States and the Political Economy of Unfree Labour

ABSTRACT

A growing body of academic and policy research seeks to understand and address the problem of contemporary unfree labour. In this article, we argue that this literature could be strengthened by a stronger conceptualization of, and more systematic attention towards, the role of national states. In particular, we argue that there is a need to move beyond simplistic conceptualisations of states as simple agents of regulation and criminal justice enforcement who respond to the problem of unfree labour, and to recognize the causal and multifaceted role that national states play in creating the conditions in which unfree labour can flourish. We propose a framework to understand and compare the ways in which national states shape the political economy of unfree labour. Focusing on the United States, we outline three arenas of governance in which national states have been particularly central to enabling the conditions for unfree labour: the regulation of labour mobility, labour market regulation, and business regulation. We conclude by reflecting on the comparative political economy research that will be required to understand the role of different states in shaping the conditions in which unfree labour thrives or is eliminated.

KEYWORDS:

• Unfree labour
• global political economy
• governance
• the state
• corporate social responsibility

Unfree labour, which encompasses exploitative practices often described as forced labour, human trafficking, and ‘modern slavery’, is a stable feature of the contemporary global economy. Unfree labour has been well documented as a core component of several industries ranging from mica mining to seafood processing, and is especially pervasive in low value-added and labour intensive activities like those found within agriculture, construction, and some forms of manufacturing (Verité 2017). Unfree labour tends to thrive in outsourced portions of labour and product supply chains (Phillips 2013, Crane et al. 2017), and has been documented to occur in both developed (Fudge and Strauss 2013, LeBaron 2015, Crane et al. 2017) and developing economies (McGrath 2013, Phillips 2013, Mezzadri 2016). Although unfree labour takes a variety of forms in the contemporary global economy, at its core unfree labour involves the use of coercion or compulsion to extract labour from workers. In the business context, unfree labour often involves deception at the point of entry into work, as well as coercion that precludes workers from exiting labour relationships that are highly exploitative. Common attributes of unfree labour include debt bondage, manipulation of contracts and credit, violence and threats of violence against workers or their families, and the predatory overcharging of workers for services such as accommodation or recruitment fees.

In recent years, both scholarly debates and policy discussions about the global problem of unfree labour have been gathering pace. While this new wave of scholarship on unfree labour has lent fresh and valuable insights into its extent, nature and dynamics, there remains a weak link – namely, the question of how national states fit into the picture. States are commonly incorporated in indirect fashion into the theoretical, analytical and empirical frameworks for understanding the political economy of unfree labour, lurking in the background but rarely being brought to the centre of attention. The result is that their central role in shaping the global conditions that facilitate the emergence and persistence of unfree labour remains obscured and under-emphasised.

In large parts of the scholarly debate, this lack of systematic attention to states stems from theoretical approaches to understanding the relationship between capitalism and unfree labour. Both Marxist and neoclassical debates about unfree labour in the capitalist economy have centred on the capital-wage labour relation, particularly around questions of value and agency (Fogel and Engerman 1974, Brass and van der Linden 1997, Rao 1999, Guérin 2013). The prioritisation of debates about whether and under what conditions unfree labour can be considered ‘capitalist’ or not – within an abstract and often economistic conception of capitalism – has meant that very little scholarship has focused explicitly on the role of states or other political actors. Rather, the state has frequently been collapsed into the capital-labour relation, by means of particular assumptions about the relationship of states to the capitalist economy. Especially in recent Marxist scholarship on unfree labour, this has led to a certain invisibility for the state and state actors, insofar as they are conceived as an extension of capital carrying little intrinsic or autonomous interest.¹

The sidelining of states is also characteristic of large parts of policy approaches to the problem of unfree labour – usually in these circles termed forced labour or human trafficking – and in the policy-based literature that has been growing rapidly since the mid-2000s (e.g. United Nations Office on Drugs and Crime 2012, International Organization for Migration 2017). One of its most evident manifestations is found in the dominant typologies of forced labour produced by international organisations and governments. The most influential typology is that of the International Labour Organization (ILO) (2005, 2014), which identifies three forms of forced labour: forced labour exploitation, forced sexual exploitation, and state-imposed forced labour. The former has globally the highest incidence, with the ILO estimating that 18.7 million (90 per cent) are exploited in the private economy, by individuals or enterprises (ILO 2014: 7). Yet the implication here – and in much other output from the ILO – is that the state is only present in one of these forms, namely state-imposed forced labour, as a key structure or agent. In short, forced labour exploitation is often taken to be perpetrated by private individuals or firms in the ‘private economy’, whose actors and dynamics are deemed separate from states.

This is not to suggest that states are entirely overlooked in the policy-based literature. Indeed, a great many policy actors with different approaches populate this arena, and states are often the subject of extensive attention, particularly in relation to issues such as immigration policy. Nevertheless, states are often considered predominantly as the vehicles for responses to forced labour, rather than as actors who play a causal role in shaping the conditions that give rise to it. Hundreds of reports, articles, and books from non-governmental organisations (NGOs), international organisations, academics, and governments illustrate this trend in important parts of the policy arena. To take just one, the ILO’s influential 2014 report, Profits and Poverty: The Economics of Forced Labour, goes as far as to portray states not only as the solution to unfree labour but also as a victim of it. It argues that ‘unscrupulous employers and criminals reap huge profits from the illegal exaction of forced labour’, harming victims, law-abiding businesses, and ‘governments and societies’, who are ‘harmed because the profits generated by forced labour bypass national tax collection systems’. The report notes that the available data on unfree labour ‘provides a compelling argument for stronger government intervention’ and that ‘measures are needed to strengthen laws and policies and reinforce inspection in sectors where the risk of forced labour is high’ (ILO 2014: 45, 47). Similarly, the United States Trafficking in Persons Report extensively discusses the efforts that states are taking to respond to various forms of unfree labour through law enforcement and criminal justice prosecution, but there is little acknowledgement of the role played by states in facilitating the use of these forms of exploitation by business (US Department of State 2017: 17).

