number of low-wage migrant-worker visas as a part of any efforts to achieve comprehensive immigration reform).

  53. [53]. 20 C.F.R. § 655.122 (including regulations governing the H-2A program which require that employers pay an amount significantly above the federal minimum wage, guarantee 75% of the season’s salary even if there is not enough work, and provide free housing); see also Muzaffar Chishti et al., supra note 9 (discussing the debate between employers and H-2B advocates over regulations proposed by the U.S. Department of Labor that would increase the protections for H-2B workers); Ted Griggs, H-2B Guest-Worker Program Under Fire Over Salary Discrepancies, Advocate (Feb. 20, 2016, 12:52 P.M.), http://theadvocate.com/news/neworleans/neworleansnews/14698791-75/h-2b-guest-worker-program-under-fire-over-salary-discrepancies (citing Louisiana Agriculture Commissioner criticizing the H-2B hiring process as “laborious” and suggesting that H-2 workers should be more expensive than resident workers, not less).

  54. [54]. Centro de los Derechos del Migrante, supra note 4.

  55. [55]. See, e.g., Karen D. Johnson–Webb, Employer Recruitment and Hispanic Labor Migration: North Carolina Urban Areas at the End of the Millennium, 54 Prof. Geographer 406, 413–14 (2002); Nestor Rodriguez, “Workers Wanted”: Employer Recruitment of Immigrant Labor, 31 Work & Occupations 453, 466–67 (2004).

  56. [56]. Johnson–Webb, supra note 55, at 413–14; Rodriguez, supra note 55, at 466–67.

  57. [57]. Immigration Reform and Control Act of 1986 (“IRCA”), Pub. L. No. 99-603, 100 Stat. 3359. The IRCA amended the Immigration and Nationality Act (“INA”), Pub. L. No. 82-414, 66 Stat. 163 (1952) (codified as amended at 8 U.S.C. § 1101). Employer sanctions are specifically set out in section 274A of the INA. For instance, an employer could be criminally liable if it arranged for an unauthorized worker to be recruited abroad and brought to the United States. See INA § 274(a)(1)(A)(i) and (iv).

  58. [58]. H-2A Temporary Agricultural Workers, U.S. Citizenship & Immigr. Servs., https://www.uscis.
    gov/working-united-states/temporary-workers/h-2a-temporary-agricultural-workers (last visited Nov. 4, 2016) (discussing requirements of the H-2A visa); H-2B Temporary Non-Agricultural Workers, U.S. Citizenship & Immigr. Servs., https://www.uscis.gov/working-united-states/temporary-workers/h-2b-temporary-non-agricultural-workers (last visited Nov. 4, 2016) (discussing requirements of the H-2B visa).

  59. [59]. See supra notes 4–5.

  60. [60]. See supra note 4.

  61. [61]. For map of that complexity in one corridor, see Centro de los Derechos del Migrante, supra note 4, at 11­–12.

  62. [62]. For Más Labor, see The H2 Obstacle Course: Where the Finish Line Affects Your Bottom Line, másLabor: The H2 Lab. Specialists, http://maslabor.com/masLabor_OLD (last visited Nov. 4, 2016). CSI Labor Services was established by Stan Eury, who is also the head of the North Carolina Growers Association (“NCGA”). It is one of the largest H-2 agencies in Mexico, headquartered in Monterrey with offices around the country. CSI does not appear to have a web presence. Eury and CSI were the subject of a highly critical investigative report in 2015. Ken Bensinger et al., The Coyote, BuzzFeed News (Dec. 29, 2015, 7:20 PM), https://www.buzzfeed.com/kenbensinger/the-coyote. In addition to NCGA, a number of major U.S.-based farm labor recruitment entities subcontract to CSI. See Wash. Farm Labor Ass’n, Washington State Agricultural H-2A Guest Workers 2013 Annual Report 17 (2013), http://wafla.org/Resources/Documents/Employer%20Services/2013%20H-2A%20Annual%20Report.pdf. According to the Washington Farm Labor Association, CSI recruited 24,000 H-2 workers in 2014. See Wash. Farm Labor Ass’n, Washington State H-2A 2014 Annual Report 4 (2014), http://www.wafla.org/resources/Documents/Employer%20Services/2014_H-2A_Annual_Report/flippagebook.swf. Many other agencies are much smaller. See Centro de los Derechos del Migrante, supra note 4, at 15 (noting that Del-Al Associates recruits between 11,000 and 13,000 workers each year).

  63. [63]. Centro de los Derechos del Migrante, supra note 4, at 11–12.

  64. [64]. The recruitment-only case is more difficult because staffing companies are easier to hold liable for recruitment fees. By definition, staffing companies operate as employers in the United States, and U.S. law directly forbids U.S. employers from charging fees for recruitment. 20 C.F.R. § 655.135(j) (2016).

  65. [65]. For an overview of recruiters’ positive functions in the context of labor migration, see Dovelyn Agunias, Guiding the Invisible Hand: Making Migration Intermediaries Work for Development 10–17 (2009), http://hdr.undp.org/sites/default/files/hdrp_2009_22.pdf. For an overview of incentives to subcontract, see generally Bernhardt, supra note 30.

  66. [66]. For fee amounts, see infra notes 93–96 and accompanying text.

  67. [67]. For example, an employer can direct a recruiter to hire or re-hire particular migrants, or leave the selection of workers up to the agency.

  68. [68]. Employers must require “any foreign labor contractor or recruiter (or any agent of such foreign labor contractor or recruiter) whom the employer engages, either directly or indirectly, in international recruitment of H-2A workers” to sign a contract barring the recruiter from “seek[ing] or receiv[ing] payments or other compensation from prospective employees.” 20 C.F.R. § 655.135(k). The new Interim Final Rule that the Department of Homeland Security and Department of Labor implemented on April 29, 2015 provides similar protections to H-2B workers. 20 C.F.R.
    § 655.20(o)–(p). The 2008 Department of Labor H-2B regulations, which the U.S. District Court for the Northern District of Florida vacated on March 4, 2015, also provided similar protections. See Perez v. Perez, No. 3:14-cv-682, 2015 BL 60402, at *5 (N.D. Fla. Mar. 4, 2015).

  69. [69]. Memorandum from Nancy J. Leppink, Acting Adm’r, U.S. Dep’t of Labor, to Reg’l Adm’rs & Dist. Dirs. (May 6, 2011), https://www.dol.gov/whd/fieldbulletins/fab2011_2.pdf. A “knew or reasonably should have known” standard is not necessarily problematic. Id. It could support systematic enforcement under the theory that fees are so common in the recruitment industry that employers should be presumed to be on notice of them, particularly when they sign a contract with a recruiter requiring the employer to pay an amount that is insufficient to permit the recruiter to cover the costs of compliance with the law. A similar theory animates the California “Brother’s Keeper” law, which makes an employer liable for wage violations when it signs a contract to supply labor at a rate that is insufficient to cover compliance with wage and hour regulations. Cal. Lab. Code § 2810 (West 2016). The lack of evidence of effective enforcement of the H-2A provision suggests that the problem may be as much an issue of political will and resources as of the law itself.

  70. [70]. The primary case establishing employer liability under FLSA for reimbursement of H-2A workers’ travel costs and visa expenses (but not recruitment fees) is Arriaga v. Florida Pacific Farms, L.L.C., 305 F.3d 1228, 1244–46 (11th Cir. 2002). In Arriaga, the Eleventh Circuit recognized that “[w]orkers must be reimbursed during the first workweek for pre-employment expenses which primarily benefit the employer, to the point that wages are at least equivalent to the minimum wage.” Id. at 1237. The court allowed H-2A workers to recover travel and visa costs under FLSA because they were incurred “for the benefit of the employer.” Id. at 1242 n.17. The court denied reimbursement for recruitment fees because the workers could not show that the employer “consented to have the recruitment fees demanded on their behalf.” Id. at 1245. But see Rivera v. Peri & Sons Farms, Inc., 735 F.3d 892, 898–99 (9th Cir. 2013) (allowing H-2A workers to recover recruitment expenses from employer under FLSA). Attempting to expand Arriaga to H-2B workers and to encompass recruitment fees, see En Banc Brief for Sec’y of Labor as Amici Curiae Supporting Plaintiffs-Appellees, Castellanos–Contreras v. Decatur Hotels, LLC, 622 F.3d 393 (5th Cir. 2010) (No. 07-30942), 2010 WL 3049082, at *2 (arguing that transportation and visa fees are “for the primary benefit or convenience of the employer” and therefore employees on H-2B visas charged for such fees should be able to recoup them as a deduction from wages under the FLSA). Both arguments were rejected by the Fifth Circuit in Castellanos–Contreras v. Decatur Hotels, LLC, 622 F.3d 393, 403 (5th Cir. 2010) (H-2B workers cannot recover travel, visa, or recruitment fees from employer under FLSA). But see Rivera v. Brickman Grp., Ltd., No. 05-1518, 2008 WL 81570, at *14 (E.D. Pa. Jan. 7, 2008) (allowing H-2B workers to recover recruitment fees from employer under FLSA where employer directed workers to use a particular fee-charging recruiter). For an overview of the legal arguments for holding employers liable for H-2A and H-2B recruitment fees, see Carr, supra note 27, at 405–10.

  71. [71]. See, e.g., Arriaga, 305 F.3d at 1241–46; Rivera, 2008 WL 81570, at *8–9.

  72. [72]. See Migration of Asian Workers to the Arab World 60 (Godfrey Gunatilleke ed., 1986) (“Corruption begins in the office of the recruiting agents. In theory, agents are to receive no payment from the migrants; they are paid a commission by the employers for recruiting on their behalf. But there is a large scope for economic rent in this transaction, as the demand for jobs far outstrips supply, and the agents take full advantage of it.”).