Our aim is to help move beyond conceptualisations of states as policy responders, or simple agents of regulation and criminal justice enforcement, towards a robust and historically informed political economy analysis that can elucidate the ways in which shifts in state power and policy have played a significant role in facilitating unfree labour. We develop this argument by first setting out a theoretical and analytical perspective on how we can think about states in the political economy of unfree labour, calling for a greater comparative research effort to illuminate the role of states in shaping the dynamics of unfree labour across the world. In Table 1 in this section, we summarise the roles that national states play in the political economy of unfree labour, in order to contextualise our own subsequent empirical discussion and provide a schematic starting point for future comparative and theoretical research.

Table 1. State strategies and unfree labour.

	Conditions of vulnerability		Conditions of profitability
Direct	Directly imposed unfree labour (e.g. prison labour) Non-enforcement of labour legislation and laws prohibiting forced labour, child labour, trafficking Maintenance of migrant work programmes that remove labour rights and protections against forced labour Failure to regulate labour recruitment industry or enforce regulation against abusive labour recruitment practices Differential distribution of labour protections on basis of gender, race, caste, ethnicity, nationality, age, place of birth, etc. Failure to extend sufficient protection to identified victims of forced labour or trafficking, or provide viable employment alternatives		Facilitation of labour recruiters and intermediaries which enable exploitation Facilitation of industries and business practices that lead to systematic abuse Non-enforcement of trade law or ethical norms prohibiting import and export of goods made using unfree labour Non-prosecution of businesses which use forced labour or fail to act proactively to remove forced labour from their supply chains Maintenance of permissive laws or legal ‘loopholes’ facilitating labour exploitation e.g. child labour
Indirect	Restrictive policies governing mobility and migration Labour market policy placing downward pressure on wages and conditions Policies that restrict workers’ rights to organisation and unionisation Education policy governing differential access to education, levels of literacy and development of capabilities Policies which distribute social and labour rights unequally on the basis of gender, race, caste, ethnicity, class, age, nationality, place of birth, etc. Non-enforcement of laws prohibiting discrimination on the above criteria Policies that facilitate the concentration of private ownership of land, food, and water		Deregulation so as to facilitate wealth creation and concentration; policies to boost competitiveness by decreasing protections for workers Promotion of corporate self-regulation and private governance in the arena of labour standards, social and welfare policy Exemption for transnational investor firms from new or existing labour and social laws Lack of regulation of debt and credit markets that facilitate debt bondage Under-resourcing or de-funding of labour inspectorates ‘Outsourcing’ of labour inspection and enforcement functions to private actors (e.g. auditors) Failure to develop effective supply chain regulation encompassing extra-territoriality

In the following sections, we take the United States (US) as an illustrative empirical focus. We have selected the US because by most measures, it is the world’s largest economy (Kose et al. 2017) and unfree labour plays a significant role within it. Although there is no reliable country-level quantitative estimate for the scale of unfree labour within the US, since prevalence research is in its early stages (LeBaron 2018),² sector- and population-based research have established unfree labour as playing a key role within several industries, including agriculture, the service sector, and domestic work (Zhang 2012, Owen et al. 2014). The ILO estimates that developed economies such as the US have the highest annual profit per victim of forced labour, with the annual profit per victim estimated at US$34,800 (compared to US$3900 in Africa and US$5000 in Asia-Pacific) (ILO 2014). Our discussion focuses on outlining three arenas of governance in which the US state has been particularly central to enabling the conditions for unfree labour to flourish. The first relates to the regulation of labour mobility, the second to labour market regulation, and the third to business regulation.

Our discussion within each of these arenas seeks to draw together and develop three literature-based propositions: (1) that states affect unfree labour through their governance of migration, and, in the contemporary US, punitive mobility and border control regimes have strengthened the industry associated with trafficking for labour exploitation; (2) that states affect unfree labour through labour market policy, and, in the contemporary US, laws that curtail workers’ rights have enhanced vulnerability to exploitation including unfree labour; and (3) that states affect unfree labour by empowering employers to exploit workers, and, in the contemporary US, the failure to enforce labour standards and the devolution of responsibility for enforcement facilitates employers’ use of unfree labour. These propositions emerge from a wide range of scholarly literatures, and particularly those on migration, labour law, and private business regulation, respectively. There is clearly some overlap between these literatures, but they have nevertheless tended to be siloed across various disciplines and academic communities and have too rarely been brought into engagement with one another. We seek to move debates onwards by synthesising and developing these propositions, and by highlighting the multi-faceted ways through which states influence vulnerability to unfree labour and businesses’ demand for it in the contemporary global economy. We return in the conclusions to the bigger picture, and draw on our scheme in Table 1 to reflect on the future comparative and theoretical agenda for the study of states in the political economy of unfree labour.

Two clarifications are in order before we proceed. The first is that our focus here is on forms of unfree labour associated with forced labour and trafficking for labour exploitation in the US economy. We do not focus directly on those varieties which occur in the social reproductive sector, such as care work or sex work, although there are important points of convergence and we are hopeful that these will be fully taken up within the future comparative political economy research required to advance this line of inquiry. We also focus on the forms of unfree labour which identifiably occur within the territorial boundaries of the US state. A comprehensive exploration of unfree labour in the US economy would need to take into account those forms of exploitation which occur outside its national borders in the global production networks which feed the US economy and into which it is integrated. However, this is beyond the scope of this paper: our focus is, in this instance, on forms of unfree labour and exploitation among workers physically located in the United States in sectors and workplaces which are directly governed by the authority of the US state.

The second concerns our use of the concept of ‘unfree labour’. This concept is contentious and its definition has been widely debated within the literature on severe exploitation (see Miles 1989, Banaji 2003, O’Connell Davidson 2010, Strauss 2012, Morgan and Olsen 2015, LeBaron 2018). It is beyond the scope of our paper to resolve these debates. It suffices to note that unfree labour involves work that is involuntary and from which workers face severe constraints on their ability to exit, but our purpose is not to resolve controversies about precisely what level of ‘involuntariness’ is required, and what form preclusion to exit must take, for labour to qualify as ‘unfree’. Yet we consider the concept of ‘unfree labour’ to be more appropriate in the context of our paper than the international legal definition of ‘forced labour’, defined in the ILO’s 1930 Forced Labour Convention (ILO 2012, 2014) as ‘all work or service which is exacted under menace of any penalty for its non-performance and for which the worker concerned does not offer himself voluntarily’. This is so for two reasons. First, the ILO definition is too narrow. In subsequent interpretations and elaborations of the 1930 definition, the ILO has expressly excluded economic forms of coercion (e.g. starvation), as well as certain forms of state-imposed forced labour (e.g. some forms of prison labour) from its definition of forced labour. We deem these to be critical to the global political economy of unfree labour, and, especially, to understanding the role of states within it. Second, the concept of forced labour is often used in ways that suggest a neat binary between ‘free’ and ‘forced’ labour. We view this as problematic since many forms of ‘free’ labour also involve considerable levels of exploitation, and in low-waged sectors workers often move between the categories of unfree labour and other forms of exploitation throughout their lives. Like most scholars who prefer the concept of unfree labour, we recognise the role of economic coercion in instantiating unfreedom and resist the notion that there is a neat separation between ‘free’ and ‘unfree’ labour, given the porousness and blurring of boundaries between these categories at the level of workers’ daily lives.

States in the global political economy of unfree labour

Unfree labour is far from being a new phenomenon. Its existence was not caused by contemporary neoliberalism at the global or national level. Unfree labour exists across a wide array of political systems, institutional forms, and modes of economic organisation. Yet, we suggest that an understanding of contemporary unfree labour needs to be connected to an understanding of neoliberal globalisation, inasmuch as this dominant set of ideological, political and economic processes has shaped the forms that unfree labour has taken in the contemporary period and the mechanisms by which they are produced and reproduced. Our key point for present purposes is that states, whether those occupying dominant positions of power in the global political economy or those which are manifestly less powerful, are among the key architects of the neoliberalism, both nationally and globally, and by extension are central to shaping the context in which contemporary unfree labour can flourish.

This may seem like a straightforward point to a critical political economist, but it is one that goes against the grain of large swathes of the literature on contemporary unfree labour, and especially liberal conceptualisations that understand unfree labour’s roots to lie in the ancient past. Kevin Bales’ influential account, for instance, explains ‘the new slavery’ in the global economy as being rooted in a trans-historical relationship between ‘the poor’ and ‘the violent’, which emerged in ancient times and continues in the contemporary capitalist economy (Bales 2012). Although Bales – and the literature more broadly – make frequent reference to the role of ‘globalisation’ in perpetuating unfree labour, such references remain superficial and stand awkwardly alongside suggestions that global economic forces can also help to eradicate unfree labour (Datta and Bales 2013). At the very least, reflection is needed on the core tension between understanding globalisation to be among the root causes of unfree labour and the search for solutions which harness powerful economic and market forces to that end. In short, there is a need for much greater clarity regarding why and how neoliberal globalisation has contributed to the dynamics of contemporary unfree labour, and the role of national states within this picture.

The international political economy literature, particularly in its critical strands, offers rich perspectives on how the agency of states has driven neoliberal globalisation across the world. Far from being ‘an inevitable outcome of inherently expansionist economic tendencies’ (Panitch and Gindin 2012: vii), powerful states embarked in the 1970s on a massive political, economic, and ideological project of global sweep and reach designed to free up market forces. Neoliberalism became a dominant policy paradigm from the 1980s onwards, as states drove forward an ‘open-ended and contradictory process of politically assisted market rule’ (Peck 2010: xii), championing individual freedoms, property rights, and unfettered markets and trade, and a reversal of the Keynesian-era idea that an active, interventionist state is necessary to counter the social impacts of market deficiencies and failures. It has entailed dramatic changes in approaches to economic regulation, including the adoption of monetarist fiscal policy, the removal of barriers to competition, and the expansion and deepening of markets and competitive pressures in many states (see Helleiner 1996, Harvey 2007).

As they implemented these and other far-reaching economic reforms designed to re-ignite growth and free up market forces, states altered their relationship with private actors. They bolstered the power of capital and business, and in particular, facilitated the accelerating power of financial capital, retail corporations and transnational corporations (TNCs) in the global South (Harvey 2007, Lichtenstein 2010, Soederberg 2010, Hamilton et al. 2011). By devolving authority over markets and corporate conduct directly to companies themselves, states opened the door to the heightened involvement of private actors in economic governance processes, such that an increasingly ‘significant degree of global order is provided by individual firms that agree to cooperate, either formally or informally, in establishing an international framework for their economic activity’ (Cutler et al. 1999: 3–4). Private governance of this sort has not arisen independently of states; rather, the ongoing ‘delegation’ or ‘outsourcing’ of key governance functions to private actors, especially in relation to regulation, reflects the active agency of states in promoting private governance and shaping the forms it takes (Bartley 2007, Abbott and Snidal 2009, Green 2014, Mayer and Phillips 2017).

As states bolstered the power and mobility of capital, they also redrew the relationship between capital and labour. While the heightened vulnerability of workers is often portrayed as resulting from the unprecedented growth and mobility of TNCs, and from the increasing emphasis on corporate self-regulation, states have played a direct role in facilitating models of profitability that have depended on diminished power and circumstances for massive portions of the labouring population. Governments removed domestic and international barriers to competition in the labour market, and enacted policies intended to undermine the social and collective power of labour, such as mass layoffs, the replacement of public sector workers with private sector workers, strikebreaking and a political offensive against unions (McNally 2011).