  73. [73]. On the phenomenon of return migrants inflating the expectations of would-be migrants, see Douglas S. Massey, Long Day’s Journey into Night: One Person’s Reflections on International Migration, 29 Qualitative Soc. 111, 114 (2006). But see David McKenzie et al., A Land of Milk and Honey with Streets Paved with Gold: Do Emigrants Have Over-Optimistic Expectations about Incomes Abroad?, 102 J. Dev. Econ. 116, 127 (2013) (finding significant under-estimation of unconditional and conditional labor earnings for would-be male emigrants).

  74. [74]. The difference in estimates depends on the methodology used. The 4:1 ratio is from Jus Semper’s Purchasing Power Parity (“PPP”) analysis of wages in the manufacturing sector. The Jus Semper Glob. All., Mexico’s Wage Gap Charts 18 (2015), http://www.jussemper.org/Resources/
    Labour%20Resources/WGC-AEM/Resources/WagegapsMexAEM.pdf. The 4:1 ratio replaced the 9:1 ratio in 2013. Id.

  75. [75]. Minimum Wages: July 2015 (EUR Per Month), Eurostat, http://ec.europa.eu/eurostat/
    statistics-explained/index.php/File:Minimum_wages,_July_2015_(%C2%B9)_(EUR_per_month)
    _YB16.png (last visited Nov. 4, 2016).

  76. [76]. Dilip Ratha et al., World Bank Grp., 22 Migration and Remittances: Recent Developments and Outlook 15 tbl.2 (2014). The hourly wage in South Korea for 2016 is 6,030 Won. South Korea Minimum Hourly Wages 1989–2016, Trading Econ., http://www.trading
    economics.com/south-korea/minimum-wages (last visited Nov. 4, 2016). The hourly wage in Vietnam for 2016 is 3,500 Dong. Vietnam Minimum Wages, Trading Econ., http://www.trading
    economics.com/vietnam/minimum-wages (last visited Nov. 4, 2016). The conversion rate for South Korean Won to Vietnamese Dong is roughly 19:1. See Exchange Rates for South Korean Won, Money Converter, http://themoneyconverter.com/KRW/Exchange_Rates_For_South_Korean_Won.aspx (last visited Nov. 4, 2016).

  77. [77]. See Philip Martin, Reducing Migration Costs and Maximizing Human Development, in 1 Global Perspectives on Migration and Development: GFMD Puerto Vallarta and Beyond 27, 31 (Irena Omelaniuk ed., 2012).

  78. [78]. See Douglas S. Massey, International Migration at the Dawn of the Twenty-First Century: The Role of the State, 25 Population & Dev. Rev. 303, 305 (1999).

  79. [79]. See infra note 96 and accompanying text. For an overview of the average fees in major migration corridors, see generally Open Working Grp. on Labour Migration & Recruitment, Recruitment Fees & Migrants’ Rights Violations, http://madenetwork.org/sites/default/
    files/Policy-Brief-Recruitment-Fees-Migrants-Rights-Violations.pdf (last visited Nov. 4, 2016). The fees charged to migrants headed for Qatar offers one example of the high costs of labor migration. For a detailed overview of how those fees vary from agency to agency and country to country, see generally Ray Jureidini, Migrant Labour Recruitment to Qatar: Report for Qatar Foundation Migrant Worker Welfare Initiative (2014), http://www.qscience.com/userimages/ContentEditor/140481
    1243939/Migrant_Labour_Recruitment_to_Qatar_Web_Final.pdf; see also Isaku Endo & Gabi G. Afram, The Qatar-Nepal Remittance Corridor: Enhancing the Impact and Integrity of Remittance Flows by Reducing Inefficiencies in the Migration Process 10 (2011), http://
    issuu.com/world.bank.publications/docs/9780821370506/27#/signin (analyzing the Nepal-Qatar migration corridor and finding that the average migrant pays $1,216 for recruitment costs, which is 2.5 times the average per capita GDP in Nepal (2009) and four to six times the monthly salary a migrant construction or service worker will earn in Qatar).

  80. [80]. Manolo I. Abella, The Role of Recruiters in Labor Migration, in International Migration: Prospects and Policies in a Global Market 201, 203 (Douglas S. Massey & J. Edward Taylor eds., 2004).

  81. [81]. Jureidini, supra note 79, at 29–30, 44–50; see also Abella, supra note 80, at 202 (“Where unemployment is high and jobs are scarce, recruiters have no difficulty finding workers who are willing to pay for placement.”).

  82. [82]. Other actors in the migration industry charge high fees for mandatory medical exams, job training, required insurance, and bribes to government officials. Banks are also key actors. Much of the fees that migrants report paying to recruiters is in fact dispersed among other actors in the recruitment industry. Recruiters often collect money from migrants to cover a range of legally permitted expenses and mandatory (if illegal) payments to officials, as well as for their own services and profit. I thank Katharine Jones, Senior Research Fellow at Coventry University, for this observation. For an overview of the concept of a “migration industry,” see generally Ninna Nyberg Sørensen & Thomas Gammeltoft–Hansen, Introduction, in The Migration Industry and the Commercialization of International Migration 1 (Thomas Gammeltoft–Hansen & Ninna Nyberg Sørensen eds., 2013). For an application of this concept to the relationship between the United States and Mexico, see Ruben Hernandez–Leon, The Migration Industry in the Mexico–U.S. Migratory System, Cal. Ctr. for Population Res. (2005), http://www.diplomatie.gouv.fr/fr/IMG/pdf/migrationindustry_mexico.pdf.

  83. [83]. In recent years, concern about abuses perpetrated by recruiters has soared. For just some of the advocacy reports since 2012 highlighting problems with recruitment and recommending reforms in the context of migrant worker visas in the United States, see generally Mary Bauer & Meredith Stewart, S. Poverty Law Ctr., Close to Slavery: Guestworker Programs in the United States (2013), http://www.splcenter.org/sites/default/files/downloads/publication/SPLC-Close-to-Slavery-2013.pdf; Centro de los Derechos del Migrante, supra note 4; Fay Faraday, Metcalf Found., Profiting from the Precarious: How Recruitment Practices Exploit Migrant Workers (2014), http://metcalffoundation.com/wp-content/uploads/2014/04/Profiting-from-the-Precarious.pdf; Int’l Lab. Recruitment Working Grp., supra note 11; Nat’l Guestworker All., Leveling the Playing Field: Reforming the H-2B Program to Protect Guestworkers and U.S. Workers, http://www.guestworkeralliance.org/wp-content/uploads/2012/06/Leveling-the-Playing-Field-final.
    pdf (last visited Nov. 4, 2016); Sukthankar, supra note 44, at 40–45; Alejandra Constanza Ancheita Pagaza & Gisele Lisa Bonnici, Quo Vadis? Recruitment and Contracting of Migrant Workers and Their Access to Social Security: The Dynamics of Temporary Migrant Labor Systems in North and Central America 40 (Instituto de Estudios y Divulgación sobre Migración, A.C., Working Paper No. 4, 2013), http://estudiosdem
    igracion.org/inedim2013/documentosypub/publicaciones/reclutamientoconportada.pdf.

  84. [84]. See supra note 9 and accompanying text.

  85. [85]. See supra note 9 and accompanying text.

  86. [86]. See supra note 4. For a discussion on the various types of illegal costs, see infra notes
    93–96 and accompanying text.

  87. [87]. Centro de los Derechos del Migrante, supra note 4, at 18; Jornaleros SAFE, supra note 4, at 14; U.S. Gov’t Accountability Office, supra note 4, at 30.

  88. [88]. Centro de los Derechos del Migrante, supra 4, at 18; Aroldo Palacios, Glob. Workers Justice All., Confiscación de Títulos de Propiedad en Guatemala Por parte de Reclutadores en Programas de Trabajadores Temporales con Visas H-2B 14–17 (2013), http://
    www.globalworkers.org/sites/default/files/Confiscacion%20de%20Titulos_Informe_Final.pdf (in Spanish only).

  89. [89]. Migration News, a journal published by UC Davis, summarizes this situation as it was presented at the United Nations High Level Dialogue on Migration and Development in 2013:

    As worker-paid fees and the opportunity costs of not working while preparing to go abroad mount, the wage wedge continues to promise high wages that will enable migration debts to be repaid quickly and savings to accumulate. Many migrant-sending governments mandate pre-departure training, where migrants often learn that their net earnings abroad will not be as high as they expected, and that high living costs abroad will make it harder to achieve savings targets. But by this time, most migrants have invested too much in migration to turn back, and they look forward to departure.

    HLD: Migration Costs, 21 Migration News, Jan. 2014, https://migration.ucdavis.edu/mn/more.php?id=3892_0_5_0.

  90. [90]. See, e.g., Centro de los Derechos del Migrante, supra note 4, at 22; Bauer & Stewart, supra note 83, at 41.

  91. [91]. See supra note 4.

  92. [92]. Julia Preston, Company Banned in Effort to Protect Foreign Students from Exploitation, N.Y. Times (Feb. 1, 2012), http://www.nytimes.com/2012/02/02/us/company-firm-banned-in-effort-to-protect-foreign-students.html; see also generally Palacios, supra note 88 (describing how recruiters in Guatemala routinely demand the deeds to migrant workers’ property at home as security for loans, and threaten to seize the properties in order to control migrants while abroad).

  93. [93]. Jornaleros SAFE, supra note 4, at 14.

  94. [94]. This amount was for recruitment services only, not travel, visa fees, or other charges. Centro de los Derechos del Migrante, supra note 4, at 8, 16.

  95. [95]. As of January 2016, the Mexican minimum daily wage is 73.04 pesos. See Minimum Wages in Mexico with Effect from 01-01-2016, WageIndicator.org, http://www.wage
    indicator.org/main/salary/minimum-wage/mexico (last visited Nov. 4, 2016).

  96. [96]. For one set of estimates of recruitment fees in different countries around the world, of which fees paid by Mexican migrants are among the lowest, see Open Working Grp. on Labour Migration & Recruitment, supra note 79, at 2–3.

  97. [97]. For H-2A workers, 20 C.F.R. § 655.120(l) (2016) requires employers to pay at least the “Adverse Effect Wage Rate” (“AEWR”), the average wage for livestock and field workers in that state as determined by the Department of Labor. In 2016, the AEWR is between $3 and $7 higher than the federal minimum wage. See Adverse Effect Wage Rates for 2016, 80 Fed. Reg. 79,614, 79,614–79,615 (Dec. 22, 2015). For H-2B workers, the employer must pay at least the “prevailing wage rate.” 20 C.F.R. § 655.10. The method of calculation of this rate is currently the subject of litigation. See supra note 68 and accompanying text.

  98. [98]. See supra note 14.

  99. [99]. Lauren A. Apgar, Econ. Policy Inst., Authorized Status, Limited Returns: The Labor Market Outcomes of Temporary Mexican Workers 17 (2015), http://www.epi.org/
    files/pdf/86437.pdf. The report concludes, “[o]verall, these findings indicate that benefits from the legal authorization given to temporary workers are offset by the inability of these workers to switch employers while working in the United States.” Id. at 20. It highlights the “unfair labor recruitment process” as another source of the impediments temporary migrant workers face. Id. at 21. Apgar also reported that H-2 workers “held jobs in occupations with educational scores 13 percent lower than the occupations held by unauthorized workers,” even though the H-2 migrants had higher education levels overall than their undocumented counterparts. Id. at 12. The education score is calculated based on “the proportion of U.S. workers in each occupation who completed one or more years of college.” Id. at 8.

  100. [100]. Costa, supra note 45, at 20 tbl.6.

  101. [101]. Id. at 22 tbl.8.

  102. [102]. Id. at 24 tbl.10.

  103. [103]. See supra notes 100–02 and accompanying text.

  104. [104]. New, at least, since the last great wave of immigration in the late 19th and early 20th centuries, when padrones recruited Italian and Greek immigrants for a range of low-wage work in the United States. Harris Freeman & George Gonos, Taming the Employment Sharks: The Case for Regulating Profit-Driven Labor Market Intermediaries in High Mobility Labor Markets, 13 Emp. Rts. & Emp. Pol’y J. 285, 304–05 (2009). Internally, recruiters also played important roles in the Great Migration of African Americans from the southern to northern United States. Joshua L. Rosenbloom, Looking for Work: Searching for Workers: American Labor Markets during Industrialization 46–114 (2002).

  105. [105]. See Cindy Hahamovitch, No Man’s Land: Jamaican Guestworkers in America and the Global History of Deportable Labor 3–6, 228, 234–35 (2011); see also generally Dovelyn Rannveig Mendoza & Kathleen Newland, Migration Policy Inst., Circular Migration and Development: Trends, Policy Routes, and Ways Forward (2007), http://www.migrationpolicy.org/research/
    circular-migration-and-development-trends-policy-routes-and-ways-forward.

  106. [106]. This pattern is particularly marked in the Gulf States, which rely on temporary migrant workers for most low-wage labor needs. For example, 84.8% of the population of the United Arab Emirates is made up of temporary migrant workers. See Froilan T. Malit Jr. & Ali Al Youha, Labor Migration in the United Arab Emirates: Challenges and Responses, Migration Pol’y Inst. (Sept. 18, 2013), http://www.migrationpolicy.org/article/labor-migration-united-arab-emirates-challenges-and-responses (“In 2013, the UAE had the fifth-largest international migrant stock in the world with 7.8 million migrants (out of a total population of 9.2 million), according to United Nations (UN) estimates.”). The corresponding figure in Qatar is 85.7%. Françoise De Bel-Air, Gulf Labour Mkts. & Migration, Demography, Migration, and Labour Market in Qatar 6 (2014), http://
    cadmus.eui.eu/bitstream/handle/1814/32431/GLMM_ExpNote_08-2014.pdf (“Indeed, in 2010, non-Qataris were estimated at 1,456,362 or 85.7 per cent of the total resident population of 1,699,435, a significant increase from previous decades.”). The use of recruiters in these areas is on the rise. Katharine Jones, Recruitment Monitoring & Migrant Welfare Assistance: What Works? 1–2 (2015), https://www.iom.int/sites/default/files/migrated_files/What-We-Do/docs/Recruitment-Monitoring-Book.pdf. Since the 1970s the numbers of recruitment agencies and brokers which expedite the international migration process have burgeoned in both Colombo Process Member States (“CPMS”) and destination states alike. Id. at 1–2. For instance, in Sri Lanka, “the number of [recruitment agencies has] increased five-fold between 1985 and 2005.” Id. at 17. In China, the number of agencies has grown from four (all large state-owned corporations) at the beginning of the 1980s, to over 3,000, now a mixture of both state-owned and private, by 2005. Id. at 48 tbl.8.

  107. [107]. This is not to say by any means that undocumented migration is a nonissue, or that the rights of the undocumented are no longer worth our attention. There are substantial undocumented populations all over the world, especially where wealthier and less wealthy countries are geographically proximate. In addition, the categories of “undocumented workers,” “authorized temporary migrant workers,” and “refugees” are overlapping and unstable. For just one example, Syrians who flee to the European Union seeking protection are likely to request refugee status, but if they are denied it, they may well become undocumented immigrants.

  108. [108]. See, e.g., World Bank Grp. & Int’l Monetary Fund, Global Monitoring Report 2015/2016: Development Goals in an Era of Demographic Change 203–08 (2016), https://openknowledge.worldbank.org/handle/10986/22547; Georgi Gotev, Migration Will Be Win-Win for World, Finds World Bank Report, EurActiv.com (Nov. 6, 2015), http://www.euractiv.com/
    section/development-policy/interview/migration-will-be-win-win-for-world-finds-world-bank-report.

  109. [109]. Labour Migration, Int’l Org. for Migration, https://www.iom.int/labour-migration (last visited Nov. 4, 2016).

  110. [110]. See generally Org. for Econ. Co-operation & Dev., Is Migration Good for the Economy? (2014), www.oecd.org/migration/OECD%20Migration%20Policy%20Debates%
    20Numero%202.pdf.

  111. [111]. In an earlier article, I contest this claim. See generally Jennifer Gordon, People Are Not Bananas: How Immigration Differs from Trade, 104 Nw. U. L. Rev. 1109 (2010). For another particularly trenchant critique, see generally Kerry Preibisch, Development as Remittances or Development as Freedom? Exploring Canada’s Temporary Migration Programs from a Rights-Based Approach, in Constitutional Labour Rights in Canada: Farm Workers and the Fraser Case 81 (Fay Faraday et al. eds., 2012).

  112. [112]. Chris Forde et al., Understanding the Connections Between Temporary Employment Agencies and Migration, 31 Int’l J. Comp. Labour L. & Indus. Rel. 357, 360–61 (2015).

  113. [113]. Mendoza, supra note 5, at 2.

  114. [114]. Dovelyn Rannveig Mendoza, of the Migration Policy Institute (“MPI”), has been publishing policy briefs on recruitment issues in the Asia-Middle East corridor since the mid-2000s; for a summary of her findings, see id. A number of excellent studies on recruitment abuses in different places around the globe have been issued in the past year alone. See generally Bassina Farbenblum et al., Migrant Workers’ Access to Justice at Home: Indonesia (2013), http://www.opensocietyfoundations.
    org/sites/default/files/migrant-worker-justice-indonesia-20131015.pdf; Jones, supra note 106; Jureidini supra note 79, at 39–44; Sarah Paoletti et al., Migrant Workers’ Access to Justice at Home: Nepal (2014), http://www.opensocietyfoundations.org/sites/default/files/migrant-nepal-report-english-20140610_1.pdf; Piyasiri Wickramasekara, Regulation of the Recruitment Process and Reduction of Migration Costs: Comparative Analysis of South Asia (2013).

  115. [115]. For one exception in legal scholarship, see Fudge, supra note 34, at 249–51, and text accompanying note 34. More attention has been paid by legal scholars to the role of staffing agencies (i.e., intermediaries that employ workers and provide their labor to other firms on a contractual basis) in violation of workers’ rights. See, e.g., Freeman & Gonos, supra note 104, at 287–92. Outside of legal scholarship, there is a small amount of work identifying recruitment intermediaries as part of global supply chains. See Forde et al., supra note 112; Stephanie Ware Barrientos, ‘Labour Chains’: Analysing the Role of Labour Contractors in Global Production Networks, 49 J. Dev. Stud. 1058 (2013).

  116. [116]. See supra note 71 and accompanying text.

  117. [117]. The United Nations defines “trafficking in persons” as:

    [T]he recruitment, transportation, transfer, harbouring or receipt of persons, by means of the threat or use of force or other forms of coercion, of abduction, of fraud, of deception, of the abuse of power or of a position of vulnerability or of the giving or receiving of payments or benefits to achieve the consent of a person having control over another person, for the purpose of exploitation. Exploitation shall include, at a minimum, the exploitation of the prostitution of others or other forms of sexual exploitation, forced labour or services, slavery or practices similar to slavery, servitude or the removal of organs.

    Protocol to Prevent, Suppress and Punish Trafficking in Persons Especially Women and Children, Nov. 15, 2000, art. 3(a) http://www.ohchr.org/EN/ProfessionalInterest/Pages/ProtocolTraffickingInPersons.aspx.