Unfree labour needs to be understood in this context, linked to an understanding of the power relations inherent in neoliberal globalisation which are articulated both at the global level and within national borders, and reflecting the interests and agency of states and economic actors. Unfree labour today thrives across the global economy, in both the global North and global South, and across a wide range of manufacturing, agricultural, extractive and other industries in global production (Andrees and Belser 2009, Shelley 2010, US Department of Labor 2012, Verité 2012, Barrientos et al. 2013). Yet, in this context, the role of states in the political economy of unfree labour remains empirically under-appreciated and theoretically marginalised, and needs to be brought more satisfactorily to the centre of the debate. Sometimes, as noted earlier, unfree labour regimes are directly institutionalised and sanctioned by states, such as in the US prison industry (LeBaron 2015) or the Gulf states’ kafala system (Behbahani 2015). But most of the time, and particularly in the arenas of exploitation that interest us here, states do not themselves cause unfree labour. Rather, through the political projects pursued to facilitate globalisation and engagement with its various processes in the ways outlined above, they put in place the conditions in which individuals and groups of people become vulnerable to unfree labour, on the one hand, and, on the other hand, the conditions in which these labour practices become feasible and coherent ‘management practices’ (Crane 2013) and profitable business models for firms, employers and users of labour. We conceptualise these two dimensions as intrinsically linked – as flip sides of a coin – wherein these two sets of conditions are created simultaneously by particular strategies of state governance and reinforce one another.

In Table 1, we offer a schematic summary of the key ways in which national states shape the political economy of unfree labour, across these two dimensions of vulnerability and profitability. Our aim in the table is to offer a framework both to contextualise our empirical discussion and to provide a starting point for further research in this area. In some respects, states can be said to play a very ‘direct’ role in the facilitation of unfree labour; in others, their role is more ‘indirect’, wherein the conditions for unfree labour arise as a result of dynamics in wider policy and political arenas. Extensive research will be required to understand and compare how such dynamics have unfolded in different types of states and different political-economic contexts. The key point for our purposes is that contemporary unfree labour cannot be understood anywhere without an appreciation of the centrality of states, nor in isolation from the broader global political-economic processes which shape their nature, power, and agency.

Table 1 sets out an expansive terrain for research which goes far beyond the possible remit of a single paper. We seek therefore to cut into it here by focusing the case of the United States, and exploring how the US state has shaped the conditions in which unfree labour is enabled. We focus our attention on the three spheres of governance outlined earlier, where, in the US context, the state has been especially central to engineering the conditions in which unfree labour can thrive: (a) the regulation of labour mobility; (b) labour market regulation; and (c) business regulation.

The regulation of labour mobility

In the United States, unfree labour is concentrated among migrant workers. It is not, as is commonly assumed, a problem associated solely with undocumented migration; rather, circumstances of unfree labour have been identified among workers who have entered the country both with and without legal status (US Department of State 2016). Similarly, conditions of unfreedom may prevail from the outset, when someone is moved into situations of forced labour from the start of the trafficking process, or may occur among migrant workers within the United States. In all cases, migration and mobility are common denominators, and constitute an important dimension of how the practices of the US state shape the conditions in which unfree labour is enabled. Here, we open up the discussion by focusing on border control and elements of the ‘migration industry’ connected with trafficking, smuggling, and labour recruitment, on the one hand, and, on the other hand, immigration policy and its tensions with labour law and trafficking policy.

Border control, the ‘migration industry’ and labour recruitment

The United States operates one of the most highly policed border systems in the world. Starting in the 1980s, successive waves of reform aimed to tighten the border, with the focus always on restricting undocumented migration into the country. In the early 1990s, as public opinion on immigration became more negative, the rhetoric of border control became increasingly strident and enforcement campaigns expanded vigorously (Andreas 2000, 2001, Nevins 2002). The terrorist attacks of 11 September 2001 heralded an even more intense focus on both immigration and border policing, characterised by a significant militarisation of the border, particularly the land border with Mexico. All of these efforts failed to make any significant impression on the scale of undocumented migration; to the contrary, we have seen a remarkably rapid adaptation by the human trafficking and smuggling industry to the new realities of US border control policy.

To this extent, while it might be tempting to see the persistence and expansion of trafficking for forced labour as the result of the increasing porousness of borders in the context of globalisation (Shelley 2010: 42), such an explanation captures little of the situation in the US and other countries which maintain restrictive immigration policies and aggressive border control. Instead, the trafficking and smuggling industries have in fact been strengthened by tighter border enforcement, as their services have become, as it were, a necessary evil for prospective migrants (Andreas 2001, Spener 2009). Many of the small, local, freelance agents involved in the smuggling industry were squeezed out by the border enforcement campaign of the 1990s, replaced by a small number of large, better organised and more skilled organisations whose ‘market position’ was significantly enhanced (Andreas 2001: 118–19). Routes, techniques of movement, networks and actors have all adapted and continue to evolve, generating an often well organised and sophisticated form of transnational crime (Zhang 2007).

More aggressive border control techniques have increased the risks for migrants, and at the same time permitted massive increases in the prices charged by the increasingly sophisticated agents of the migration industry (Andreas 2001: 116). High profits increase the incentives to traffickers and smugglers to enter the industry and accept its risks, but also mean that people who make use of their services pay huge sums of money which render them bound by very large debts to these agents. The extent of the debt accrued in the context of a highly policed border enforcement regime augments the unfreedom that can be imposed on a migrant worker. This is most straightforwardly the case in trafficking situations, where a worker is trafficked into a situation of unfree labour on the basis of the costs incurred in the movement across borders. Yet, it is also the case for ostensibly ‘free’ migrant workers who owe money to smugglers, whose vulnerability to severe exploitation and forced labour is enhanced by the need to meet the onerous terms of debt repayments, alongside the mechanisms of coercion and exploitation that are facilitated by their undocumented status. Migrants can also be forced by smugglers to work at specific sites until the fees are repaid (Verité 2015).

Beyond the trafficking and smuggling industries, labour recruiters occupy an increasingly salient place in the migration industry across the world. While smugglers charge for the passage across the border, labour recruiters charge to secure visas and place a person at work in the United States. A vast network of labour recruiters operates both legally and illegally to supply an array of industries in the US with temporary foreign workers, especially in sectors such as agriculture, construction, and manufacturing. The use of recruiters and contractors has been shown in the United States and elsewhere to be a means by which employers have sought to evade legal responsibility for hiring unauthorised workers by involving third parties (Barrientos 2013, Phillips 2013), and such tactics of evasion have been spurred by policy initiatives to increase sanctions on employers, as in the US Immigration and Control Act of 1986 (Bump et al. 2011: 12, Hernández-Leon 2013: 29).