  118. [118]. Janie A. Chuang, Giving as Governance?: Philanthrocapitalism and Modern-Day Slavery Abolitionism, 62 UCLA L. Rev. 1516, 1544–47 (2015). The anti-trafficking approach also has a tendency to cast migrants as hapless “victims,” without recognition that they have exercised agency in the process of migration and also have the capacity to do so in response to exploitation. Chuang, supra note 26, at 612; Johan Lindquist, Images and Evidence: Human Trafficking, Auditing, and the Production of Illicit Markets in Southeast Asia and Beyond, 22 Pub. Culture 223, 226 (2010); Ben Rogaly, Migrant Workers in the ILO’s Global Alliance Against Forced Labour Report: A Critical Appraisal, 29 Third World Q. 1431, 1432 (2008). A critical part of the human-supply-chain perspective, by contrast, is an insistence on the recognition of migrant workers’ agency in the process of migration and in efforts to combat the problems that they face, even if both are often exercised within a very constrained range of options. These same critics argue for a view of the exploitation of migrants as a continuum, rather than a bright line between bad actors and everyone else. On the continuum in the context of undocumented workers, see Kathleen Kim, Beyond Coercion, 62 UCLA L. Rev. 1558, 1582 (2015).

  119. [119]. Chuang, supra note 118, at 1520; see also Ontiveros, supra note 26, at 17–18; Rogaly, supra note 118, at 1435, 1444. For an example of an advocacy report that takes the “bad actor” view, see generally Meghan Benton, Migration Policy Inst., Spheres of Exploitation: Thwarting Actors Who Profit from Illegal Labor, Domestic Servitude, and Sex Work (2014), http://www.
    migrationpolicy.org/research/migration-exploitation-illegal-labor-domestic-servitude-sex.

  120. [120]. KnowTheChain, https://www.knowthechain.org (last visited Nov. 4, 2016).

  121. [121]. The U.S. State Department offers a brief summary of the challenges. See generally Office to Monitor & Combat Trafficking in Persons, The Challenges of Certification in Addressing Forced Labor (2014), http://www.state.gov/documents/organization/228265.pdf.

  122. [122]. For an excellent analysis of the labor supply chain and its impact on working conditions, particularly where it intersects with product supply chains, see generally Jean Allain et al., Report: Forced Labour’s Business Models and Supply Chains (2013), http://pure.qub.
    ac.uk/portal/en/publications/report-forced-labours-business-models-and-supply-chains(67757
    de8-90e2-4154-bde4-bd34d839cc17).html.

  123. [123]. There is unfortunately no reliable data available on what percent of U.S. low-wage temporary migrants work for companies in supply chains topped by a major retailer or brand. However, agribusiness, food processing, and especially food retail stores have consolidated dramatically over the past two decades. Food & Water Watch, Consolidation and Buyer Power in the Grocery Industry 1 (2010), https://www.foodandwaterwatch.org/sites/default/files/consolidation_buyer_
    power_grocery_fs_dec_2010.pdf. It is thus reasonable to assume that almost all H-2A (agricultural) employers, and H-2B employers that process and pack food, sell their products to a large retail chain. Allain et al., supra note 122; see also generally Greg Asbed & Sean Sellers, The Fair Food Program: Comprehensive, Verifiable and Sustainable Change for Farmworkers, 16 U. Pa. J.L. & Soc. Change 39 (2013). For example, Hershey’s and Wal–Mart have both been singled out as having migrant worker recruitment abuses in their food supply chains. See generally Nat’l Guestworker All., Raising the Floor for Supply Chain Workers: Perspective from U.S. Seafood Supply Chains (2016), http://asia.floorwage.org/workersvoices/reports/raising-the-floor-for-supply-chain-workers-perspecti
    ve-from-u-s-seafood-supply-chains/view; Julia Preston, Pleas Unheeded as Students’ U.S. Jobs Soured, N.Y. Times (Oct. 16, 2011), http://www.nytimes.com/2011/10/17/us/hershey-foreign-exchange-students-pleas-were-ignored.html. Big names in the hospitality industry have had similar issues. See Ken Bensinger et al., The Pushovers, BuzzFeed (May 12, 2016, 2:06 PM), https://www.buzz
    feed.com/kenbensinger/the-pushovers (identifying Disney and Trump as two brands that hire H-visa workers directly).

  124. [124]. The following description and Part III.B are adapted from Jennifer Gordon, Global Labour Recruitment in a Supply Chain Context 12–17 (2015), http://www.ilo.org/public/
    libdoc/ilo/2015/115B09_97_engl.pdf.

  125. [125]. See supra notes 61–63 and accompanying text.

  126. [126]. See supra notes 86–88 and accompanying text.

  127. [127]. See infra note 130 (discussing the intersection of human and product supply chains in the context of consolidation of the food industry); see also Allain et al., supra note 122, at
    39–43.

  128. [128]. For a complex and informative set of diagrams illustrating product and labor supply chains and their intersection in the context of forced labor, see Allain et al., supra note 122, at 45–53.

  129. [129]. Id. at 42–43; Weil, supra note 30, at 9–10. But see Bernhardt, supra note 30, at 11–12.

  130. [130]. Many states exempt farm employers from the mandate to provide workers compensation insurance for job-related illnesses or injuries, but Washington State is not one of them. See Workers’ Comp. Servs., Wash. State Dep’t of Labor & Indus., Employers’ Guide to Workers’ Compensation Insurance in Washington State 2–3, http://www.lni.wa.gov/
    IPUB/101-002-000.pdf (last visited Nov. 4, 2016).

  131. [131]. Recruitment agencies play major roles in increasing the market for their own services. Forde et al., supra note 112, at 358, 360–61; Benjamin Hopkins & Chris Dawson, Migrant Workers and Involuntary Non-Permanent Jobs: Agencies as New IR Actors?, 47 Indus. Rel. J. 163, 176 (2016).

  132. [132]. The degree to which Presser and other actors profit depends on how much cost pressure there is from above them in the supply chain. If the top firm in the supply chain is demanding much lower prices from them, the move to using temporary migrant labor may just allow them to stay afloat.

  133. [133]. Gary Gereffi, Globalization and the Demand for Governance, in The New Offshoring of Jobs and Global Development 39, 47–49 (Int’l Institute for Labour Studies ed., 2006), http://unstats.
    un.org/unsd/trade/s_geneva2011/refdocs/RDs/Offshoring%20of%20Jobs%20(Gereffi%20-%20Dec%202005).pdf.

  134. [134]. Id. at 47.

  135. [135]. Id. at 47–49.

  136. [136]. Int’l Labour Org., Guide to Private Recruitment Agencies: Regulation, Monitoring and Enforcement 5–6 (2007), http://www.ilo.org/wcmsp5/groups/public/
    ---ed_norm/---declaration/documents/instructionalmaterial/wcms_083275.pdf.

  137. [137]. Robyn Magalit Rodriguez, Migrants for Export: How the Philippine State Brokers Labor to the World 116–40 (2010); see also Centro de los Derechos del Migrante, supra note 4, at 24; Jones, supra note 106, at 54–56; Paoletti et al., supra note 114, at 152–53.

  138. [138]. For specific examples of how this conflict plays out, see Rodriguez, supra note 137, at 126–130 (describing and analysing the Philippines government’s response to an incident with Filipino migrants to Brunei); and Jureidini, supra note 79, at 118–20 (describing one recruiter’s experience with the Banglaseshi government’s focus on increasing workers migrating abroad).

  139. [139]. See World Bank Grp., Migration and Remittances Factbook 5 (3d ed. 2016), http://
    siteresources.worldbank.org/INTPROSPECTS/Resources/334934-1199807908806/4549025-1450455807487/Factbookpart1.pdf.

  140. [140]. Clare Ribando Seelke, Cong. Research Serv., R43616, El Salvador: Background and U.S. Relations 13 (2016), https://www.fas.org/sgp/crs/row/R43616.pdf.

  141. [141]. Personal Remittances, Received (% of GDP), World Bank Grp., http://data.
    worldbank.org/indicator/BX.TRF.PWKR.DT.GD.ZS?locations=NP (last visited Nov. 4, 2016).

  142. [142]. For an example of such a reaction, see Nicole Constable, Maid to Order in Hong Kong: Stories of Migrant Workers 86–88 (2d ed. 2007).

  143. [143]. The Mexican human-rights organization ProDESC has recently broken new ground by convincing the Mexican government to inspect a recruitment agency for the first time, resulting in an administrative fine for multiple violations. A criminal case for fraud is pending. For a description of how this came about, and particularly of the roles migrant workers themselves have played in demanding accountability from the Mexican government, see Jennifer Gordon, Solidarity Ctr., The Transformation of Work: Challenges and Strategies—Roles for Workers and Unions in Regulating Labor Recruitment in Mexico (2015), http://www.
    solidaritycenter.org/wp-content/uploads/2015/05/Migration.Roles-for-Workers-and-Unions-in-Regulating-labor-Recruitment-in-Mexico.Jennifer-Gordon-Fordham.5.15.pdf.

  144. [144]. For an overview of Bilateral Labour Migration Agreements in OECD countries, see Daniela Bobeva & Jean-Pierre Garson, Overview of Bilateral Agreements and Other Forms of Labour Recruitment, in Migration for Employment: Bilateral Agreements at a Crossroads 11,
    11–12 (2004). For an overview within North and Central America, see Ancheita & Bonnici, supra note 83. For an overview between India and its migrants’ destination countries, see generally Piyasiri Wickramasekara, Something is Better than Nothing: Enhancing the Protection of Indian Migrant Workers Through Bilateral Agreements and Memoranda of Understanding (2012). For an overview between Nepal and its migrants’ destination countries, see Paoletti et al., supra note 114, at 88; and Gordon, supra note 108 at 1126–28.

  145. [145]. See Gordon, supra note 108, at 1128–29; see also Jones, supra note 106, at 8; Wickramasekara, supra note 144, at 4.

  146. [146]. The title of a recent study of India’s bilateral migration agreements by long-time migration expert and former ILO Migration Specialist Piyasiri Wickramasekara offers an apt description of their effectiveness in most other contexts as well. Wickramasekara, supra note 144.

  147. [147]. See Gordon, supra note 108, at 1127 (regarding the United States); see also Jureidini, supra note 79, at 3 (regarding Gulf States).

  148. [148]. Recently, some destination governments have begun to take more responsibility for addressing recruitment fees. See infra Part IV. A number of destination countries beyond those noted in Part IV have laws on the books that forbid or limit recruitment fees. See supra note 68 and accompanying text (discussing U.S. law regarding migrant recruitment fees). The problem in those cases is not the law, but inadequate or ineffective enforcement. When pressed on the issue of enforcement, such governments tend to resort to the “no jurisdiction” response. The reality is more that there is a lack of political will to do what is necessary to enforce them. One obstacle that would need to be overcome, for example, is the lack of coordination between destination country agencies charged with addressing trafficking, those overseeing labor recruitment and migration, and those that enforce labor standards. See, e.g., Jones, supra note 106, at 68–69; Fudge, supra note 34, at 244–46.

  149. [149]. The International Convention on the Protection of the Rights of All Migrant Workers and Members of Their Families protects a wide array of migrant rights, but no major destination country has ratified it. International Convention on the Protection of the Rights of All Migrant Workers and Members of Their Families, Dec. 18, 1990, 45 U.N. 158, http://www.ohchr.org/
    EN/ProfessionalInterest/Pages/CMW.aspx; International Convention on the Protection of the Rights of All Migrant Workers and Members of their Families, United Nations, https://treaties.un.
    org/Pages/ViewDetails.aspx (last visited Nov. 4, 2016) (listing the current signatories of the convention).

  150. [150]. The ILO Private Employment Agencies Convention (“Convention 181”) and the accompanying Recommendation sets parameters for recruitment. Private Employment Agencies Convention, June 19, 1997, ILO No. 181, http://www.ilo.org/dyn/normlex/en/f?p=NORMLEX
    PUB:12100:0::NO::P12100_INSTRUMENT_ID:312326. While Article 7.1 of Convention 181 prohibits recruitment fees, Article 7.2 permits signatory governments to allow fees for some types of workers and recruitment-related services. Id. at art. 7.2. Exceptions to the provision of Article 7.1 must be “[i]n the interest of the workers concerned, and after consulting the most representative organizations of employers and workers.” Id. at art. 7.1. For a list of countries that have ratified the Convention, see Ratifications of C181 - Private Employment Agencies Convention, 1997 (No. 181), Int’l Labour Org., http://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:11300:0::NO::P11300_
    INSTRUMENT_ID:312326 (last visited Nov. 4, 2016). Finally, the 2014 ILO Forced Labour Protocol and its accompanying Recommendation emphasizes the importance of combatting abusive and fraudulent recruitment of workers as a critical part of the effort to prevent forced labor. See generally Int’l Labour Org., Protocol to the Forced Labour Convention, 1930 (2014), http://www.ilo.org/wcmsp5/groups/public/---ed_norm/---relconf/documents/meetingdocument/
    wcms_248900.pdf (containing provisions on recruitment in Article 2); Int’l Labour Org., Recommendation on Supplementary Measures for the Effective Suppression of Forced Labour (2014), http://www.ilo.org/wcmsp5/groups/public/---ed_norm/---relconf/documents/meetingdoc
    ument/wcms_248908.pdf (containing specific proposals for state action to combat recruitment fraud and abuse in Supplementary Measure 8).

  151. [151]. See supra Part IV (discussing governance gaps).

  152. [152]. There is a healthy debate over whether what this Article terms the “routine abuses of labor recruitment” should also be understood, and addressed, as forced labor or debt bondage. See, e.g., Klara Skrivankova, Joseph Rowntree Found., Between Decent Work and Forced Labour: Examining the Continuum of Exploitation 17, 22 (2010), http://www.gla.gov.uk/media/1585/
    jrf-between-decent-work-and-forced-labour.pdf; Chuang supra note 26, at 639; Rogaly, supra note 118, at 1436; see also generally Int’l Labour Org., A Global Alliance Against Forced Labour: Global Report Under the Follow-up to the ILO Declaration on Fundamental Principles and Rights at Work 2005, http://www.ilo.org/wcmsp5/groups/public/---ed_norm/---declaration/documents/
    publication/wcms_081882.pdf; Rebecca Smith, Guest Workers or Forced Labor?, 16 New Lab. F. 70 (2007). The distinction in many cases is legally important, because more remedies are available for forced labor than for violations of statutes about recruitment, where they exist, and also politically important as a way to frame the issue publicly. It is not necessary to decide this question in order to decry these conditions as unfair to migrants, and to believe that replacing them with more equitable arrangements is an important goal.

  153. [153]. United Nations Conference on Trade and Dev., World Investment Report 2013: Global Value Chains: Investment and Trade for Development 135 (2013), http://unctad.org/
    en/publicationslibrary/wir2013_en.pdf (stating how “UNCTAD estimates that about 80 per cent of global trade (in terms of gross exports) is linked to the international production networks of TNCs”).

  154. [154]. Tatyana P. Soubbotina & Katherine A. Sheram, Beyond Economic Growth: Meeting the Challenges of Global Development 66 (2000).

  155. [155]. Weil, supra note 30, at 10–11. Drivers of transnational production other than reduced labor costs include the promise of escape from the regulatory environment of the transnational corporation (“TNC”)’s home country, and savings growing from clusters of production. On the advantages of production clusters abroad, see Charles Duhigg & Keith Bradsher, How the U.S. Lost Out on iPhone Work, N.Y. Times (Jan. 21, 2012), http://www.nytimes.com/2012/01/22/business/apple-america-and-a-squeezed-middle-class.html. Drivers of transnational or domestic subcontracting and other fragmented or “fissured” structures of production, other than reduced labor costs, include the rise of the business philosophy that firms should focus on “core competencies” and shed all other functions in pursuit of improved efficiency, and the fact that technology has made it possible for the lead firm to coordinate and control its fissured production. Weil, supra note 30, at 49–58, 60–61; see also Annette Bernhardt, Labor Standards and the Reorganization of Work: Gaps in Data and Research 10–11 (Inst. for Research on Labor & Emp’t, Working Paper No. 100-14, 2014), http://www.irle.berkeley.edu/working
    papers/100-14.pdf.

  156. [156]. Again, this is a simplified account of a complex phenomenon. See Bonacich & Wilson, supra note 30.

  157. [157]. The two situations that have received the most media attention in the past few years are the Rana Plaza garment factory collapse in 2013, and a series of worker suicides, chemical explosions, and child labor allegations at Foxconn, the Chinese firm to which Apple contracts much of its production. For descriptions of each, see David Barboza, After Suicides, Scrutiny of China’s Grim Factories, N.Y. Times (June 6, 2010), http://www.nytimes.com/2010/06/07/
    business/global/07suicide.html; Jim Yardley, Report on Deadly Factory Collapse in Bangladesh Finds Widespread Blame, N.Y. Times (May 22, 2013), http://www.nytimes.com/2013/05/23/world/
    asia/report-on-bangladesh-building-collapse-finds-widespread-blame.html.

  158. [158]. The quote is from Martha Benson, who at the time represented Nike in Asia. Tim Larimer, “Sneaker Gulag: Are Asian Workers Really Exploited?”, Time Int’l 30, May 11, 1998, cited in Debora L. Spar, Hitting the Wall: Nike and International Labor Practices 1, 2 (Harvard Business School Case 9-700-047, Sept. 6, 2002) (on file with author).

  159. [159]. Richard M. Locke, The Promise and Limits of Private Power: Promoting Labor Standards in a Global Economy 12–17 (2013) (describing how the production demands TNCs make of their suppliers conflict with the commitments in their codes of conduct, sending market signals that make clear that fast turnaround and low price are the highest priority); Weil, supra note 30, at 184–201 (explaining how the law fails to regulate fragmented workplace relations).

  160. [160]. Archon Fung et al., Can We Put an End to Sweatshops? 4 (Joshua Cohen & Joel Rogers eds., 2001).

  161. [161]. Locke, supra note 159, at 20; Mark Anner et al., Toward Joint Liability in Global Supply Chains: Addressing the Root Causes of Labor Violations in International Subcontracting Networks, 35 Comp. Lab. L. & Pol’y J. 1, 5 (2013); Dara O’Rourke, Outsourcing Regulation: Analyzing Nongovernmental Systems of Labor Standards and Monitoring, 31 Pol’y Stud. J. 1, 20–23 (2003); Declan Walsh & Steven Greenhouse, Inspectors Certified Pakistani Factory as Safe Before Disaster, N.Y. Times (Sept. 19, 2012), http://
    www.nytimes.com/2012/09/20/world/asia/pakistan-factory-passed-inspection-before-fire.html.