Between 2007 and 2016, it is estimated that intermediaries were involved in securing visas for around 80 per cent of the 2 million foreign workers approved for low-skilled jobs in agriculture and other sectors (Twohey et al. 2016). Some provide documented workers to employers through legal channels; others facilitate the recruitment of undocumented workers, favoured by employers for their easy ‘disposability’ and the absence of requirements in relation to wages and conditions. Exploitative practices are rampant, with all the hallmarks of trafficking and unfree labour: the illegal charging of exorbitantly high recruitment fees, making deductions from wages and manipulating the terms of repayment so as continually to increase the debt, placing workers in squalid living and transportation conditions, and engaging in deception concerning the conditions, location and type of work. Given that workers are usually paid (in cash) by the recruiter rather than the user of their labour, it can be difficult for the latter or for labour inspectors to determine the true level of wages and the extent of violations (Verité 2010, 2015, Andrees et al. 2015: 2–3).

NGOs and others have drawn attention to the lack of regulation of labour recruiters, but passage of appropriate legislation has been obstructed. The most significant attempt at comprehensive immigration reform was the bipartisan Border Security, Economic Opportunity, and Immigration Modernization Act (S744) that was passed by the Senate in mid-2013. This bill sought to combine another renewed focus on border security with the creation of paths to citizenship for undocumented immigrants, an updated visa system, and a new set of regulatory mechanisms pertaining to labour recruiters which were widely thought capable of significantly enhancing workers’ protections. However, the House of Representatives refused to consider this bill or any other proposal for comprehensive reform that went beyond the ‘enforcement narrative’ that came in the 2000s to dominate political debate (Avedaño and Fanning 2013). Thus a fragmentation of policy responsibility remains, in which recruitment agency licensing or registration regulations are generally determined at local state level, while immigration and labour regulations pertaining to migrants, including temporary migrant workers, are primarily issued at the federal level (Andrees et al. 2015: 46). An exception at the federal level is the legislation encased in Executive Order 13627 of 2012, Strengthening Protections against Trafficking of Persons in Federal Contracts, which makes some provision relating to recruitment practices through the mechanisms of supply chains.

Some regulatory provisions are also contained in key visa programmes, such as H-2A and H-2B, where directly or indirectly charging foreign workers job placement, recruitment, or other fees related to employment is prohibited, and disclosure of the terms of employment is required. In 2015, the Departments of Labor and Homeland Security issued new H-2B rules which required disclosure of foreign labour recruiters and their sub-contractors, and prohibited retaliation against workers (US Department of State 2016). However, the problem is enforcement, particularly where high fee charging and violations of anti-discrimination laws occur in countries of origin, such as Mexico, avoiding the reach and scrutiny of regulators in the United States (Andrees et al. 2015: 50). At the same time, other policy areas have been thought to work against these goals: the Trafficking in Persons report issued annually by the State Department indicates that NGOs have raised concerns that provisions in the Fiscal Year 2016 Appropriations Act increased some H-2B workers’ vulnerability to trafficking (unfree labour) through such measures as reducing wage guarantees, employers’ accountability for recruiting abuses, transparency and oversight (US Department of State 2016).

Immigration policy, criminalisation, and labour law

In US immigration policy in the 2000s, the focus of the enforcement narrative came to be undocumented migrants themselves, rather than the smugglers, traffickers or employers who exploit them or subject them to forced labour (Shelley 2010: 259). Deportations of undocumented migrants started to increase dramatically in the late 1980s, characterised by a sharp incline not in deportations of undocumented workers per se, but rather in deportations of undocumented migrants convicted of crimes, alongside summary deportation or ‘expedited removal’ at the border enabled by new legislation in 1996 (Ellerman 2009: 22–4). There remain demanding quotas for the detention and removal of undocumented migrants: in 2014, congressional mandates required US Immigration and Customs Enforcement to fill a daily average of 34,000 beds in detention centres (Brennan 2014). These developments have inevitably intersected with a wider ‘criminalisation’ of migration, particularly since 9/11 and the passage of the Patriot Act, and the devolution of authority to states to pass immigration legislation which has focused on the criminalisation of undocumented migrants (Sassen 2010). Particularly notorious instances of state-level legislation of this kind were enacted in Arizona in 2010 and Alabama in 2011, the latter case described as involving ‘the most draconian and oppressive set of provisions that this country … has seen since the era of segregation’ (Caballero 2011).

One of the principal mechanisms of coercion and control over migrant workers is the threat of denunciation to immigration authorities, and the already powerful disincentives to complain are vastly enhanced in this context of criminalisation (Weil and Pyles 2006, Verité 2015). Employers hiring undocumented workers have well-developed mechanisms for mobilising immigration law in their favour, which include failing to comply with obligations to verify immigration status until a worker tries to file a complaint or engage with a trade union, at which point employment can be terminated with no penalty to the employer, retaliatory reporting to the immigration authorities as a means of disposing of workers without incurring obligations under labour law, and loaning undocumented workers equipment to set themselves up as self-employed, before hiring them as sub-contractors (Bump et al. 2011: 10, 13). Fines and penalties for hiring undocumented workers are also sometimes incorporated into a firm’s business model, where the benefits and profits are deemed to outweigh the potential cost and risk, and firms frequently close and reopen under different names in order to avoid sanction. One widely reported case of this practice involved a garments factory in New York City manufacturing for GAP, Macy’s and other retailers, which withheld some US$5.5 million in wages from its workforce, comprising mainly Chinese immigrants working under conditions that approximated forced labour (Verité 2015).

In this context, labour inspectors frequently report concerns that they are unable to reach undocumented workers and act on exploitation in the workplace, including in instances of forced labour (Bump et al. 2011; author’s interviews³). Legal precedents also exist – such as the often-cited Hoffman Plastic Compounds vs. the National Labor Relations Board (NLRB) case of 2002 – in which undocumented workers were deemed not to be entitled to back pay when illegally dismissed for being involved in union activity. This ruling, found by both the International Labour Organization and the Inter-American Court of Human Rights to have violated international legal obligations, has not always been upheld as a precedent in subsequent US cases, and undocumented workers are entitled to claim rights and protection under other legislative umbrellas, such as the Fair Labor Act, Civil Rights Act or health and safety legislation. Nevertheless, there is evidence that Hoffman exacerbated perceptions among migrant communities that they lack protection in the workplace, and has been wielded by employers to dissuade workers from unionising activities (Human Rights Watch 2005).