  162. [162]. The Ninth Circuit dismissed claims brought by overseas workers against Wal–Mart based on Wal–Mart’s code of conduct, on the grounds that corporate commitments expressed in a code of conduct did not create rights for third parties. Doe I v. Wal–Mart Stores, Inc., 572 F.3d 677, 681 (9th Cir. 2009). Hopes that the Alien Tort Claims Act could be applied to hold U.S. firms liable for workplace violations in their supply chains were dashed by Kiobel v. Royal Dutch Petroleum Co., 133 S. Ct. 1659, 1669 (2013). For a useful exploration of other possible theories of enforcement for codes of conduct, see James J. Brudney, Envisioning Enforcement of Freedom of Association Standards in Corporate Codes: A Journey for Sinbad or Sisyphus?, 33 Comp. Lab. L. & Pol’y J. 555, 576–97 (2012). The Rana Plaza collapse was the spur for one of the first multi-stakeholder initiatives through which TNCs made expressly enforceable commitments as to improvements in the working conditions at their subcontracted factories. Statement on the Occasion of the One-Year Anniversary of the Rana Plaza Building Collapse, Accord on Fire & Building Safety in Bangl., (Apr. 23, 2014), http://bangladeshaccord.org/2014/04/the-one-year-anniversary-of-the-rana-plaza-building-collapse [hereinafter Statement]. Benjamin Hensler & Jeremy Blasi offer a useful analysis of the advances in the Accord. Benjamin Hensler & Jeremy Blasi, Making Global Corporations’ Labor Rights Commitments Legally Enforceable: The Bangladesh Breakthrough, Worker Rts. Consortium (June 18, 2013), http://www.cleanclothes.org/resources/recommended-reading/
    making-global-corporations2019-labor-rights-commitments-legally-enforceable-the-bangladesh-breakthrough.

  163. [163]. Locke, supra note 159, at 13–14.

  164. [164]. Id. at 20.

  165. [165]. Id. at 156–73; Matthew Amengual, Complementary Labor Regulation: The Uncoordinated Combination of State and Private Regulators in the Dominican Republic, 38 World Dev. 405, 406 (2010); Kevin Kolben, Dialogic Labor Regulation in the Global Supply Chain, 36 Mich. J. Int’l L. 425, 427 (2015); Kevin Kolben, Integrative Linkage: Combining Public and Private Regulatory Approaches in the Design of Trade and Labor Regimes, 48 Harv. Int’l L. J. 203, 205–06 (2007); Paul Verbruggen, Gorillas in the Closet? Public and Private Actors in the Enforcement of Transnational Private Regulation, 7 Reg. & Governance 512, 514 (2013); see also Tim Bartley, Transnational Governance as the Layering of Rules: Intersections of Public and Private Standards, 12 Theoretical Inq. L. 517, 523 (2011) (stating that “a more promising [view of private regulation] involves paying attention to its substantive intersections with domestic law, regulation, and other rules”); Salo V. Coslovsky & Richard Locke, Parallel Paths to Enforcement: Private Compliance, Public Regulation, and Labor Standards in the Brazilian Sugar Sector, 41 Pol. & Soc’y 497, 498 (2013) (“Recent research has suggested that... a combination of private and public interventions is necessary to promote labor standards in globally dispersed supply chains.”).

  166. [166]. See e.g., Alan Hyde, Legal Responsibility for Labour Conditions Down the Production Chain, in Challenging the Legal Boundaries of Work Regulation, 83, 92–97 (Judy Fudge et al. eds., 2012); see also Mark Barenberg, Widening the Scope of Worker Organizing: Legal Reforms to Facilitate Multi-Employer Organizing, Bargaining, and Striking 55–56 (2015), http://
    rooseveltinstitute.org/wp-content/uploads/2015/10/Widening-the-Scope-of-Worker-Organizing.pdf; Jeremias Prassl, The Concept of the Employer 42–54 (2015) (discussing current systems of UK and EU employer regulation and the five functions of an employer that could be a basis for an expanded view of liability under these systems); Catherine Ruckelshaus et al., Nat’l Emp. Law Project, Who’s the Boss: Restoring Accountability for Labor Standards in Outsourced Work 38–40 (2014), http://www.nelp.org/content/uploads/2015/02/Whos-the-Boss-Restoring-Accountability-Labor-Standards-Outsourced-Work-Report.pdf. See generally Simon Deakin, Commentary, The Changing Concept of the ‘Employer’ in Labour Law, 30 Indus. L.J. 72 (2001); Judy Fudge, Fragmenting Work and Fragmenting Organizations: The Contract of Employment and the Scope of Labour Regulation, 44 Osgoode Hall L.J. 609 (2006). A rise in subcontracting and other structures that distance the end user from responsibility for the employment of workers, such as franchising and misclassification of employees as independent contractors across industries, have spurred these proposals. Some industries have longstanding traditions of subcontracting, including agriculture, garment, and construction, and some of the solutions gaining currency today have been tried in those contexts. For a particularly interesting examination of the history of joint liability in the garment industry, see generally Anner et al., supra note 161.

  167. [167]. For a review of joint liability provisions in United States employment laws, see Ruckelshaus et al., supra note 166, at 33–38; David Weil, Improving Workplace Conditions Through Strategic Enforcement: A Report to the Wage and Hour Division 13, 30, 80–83 (2010), http://www.dol.gov/whd/resources/strategicEnforcement.pdf.

  168. [168]. California has led the way among the states in joint employment liability. For example, AB 1897, signed into law in 2014, makes it easier to hold firms liable for wage and workers compensation violations related to workers with whom they have a contracting rather than employment relationship. Cal. Lab. Code § 2810.3(b)(1)–(2) (West 2016).

  169. [169]. Airborne Express, 338 N.L.R.B. 597, 597 n.1 (2002) (noting the requirement of direct control as a central part of the NLRB joint employer analysis). The NLRB later overruled the direct-control requirement in Browning–Ferris Industries of California, Inc., 362 N.L.R.B. 186, 15 (2015) (“We return to the traditional test used by the Board . . . . The Board may find that two or more entities are joint employers of a single work force if they are both employers within the meaning of the common law, and if they share or codetermine those matters governing the essential terms and conditions of employment.”).

  170. [170]. U.S. Dep’t of Labor, Wage & Hour Div., Administrator’s Interpretation No.
    2016-1: Joint Employment Under the Fair Labor Standards Act and Migrant and Seasonal Agricultural Worker Protection Act 11 (2016), http://www.dol.gov/whd/flsa/Joint_
    Employment_AI.pdf. As Administrator David Weil noted, consistent with the broad definition of joint employment embedded in the FLSA, wage and hour law looks to “the core question of whether the employee is economically dependent on the potential joint employer who, via an arrangement with the intermediary employer, is benefitting from the work” to determine whether the lead firm should bear joint liability for the direct employer’s violations. Id.

  171. [171]. Browning–Ferris Indus. of Cal., Inc., 362 N.L.R.B. at 9.

  172. [172]. Fair Labor Standards Act, 29 U.S.C. §§ 212(a), 215(a)(1) (2012). For a description of the Department of Labor’s application of the “hot goods” provision, see Weil, supra note 30, at 227–28.

  173. [173]. U.S. Dep’t of Labor, Wage & Hour Div., Fact Sheet #80: The Prohibition against Shipment of “Hot Goods” Under the Fair Labor Standards Act (2014), https://www.dol.
    gov/whd/regs/compliance/whdfs80.pdf. “Back wages” in this context refers to the difference between what the worker actually earned and the wage required by law.

  174. [174]. Unless Congress explicitly legislates to the contrary, U.S. labor and employment laws do not apply abroad, even to U.S. companies. See, e.g., Foley Bros. Inc., v. Filardo, 336 U.S. 281, 289–90 (1949) (FLSA does not apply to employees of U.S. firms who are stationed overseas); Comp. Sci. Raytheon, 318 N.L.R.B. 966, 968 (1995) (NLRA does not apply outside United States). In the employment arena, Congress has only taken such action once. In 1991, the Supreme Court held that the anti-discrimination provisions of Title VII of the Civil Rights Act of 1964 did not apply abroad. See E.E.O.C. v. Arabian Am. Oil Co., 499 U.S. 244, 246 (1991). Congress quickly amended the Act to make clear that it did intend to protect U.S. citizens who were employed by U.S. firms overseas. See Civil Rights Act of 1991, Pub. L. No. 102–166, § 101, 105 Stat. 1071 (codified at 42 U.S.C. § 1981 (1991)). The Age Discrimination in Employment Act, 29 U.S.C. § 621 (2012), and the Americans with Disabilities Act, 42 U.S.C. § 12101 (2012), extend similar protections.

  175. [175]. In other contexts, however, “mega-suppliers” such as FoxConn are challenging the dynamics of this power relationship, because they hold enough market share that brands must bargain with them to at least some extent. Such bi- or multi-polar value chains are particularly likely to emerge where there are fewer major suppliers in an industry than there are brands, as with the U.S. auto and electronics industries. See also Gary Gereffi, Global Value Chains in a Post-Washington Consensus World, in 21 Review of International Political Economy 9, 16–17 (2014).

  176. [176]. Anner et al., supra note 161, at 21–24, 35–36. Note that suppliers often do jobs for multiple brands. Locke, supra note 159, at 8. It is rarely feasible to make improvements that only impact the work for one supplier.

  177. [177]. An exception is the Fair Food Program of the Coalition of Immokalee Workers. The Fair Food Program, http://www.fairfoodprogram.org (last visited Nov. 4, 2016); see also Gordon, supra note 124, at 39–41. See generally Asbed & Sellers, supra note 123. Almost all U.S. garment brands have refused to sign the recent Bangladesh Accord on Fire Safety. Statement, supra note 162; see also Anner et al., supra note 161, at 27–31; see also Hensler & Blasi, supra note 162, at 3–5.

  178. [178]. Locke, supra note 159, at 126–55; Anner et al., supra note 161, at 8–14. I thank Mark Barenberg and Jeremy Blasi for their insights on this point, shared during separate conversations with the author.

  179. [179]. Locke, supra note 159, at 105–25.

  180. [180]. See supra notes 156–58 and accompanying text (noting the same exceptions).

  181. [181]. Asbed & Sellers, supra note 123, at 45–46; Gordon, supra note 124, at 46–47.

  182. [182]. The past decade has seen the development of a rich body of scholarship on monitoring and enforcement, particularly in the political science literature. While it is beyond the scope of this Article to capture the insights it offers, interested readers should see Locke, supra note 159; see also generally Mark Barenberg, Toward a Democratic Model of Transnational Labour Monitoring?, in Regulating Labour in the Wake of Globalisation: New Challenges, New Institutions 37 (Brian Bercusson & Cynthia Estlund eds., 2007); Jill Esbenshade, Monitoring Sweatshops: Workers, Consumers, and the Global Apparel Industry (2004); Janice Fine & Jennifer Gordon, Strengthening Labor Standards Enforcement Through Partnerships with Workers’ Organizations, 38 Pol. & Soc’y 552 (2010); David Weil, Public Enforcement/Private Monitoring: Evaluating a New Approach to Regulating the Minimum Wage, 58 Indus. & Lab. Rel. Rev. 238 (2005).

  183. [183]. Frederick Mayer & Gary Gereffi, Regulation and Economic Globalization: Prospects and Limits of Private Governance, 12 Bus. & Pol. 1, 8–11 (2010).

  184. [184]. Id. at 8; Verbruggen, supra note 165, at 521. Recently, the assumption that the brand or retailer is the most powerful actor is being challenged by the growth of large and powerful intermediary actors in the global economy, including both logistics firms (e.g. FedEx) and manufacturers (e.g., Foxconn). With the exception of mega-retailers (e.g., Wal–Mart) or brands (e.g., Apple), such intermediaries may wield more market power than the “lead” firms with which they contract. In addition, a substantial amount of global production is not for recognizable brands. Cf. Mayer & Gereffi, supra note 183, at 14.

  185. [185]. See supra note 166 and accompanying text.

  186. [186]. Amengual, supra note 165, at 412–13.

  187. [187]. The theory of comparative advantage was introduced by David Ricardo. Ricardo, supra note 10; Gordon, supra note 108, at 1111–12.

  188. [188]. In my article, People Are Not Bananas: How Immigration Differs from Trade, I offer a full version of the argument I recap in brief in this paragraph and the following. See generally Gordon, supra note 108.

  189. [189]. See Michael J. Trebilcock, The Law and Economics of Immigration Policy, 5 Am. L. & Econ. Rev. 271, 275–76 (2003) (stating “neoclassical economic theory suggests that an optimal immigration policy would be not to have one at all”); Terrie L. Walmsley & L. Alan Winters, Relaxing the Restrictions on the Temporary Movement of Natural Persons: A Simulation Analysis, 20 J. Econ. Integration 688, 722–23 (2005).

  190. [190]. The European Union has always been the principal exception here, as it permits low-wage workers who are E.U. citizens to migrate from one E.U. country to another. There has been a new development in this area since I published People Are Not Bananas, with a trend toward increasing free movement between South American Countries. Diego Acosta Arcarazo, Is Free Movement in Europe an Anomaly? The New Open Borders Policy in South America, EU L. Analysis (Nov. 14, 2014), http://
    eulawanalysis.blogspot.com/2014/11/is-free-movement-in-europe-anomaly-new.html.

  191. [191]. See supra note 12 and accompanying text.

  192. [192]. Anti-discrimination law, as applied to U.S. citizens employed by U.S. firms abroad, is one exception. See supra note 183. However, no U.S. anti-discrimination law applies to non-citizens outside the United States, even if the discriminatory act is carried out by or at the behest of a U.S. employer. See Reyes–Gaona v. N.C. Growers Ass’n, 250 F.3d 861, 865 (4th Cir. 2001).

  193. [193]. Craig Becker, Labor Law Outside the Employment Relation, 74 Tex. L. Rev. 1527, 1537 (1996). There is a growing interest among agencies, courts, and advocates in using the joint-employer doctrine to break through this limitation. See generally Browning–Ferris Indus. of Cal., Inc., 362 N.L.R.B. 186 (2015).

  194. [194]. Samuel Laryea & Will Hughes, How Contractors Price Risk in Bids: Theory and Practice, 26 Construction Mgmt. & Econ. 911, 913–14 (2008).

  195. [195]. Richard Locke, The Promises and Limits of Private Power 12–14 (2013).

  196. [196]. See David Weil, Regulating Noncompliance to Labor Standards: New Tools for an Old Problem, 45 Challenge 47, 51–54 (2002).

  197. [197]. See generally Kendra Strauss & Judy Fudge, Temporary Work, Agencies and Unfree Labour: Insecurity in the New World of Work, in Temporary Work, Agencies and Unfree Labour: Insecurity in the New World of Work 1, 3 (Judy Fudge & Kendra Strauss eds., 2014).

  198. [198]. Jennifer Gordon & R. A. Lenhardt, Rethinking Work and Citizenship, 55 UCLA L. Rev. 1161, 1212–13 (2008).

  199. [199]. Barrientos, supra note 115, at 1065–66.

  200. [200]. See supra Part III.B.

  201. [201]. See supra Part III.B.

  202. [202]. See supra Part III.B.

  203. [203]. See supra note 199 and accompanying text.

  204. [204]. Philip Martin, Int’l Labour Org., Merchants of Labor: Agents of the Evolving Migration Infrastructure 14 (2005); see also Abella, supra note 80, at 203.

  205. [205]. See Abella, supra note 80, at 201–03.

  206. [206]. Jureidini, supra note 79, at 29, 44–50 (regarding bribes); Martin, supra note 204, at 4; Abella, supra note 80, at 207.

  207. [207]. For an illustration of the centrality of the “control” concept to even the most innovative proposals in the employment context, see Ruckelshaus et al., supra note 166, at 38–40.

  208. [208]. See Jureidini, supra note 79, at 34–35 (on the practice of double-dipping—that is, charging both workers and employers).

  209. [209]. See supra notes 68–69 and accompanying text.

  210. [210]. See supra Part II.

  211. [211]. See supra note 83 and accompanying text.

  212. [212]. Mijke Houwerzijl & Terry Wilkinson, The Effects of EU Law on the Social and Economic Goals of Europe 2020: A Decision Theoretic Approach to Wage Liability Regimes in Modern Europe, 14 German L.J. 1981, 1991 (2013).

  213. [213]. Id. at 1992.

  214. [214]. See supra notes 169–72 and accompanying text.

  215. [215]. Indeed, some U.S. employment laws are explicit about this. See Questions and Answers About Race and Color Discrimination in Employment, U.S. Equal Emp. Opportunity Comm’n, http://
    www.eeoc.gov/policy/docs/qanda_race_color.html (last visited Nov. 4, 2016) (“Title VII prohibits race and color discrimination in every aspect of employment, including recruitment, hiring, promotion, wages, benefits, work assignments, performance evaluations, training, transfer, leave, discipline, layoffs, discharge, and any other term, condition, or privilege of employment.” (emphasis added)).

  216. [216]. See id. When recruitment takes place through an intermediary within the United States, an employer may still be held responsible. Best Practices for Employers and Human Resources/EEO Professionals, U.S. Equal Emp. Opportunity Comm’n, http://www.eeoc.gov/eeoc/initiatives/e-race/bestpractices-employers.cfm (last visited Nov. 4, 2016) (“When using an outside agency for recruitment, make sure the agency does not search for candidates of a particular race or color. Both the employer that made the request and the employment agency that honored it would be liable.”). Yet when recruitment takes place through an intermediary outside the U.S., such as a labor recruiter in Mexico, even explicit discrimination is not actionable. For example, in
    Reyes–Gaona v. North Carolina Growers Association, 250 F.3d 861, 863, 866 (4th Cir. 2001), a case where it was uncontested that the recruiter had told the plaintiff in Mexico that the U.S. employer had a policy of not hiring anyone older than 40, the court held that even express age discrimination by a recruiter in Mexico on behalf of a U.S. employer was not actionable, because the refusal to hire did not take place within U.S. territory.

  217. [217]. U.S. anti-discrimination laws apply to employers of temporary migrant workers. However, there is no ban on discrimination in the H-2A program, and the position of the DOL is that it is beyond its jurisdiction to investigate employers for violations of U.S. laws prohibiting discrimination in hiring when the discrimination takes place at the recruitment stage abroad. See Etan Newman, Farmworker Justice, No Way to Treat a Guest: Why the H-2A Agricultural Visa Program Fails U.S. and Foreign Workers 26–27, http://www.farmworkerjustice.org/
    sites/default/files/documents/7.2.a.6%20fwj.pdf; see also Reyes–Gaona, 250 F.3d at 863 (where a recruiter reported that its discrimination on the basis of age was mandated by the U.S. employer’s explicit policy, but the plaintiff’s discrimination claim failed because the discrimination occurred outside the United States).

  218. [218]. See Reyes–Gaona, 250 F.3d at 863.

  219. [219]. For a critique of the territorial nature of the law governing employment, see generally Guy Mundlak, De-Territorializing Labor Law, 3 L. & Ethics Hum. Rts. 189 (2009).

  220. [220]. For a typical dismissal of an agency theory in the recruitment fee context see Arriaga v. Florida Pacific Farms, L.L.C., 305 F.3d 1228, 1244–46 (11th Cir. 2002).

  221. [221]. Restatement (Third) of Agency § 1.01 (Am. Law Inst. 2006).

  222. [222]. In Arriaga, the court rejects the agency argument to require the employer to compensate workers for fees charged by the recruiter on the grounds that the employer did not authorize or “apparently authorize[]” (i.e. give implicit consent to) its recruiter to charge fees. Arriaga, 305 F.3d at 1245 (quoting Restatement (Second) of Agency § 159 (Am. Law Inst. 1958)).

  223. [223]. Hyde, supra note 166, at 94; see also generally Brishen Rogers, Toward Third-Party Liability for Wage Theft, 31 Berkeley J. Emp. & Lab. L. 1 (2010).