There is consequently a considerable tension in the US context between labour law and immigration law, where the latter takes precedence over the former in many instances of exploitation, including forced labour. Extensive provision now exists for identified victims of trafficking, including the availability of T-visa status to enable them to remain and cooperate with prosecution efforts. However, very few such visas have been granted – only 3269 between 2002 and 2012, a number well below the limits established by the Trafficking Victims Protection Act (TVPA) first enacted in 2000 (Avendaño and Fanning 2013). This situation reflects the political tenor of the immigration policy context, where quotas for the identification and removal of undocumented migrants often create incentives not to look too closely for evidence of trafficking, and demanding criteria are deployed in considering whether a person qualifies as a victim of trafficking. Immigration policy thus continues to stand in tension with both labour law and trafficking law in the United States – the two arenas which are critical to addressing the problems of unfree labour and labour exploitation.

Labour market regulation

The prevailing assumption in several strands of scholarship is that unfree labour occurs the absence of labour market regulation. Yet, the re-writing of US labour and employment law and changing patterns of enforcement have actively engineered a generalised lack of protection for workers, particularly in low-paid, low-skilled, and low value-added segments of the labour market. States and municipalities across the country have passed laws restricting the minimum wage, and curtailing workers’ rights to refuse overtime and unsafe work, access sick leave, and recover unpaid wages. At both federal and state levels, legislation has been enacted to undermine unions and to restrict the assertion of collective bargaining rights. By de- and re-regulating the labour market in this way, the US state has been the institutional architect of a market that increases workers’ vulnerability to unfreedom, and makes unfree labour a viable business model for firms. We focus here on two key trends: the overall weakening of legal protections and conditions for workers; and the strategic deployment of unfree labour as a strategy to discipline ‘free’ workers.

Reshaping the US labour force

As has been well documented in the literature on labour market regulation (Peck 2001, Weil 2014), the US state has actively reshaped the labour force through policies that have undermined labour conditions and wages. This has been part of the broader political strategy to bolster corporate profitability that we sketched earlier, which involves deepening the vulnerability of the workforce, especially in sectors that are spatially fixed and cannot be outsourced overseas (such as agriculture, services, retail, construction, and care). The increased reliance on migrant labour (documented and undocumented) as the bedrock of this vulnerable workforce represents a core dimension of this political strategy.

At the federal and state levels, legislatures have enacted measures to alter dramatically the compensation and conditions of work. For an overwhelming proportion of US workers, particularly those in middle- and low-waged sectors, wages have steadily fallen since the 1980s as governments have supported employers’ commitments and initiatives to cut and suppress wages, and as minimum wage laws have been overturned across several states. These measures have been highly effective. From 1979 to 2014, US Bureau of Labor Statistics data indicate that, although real earnings have increased for the highest earners, earnings have remained ‘basically unchanged’ for the lowest-paid 10 per cent of earners (2015a: 2). Trends in the 2000s were even more pronounced. According to a recent study of low wages by the University of California Berkeley Labor Center, ‘inflation-adjusted wage growth from 2003 to 2013 was either flat or negative for the entire bottom 70 percent of the wage distribution’ (Jacobs et al. 2015; emphasis in the original).

Downward pressure on wages has not simply resulted from abstract market forces, but rather has been part of purposeful state strategy to spur corporate profitability and growth, and withdraw from its traditional role as regulator of the bottom end of the labour market. This strategy has been articulated in a number of ways. Federal and state-level governments have implemented deliberate and proactive legislative measures to decrease wages, such as by undermining and repealing prevailing and living wage policies, limiting the numbers of workers covered by such policies, and preventing cities and states from passing prevailing and living wage policies in the first place. Between 2011 and 2013 alone, over 105 bills were introduced that aimed to weaken or repeal wage standards (National Employment Law Project 2013). By abandoning commitments to welfare-based protections and institutionalising policies known as ‘workfare’ (Peck 2001), the state has aggressively withdrawn the public protections and entitlements for vulnerable workers which had previously curtailed downward pressures on wages. The US government has removed restrictions previously placed on employers’ strategies to lower labour costs, such as through increased use of offshoring, temporary workers and, as noted above, employment agencies (Brown et al. 2014, Bonacich and Appelbaum 2000). At the federal and state levels, there has also been a widespread refusal to modernise labour laws, such as those guaranteeing workers overtime and leave, with eligibility for overtime declining from 65 per cent of salaried workers in 1975 to 11 per cent in 2014 (Conti 2014). According to one 2015 study, the peak inflation-adjusted value of the minimum wage has declined by 24 per cent since 1968, despite a doubling of productivity and increases in education and skill levels among low-wage workers (Economic Policy Institute and National Employment Law Project 2015: 1). In contrast to its more active role in wage policy over much of the twentieth century, the U.S. Chamber of Commerce captured the position of the US state when it argued in the early 2000s that ‘we don’t think the government ought to be in the business of setting wages’ (Washington Times 2002).

Declining wages are compounded by endemic wage theft, wherein employers violate wage laws, overtime, and other employment law with impunity (Bernhardt et al. 2009a). A 2009 study of 4387 workers in Chicago, Los Angeles, and New York City found that the average low-wage worker lost $2634 a year in unpaid wages, which totalled 15 per cent of their income (Bernhardt et al. 2009a, 2009b). The prevalence of piece rate wages in many sectors, such as garments, has also enabled minimum wage laws to be circumvented. Verité reports research from 2013 which indicated that 60 per cent of Los Angeles garment workers were paid less than the minimum wage, and 90 per cent did not receive overtime pay, even when working in excess of 40 hours per week (Verité 2015: 5). US Department of Labor inspections in Southern California’s garment industry between 2007 and 2012 detected violations in 93 per cent of cases, and $11 million in back wages owed to some 11,000 workers, and a report in 2010 estimated that $26.2 million per week was withheld or stolen from garment workers in Los Angeles alone (see also Bonacich and Appelbaum 2000, Verité 2015: 5). Significantly, continuing our previous theme, immigrant workers are nearly twice as likely to experience employment violations as US-born workers (Bernhardt et al. 2009b).