  224. [224]. Hyde, supra note 166, at 94; Rogers, supra note 223, at 16–17.

  225. [225]. W. Page Keeton et al., Prosser and Keeton on the Law of Torts 536–37 (5th ed. 1984); see also generally Israel Gilead, On the Justifications of Strict Liability, in European Tort Law 28 (Helmut Koziol & Barbara C. Steininger eds., 2005). Gilead identifies two kinds of strict liability, one fault-based and the other, the “real” strict liability, which “is imposed not because there is something wrong with the activity but in order to change the results, the outcomes, of that activity. The justifications for this outcome-based strict liability should be found in the reasons that call for and justify changes in the activity’s outcomes.” Gilead, supra, at 31. Gilead further argues that:

    When the overall effects of [desirable activities] are examined, taking into account who gains from the activity and who loses, there are situations in which the outcomes may be considered unfair or as such that should be made fairer. In such situations, strict liability may be justified if and to the extent that it can increase fairness by shifting the costs of harm from the ‘passive’ activity that incurs it to the ‘active’ activity that inflicts it, thereby reallocating the benefits and burdens of the collision.

    Gilead, supra, at 39.

  226. [226]. Stephen R. Perry, Responsibility for Outcomes, Risk, and the Law of Torts, in Philosophy and the Law of Torts 72, 82 (Gerald J. Postema ed., 2001).

  227. [227]. Id. at 81.

  228. [228]. Id. at 84.

  229. [229]. See Rogers, supra note 223, at 42–47. An example of an enabling tort is where the owner of a car who leaves the key in the ignition is held liable when someone then steals the vehicle and uses it to commit a crime. Id. at 43.

  230. [230]. In a perfectly competitive (and therefore efficient) market, firms can look for workers willing to take lower wages, and workers can look for jobs that pay more. See George Borjas, Labor Economics 144 (6th ed. 2012).

  231. [231]. Id.

  232. [232]. This is classic fee-splitting. See George Gonos, Fee-Splitting Revisited: Concealing Surplus Value in the Temporary Employment Relationship, 29 Pol. & Soc’y 589, 592–94 (2001).

  233. [233]. See U.S. Gov’t Accountability Office, supra note 4, at 25–26.

  234. [234]. See R. H. Coase, The Problem of Social Cost, 3 J.L. & Econ. 1, 13 (1960) (discussing taking liability costs into account to decide courses of action).

  235. [235]. Deakin, supra note 166, at 81; see also Hyde, supra note 166, at 83–84. Hyde proposes a regime under which “purchasers of services may structure those relations in any way that is efficient, but may not, thereby, relieve themselves entirely of responsibility for the worst labour conditions.” Id. at 85.

  236. [236]. See Perry, supra note 226, at 119. In advancing what he terms an “avoidability-based theory of outcome-responsibility,” Perry defines control in a way that applies to the employer in the human supply chain context: “if an agent knew of the possibility that his action would or might cause a certain outcome, and if he had the capacity to act otherwise... then he could have avoided bringing the outcome about. In other words, control in this context means avoidability.” Id. at 84, 119. It is not clear, to be fair, whether Perry would see the human-supply-chain context as one where his theory would call for the imposition of strict liability, or whether he might argue that a negligence standard was more appropriate. Id. at 119–20. But see generally Kenneth S. Abraham, Strict Liability in Negligence, 61 DePaul L. Rev. 271 (2012) (arguing that liability imposed in negligence is not always distinguishable from strict liability).

  237. [237]. See supra notes 79–83 and accompanying text.

  238. [238]. Fudge, supra note 166, at 640.

  239. [239]. Id.

  240. [240]. Fair Labor Standards Act, 29 U.S.C. § 215(a)(1) (2012).

  241. [241]. Faced with a regime that renders them strictly liable for their outside recruiter’s abuses, firms have several options. They can bring recruitment in house. If they do not want to do so, one obvious option would be to monitor their recruiters more closely. This is currently impractical given the lack of origin-country enforcement, and the cost and complexity of a U.S. employer itself determining what its recruiter and the recruiters’ brokers are actually doing on the ground in Mexico, but might become more feasible if coupled with an effective origin-country government regime. Another option would be to build a safe harbor into the strict liability rules, so that an employer would be exempt from liability if it used a designated set of “good” recruiters, or a recruiter that was bonded or insured. Finally, a firm could stop using temporary migrant workers entirely, turning instead to undocumented or legal resident labor.

  242. [242]. For an examination of the most efficient cost-avoider in the product supply chain context, see Rogers, supra note 223, at 41–42.

  243. [243]. Workers Compensation is an example of a statutory strict liability regime in the workplace context that responds to these policy goals. Keeton et al., supra note 225, at 568–73.

  244. [244]. See supra note 97 and accompanying text.

  245. [245]. See supra notes 185–88 and accompanying text.

  246. [246]. See generally Ley Federal del Trabajo [LFT], art. 14, Diario Oficial de la Federación [DOF] 01-04-1970, últimas reformas DOF 21-05-2014 (Mex.). For a summary of key provisions in Mexico’s recently reformed recruitment statute and regulations, see Gordon, supra note 143, at 3–5.

  247. [247]. A 2013–2014 fellowship from the Open Society Foundations funded much of this research, and I am deeply grateful for their support.

  248. [248]. Gordon, supra note 124.

  249. [249]. See Int’l Labor Recruitment Working Grp., supra note 11, at 6, 34–35 (discussing the “Employer Accountability” principle). The International Labor Recruitment Working Group (“ILRWG”) has strongly supported the inclusion of joint and several liability for employers that use recruiters at the federal and state level. ILRWG advocated strongly for California’s Foreign Labor Recruitment Law, S.B. 477, 2013-4 Leg., Reg. Sess. (Cal. 2013), which passed in late 2014. As of July 1, 2016, the law mandates that foreign labour recruiters register with the California Labor Commissioner and disclose all terms of employment, and bars recruiters from charging workers for their services. Employers that use registered recruiters are protected from liability for the recruiter’s violations of the law. Centro de los Derechos del Migrante has used political pressure to move employers to take action to address recruitment abuses in the Maryland crab industry. Interview with Rachel Micah–Jones, Executive Director, CDM (July 30, 2014) (on file with author). The Global Workers Justice Alliance has argued for mandating transparency in the recruitment supply chain. See Glob. Workers Justice All., Why Transparency in the Recruiter Supply Chain is Important in the Effort to Reduce Exploitation of H-2 Workers: A Global Workers Justice Alliance Position Paper 2 (2011), http://www.global
    workers.org/sites/default/files/recruiter_supply_chain_disclosure_gwja_sept_2011.pdf.

  250. [250]. See supra notes 220–22 and accompanying text.

  251. [251]. S. Poverty Law Ctr., Culture Shock: The Exploitation of J-1 Cultural Exchange Workers 3–5 (2014), https://www.splcenter.org/sites/default/files/d6_legacy_files/downloads/
    publication/j-1_report_v2_web.pdf.

  252. [252]. NGA is currently urging U.S.-based multinational brands to join an anti-forced labor accord, which would require suppliers to prohibit retaliation, including by recruiters. Michelle Chen, What If Your Ability to Stay in This Country Depended on Your Employer?, Nation (June 12, 2014), https://www.thenation.com/article/what-if-your-ability-stay-country-depended-your-emp
    loyer; see also generally Nat’l Guestworker All., The Forced Labor Prevention Accord (draft on file with author); Telephone Interview with J.J. Rosenbaum & Jacob Horwitz, Nat’l Guestworker All. (Apr. 25, 2014) (on file with author).

  253. [253]. Gordon, supra note 124, at 29. Ethiopia and Indonesia have somewhat similar provisions, but rarely enforce them even against recruiters from those countries. Id.

  254. [254]. Id. at 32.

  255. [255]. The licensing scheme is so stringent that most applicants are rejected; only 22 were approved in the program’s first five years. Id. at 22–23.

  256. [256]. Id. at 23–24.

  257. [257]. Id. at 26.

  258. [258]. Gordon, supra note 124, at 26–27.

  259. [259]. Id.

  260. [260]. Id. at 27–28.

  261. [261]. Id.

  262. [262]. Id. at 28.

  263. [263]. Id.

  264. [264]. Id.

  265. [265]. See infra text accompanying notes 266–70.

  266. [266]. Gordon, supra note 124, at 33–36.

  267. [267]. Id. at 34.

  268. [268]. Id. at 36–39.

  269. [269]. Id. at 39–41.

  270. [270]. Id. at 40.

  271. [271]. Greg Asbed, Worker-Driven Social Responsibility (WSR): A New Idea for a New Century, Huffington Post (last updated Aug. 16, 2014), http://www.huffingtonpost.com/greg-asbed/workerdriven-social-respo_b_5500104.html.

  272. [272]. Id.

  273. [273]. See supra notes 185–86 and accompanying text.

*

Professor, Fordham University School of Law. 



The author would like to thank the following people for their comments on earlier versions of this Article: Aditi Bagchi; James Brudney; Janie Chuang; Harris Freeman; Judy Fudge; Tracy Higgins; Clare Huntington; Ethan Leib; Brishen Rogers; and Linda Sugin. I deeply appreciate the support of the Open Society Foundations Fellowship and Fordham University School of Law for my research and writing on this topic. Finally, I am very grateful to Alison Shea of the Fordham Law Library and Sarah Jaramillo, formerly of the Fordham Law Library and now at the N.Y.U. Law Library, and to my research assistants on labor recruitment, Alex Cárdenas, Janice Chua, Lauren Cooperman, Naseem Faqihi, Angelica Kang, Henry Parr, Natalie Puzio and Aaliya Zaveri.