Declines in workers’ compensation have been paralleled by decreased job quality, including increases in part-time and insecure and ‘flexible’ employment. Several recent studies have documented the degradation of work and steep declines in labour standards that have occurred across many sectors of the US economy, which have often been spurred by strategies on the part of employers to reduce and evade obligations to labour and employment law through the use of temporary and contract labour, most often, as noted, in the form of migrant workers (Vosko 2005, 2011, Appelbaum et al. 2003, Bernhardt et al. 2009b, Kalleberg 2011, Fudge and Strauss 2013). As the state has re-regulated the labour market to foster ‘flexible’ employment, firms have increasingly used temporary employment agencies – one of the fastest growing sectors of the US economy since the 1980s (Peck, Theodore and Ward 2005, Theodore and Peck 2013). The US Government Accountability Office estimates that 40.4 per cent of the US workforce is now comprised of ‘contingent’ workers, as compared to 30.6 per cent in 2005 (Pofeldt 2015). By comparison, throughout the 1970s, the national number of temporary workers was as low as 250,000 (Theodore and Peck 2013: 27). For workers, the turn toward contingent employment has entailed an increase in job opportunities, but an erosion of job security, lower levels of benefits, and a decreased ability to exert rights and contest coercive and unlawful management strategies given the threat and possibility for immediate termination (ILO 2011, Barrientos 2013).

Finally, the state has weakened the power of unions and of workers to engage in collective action. Although an anti-union agenda has been firmly in place since the early 1980s, this intensified over the 2000s under pressure from large corporate lobbies, including Americans for Prosperity, National Association of Manufacturers, and Club for Growth (Lafer 2013: 8–9). At least 25 states have enacted ‘Right to Work’ statutes (either by law or constitutional amendments), which prohibit union security agreements and make it illegal for unions to require employees to pay union dues. These trends intensified in the aftermath of the 2008–2009 financial crisis. Several states passed bills to ‘repair budgets’ that undermined collective bargaining rights. Between 2011 and 2012, ‘four states passed laws restricting the minimum wage, four lifted restrictions on child labor, and 16 imposed new limits on benefits for the unemployed’ (Lafer 2013: 3). There is evidence that this has been a largely political strategy, inasmuch as there was little correspondence between where these laws were enacted and where the economic problems were most severe (Lafer 2013: 6). In the face of these pressures, trade union density has fallen across the US, declining from 27.4 per cent in 1970 to 10.8 per cent in 2013 (OECD 2015). Such shifts have undermined the role of unions as workplace regulators, and especially, as advocates for workers’ rights. By extension, they have created a situation of porousness at the bottom end of the labour market, where vulnerable workers slide from exploitation to unfreedom and forced labour. It is well understood that unfree labour is concentrated in sectors and workplaces with no or very low levels of union activity, where workers are socially isolated and lack political voice, and can be coercively deprived of the ability collectively to challenge the conditions in which they work (Phillips 2013).

Strategic deployment of unfree labour

An understudied component of labour market restructuring in the US has been the deployment of unfree labour in ways that have re-shaped the labour force. The primary form this has taken to date has been the strategic substitution of prison workers for workers circulating within the market (LeBaron 2015).⁴ Since 1979, when the government passed the Prison Industry Enhancement Act re-authorising the use of profitable prison labour by private companies, both the state and federal prison industries have expanded.⁵ Prisoners perform a vast range of tasks across several sectors within and outside prison walls, from building high-end custom motorcycles (California Prison Industry Authority 2015) to milking goats and cows to produce artisanal cheeses sold at US supermarkets (Alsever 2014). Prisoners working for the state and federal government are paid between $0.00 and $2.00 per hour, while those who work in the state prison industries for private companies are paid the minimum wage ($7.25). Yet 80 per cent of their pay can be appropriated by the state towards prisoners’ room and board at the prison, victim compensation, and family support.

Two trends in this context are most relevant to our discussion. First, states have begun to substitute prisoners for skilled, well-paid public sector workers in such activities as firefighting and data entry services for state governments. For instance, in California, which is home to one of the largest prison industries in the country, over 4400 inmates are currently deployed as firefighters. Paid between $1.45 and $3.90 per day compared to a non-inmate firefighter’s typical hourly mean wage of $34.44, the California Department of Corrections and Rehabilitation estimates that the use of inmates firefighters saves the state $1 billion per year (Lewis 2014, California Department of Corrections and Rehabilitation 2015, US Bureau of Labor Statistics 2015b). The use of prison labour in high-skilled, high-waged industries is a cost-cutting strategy amidst steadily rising costs of incarceration and persistent fiscal crisis in states like California. But it is also a strategy of labour discipline. Firefighters are heavily unionised, and their unions are among the country’s strongest public employee organisations. State deployment of prisoners in this industry, therefore, not only demonstrates to high-skilled, high-paid public employees that they are dispensable and undercuts their bargaining power, but also exerts downward pressure on wages and working conditions since prisoners have little to no ability to contest their compensation or conditions.

Second, prisoners have been substituted for low-skilled, low-waged, migrant workers (who often have irregular immigration status), particularly in the agricultural sector. In the face of a wave of anti-immigration laws and crack-downs on undocumented migrant workers in states such as Idaho, Georgia, and Alabama (Powell 2012), prisoners have been contracted out to farms to work harvesting, processing, and sorting agricultural products (Millman 2011, Colorado Correctional Industries 2015). In some states, the use of convict labour in agriculture has become sizable. In Arizona, for instance, in 2011 (the last year for which an estimate is available), inmates worked 1.3 million hours for commercial growers (Millman 2011). As in firefighting, the use of prison labour in the low-waged, agricultural sector is a strategy of both cost saving and labour discipline. In most states, prisoners are paid minimum wage, with the government appropriating 80 per cent of prisoner wages towards the costs of incarceration. At the same time, governments are strategically deploying prison labour to further erode labour standards and curtail the bargaining power of low-waged, often migrant agricultural workers.

Business regulation

Unfree labour is generally understood to be perpetrated by private individuals and firms. Yet, the state’s redefinition of its relationship with corporate actors, and changing patterns of regulatory enforcement over the last several decades, have fostered the context in which individual employers and firms can impose unfree labour with relative impunity. At both national and state levels, institutions designed to enforce labour and employment law have had their budgets and staff reduced, and governments have increasingly devolved authority and responsibility for working conditions onto employers. State strategies have also been oriented to promoting private governance and (voluntary) corporate self-regulation, often under the umbrella of corporate social responsibility (CSR). By re-drawing its relationship to firms and significantly devolving regulation for labour standards to CSR-type initiatives, the US state has created the conditions of possibility for business models configured around systematic practices of labour exploitation and unfreedom. We focus here on two key trends: the decline in state-based enforcement of labour standards; and the rise of private labour governance initiatives.

Declining state-based enforcement of labour standards

At the federal level, the US Department of Labor is tasked with enforcing over 180 labour laws, including the Fair Labor Standards Act (FLSA), which sets standards for wages and overtime, and the Occupational Safety and Health (OSH) Act, which sets standards for health and safety. Within the Department of Labor, the two most significant enforcement divisions are the Wage and Hour Division, which administers the FLSA, the Migrant and Seasonal Agricultural Worker Protection Act, and several other laws, and the Occupational Safety and Health Administration, which administers the OSH Act. Each US state also has a Department of Labor tasked with enforcing state-level labour laws.

There has been a steep decline in federal labour standards enforcement since the late 1970s. The budgets of the Wage and Hour Division, the Occupational Safety and Health Administration, and several other enforcement agencies have stagnated or declined since the 1970s (Weil 2010). Persistent staff cuts and budget cuts have left public labour standards enforcement agencies with fewer investigators, and able to conduct fewer investigations: between 1998 and 2009, the number of investigators was reduced by 22 per cent, from 942 to 731, and the number of investigations handled per investigator fell by 40 per cent, from 53 to 32. At the same time, the number of workplaces and workers in the US has grown, expanding the scope of enforcement for each agency. The net result is dramatic: David Weil estimates that there has been a net 53 per cent decline in investigations per establishment between the late 1970s and 2007s (Weil 2010: 6). To present the numbers differently, from the time that the federal minimum wage law was established in 1941 to the end of the 2010s, and in spite of the fact that the number of laws requiring enforcement has grown dramatically, the number of inspectors per worker had been cut tenfold: from one inspector for every 11,000 workers in 1941 to one inspector for every 141,000 workers in 2008 (Lafer 2013: 29).

State-level enforcement agencies have experienced similarly significant staff and budget cuts, curtailing their scope and remit (Peck 2001, Weil 2010). In many states, this has occurred as a dimension of broader budget cuts, while others, such as Ohio, have publicly singled out enforcement of labour law as especially necessary for de-funding. By rebalancing the budgets in these ways, states have sent a very clear message to firms: that employers can break labour law with relative impunity, as states will not enforce these laws, nor safeguard workers covered by them. Employers now face a trivial chance of being inspected in any given year, such that ‘an employer would have to operate for 1000 years to have even a 1 per cent chance of being audited by Department of Labor inspectors’ (Lafer 2013: 29).

Changing patterns of enforcement mean that when inspections do occur, inspectors are not necessarily targeting the employers or portions of supply chains in which the worst forms of abuse and exploitation are known to thrive. As the reduction in the size and role of enforcement agencies has created very challenging circumstances for inspectorates, agencies have shifted away from random inspection and monitoring and towards responding to individual complaints (Weil 2010, 2014), and there is little evidence to suggest that complaints target the worst abuses (Weil and Pyles 2006, Weil 2014). To the contrary, as noted earlier, the worst labour abuses, including those associated with unfree labour, tend to go unreported given concerns about personal safety and immigration consequences, lack of knowledge about employment law, or the fear of employer retaliation (Anner 2012, Hilgert 2013, Verité 2014).

The rise of private labour governance

The active withdrawal of the state from regulatory and enforcement functions has enabled – and purposefully promoted – the rise of a private labour governance regime, which has included a range of voluntary, industry-led tools and initiatives ranging from codes of conduct to certification schemes and self-monitoring through private audit firms (Ebenshade 2004, 2012, Taylor 2011, Fransen 2012, LeBaron and Lister 2015). The US state’s reduction of enforcement of labour standards has been buttressed by the claim that companies are largely capable of setting and enforcing their own labour standards, and where they fall short, it is the job of consumers to hold them accountable. As part of the broader trends documented in the previous section, governments have increasingly devolved authority and responsibility for working conditions onto employers, presenting labour and employment law and standards as a relation between workers and employers, and indeed between employers and consumers, rather than one that is mediated by and fundamentally shaped by the state (Esbenshade 2012).

Recent legislation to address forced labour and human trafficking in supply chains has codified this turn towards corporate self-regulation, effectively writing the state out of labour standards regulation and enforcement. California’s 2012 Transparency in Supply Chains Act paved the way, requiring manufacturers and retailers with annual global profits exceeding US$1 million and conducting business in California to disclose the voluntary measures (if any) they are taking to verify their supply chains against trafficking and forced labour. Similar federal legislation, the Business Supply Chain Transparency on Trafficking and Slavery Act of 2015, has recently been introduced and is under consideration by Congress. This wave of transparency legislation is significant because it contains no binding requirements on firms, nor sanctions for non-compliance (Phillips 2015, Phillips et al. 2016, LeBaron and Rühmkorf 2017). Rather, the role taken on by the state is one of encouraging businesses to take ethical considerations seriously and facilitating the communication of information between businesses and consumers.

As part of this private labour governance regime, companies have increasingly turned to the use of private audit firms to monitor standards within their supply chains. Although auditors are often presented as an alternative to state-based enforcement, there are crucial differences between public and private enforcement agencies. Private auditors do not have powers of investigation or prosecution. The FLSA gives the US Department of Labor both powers of civil litigation (e.g. to recover back wages) and criminal prosecution (e.g. to prosecute wilful violations). The Department of Labor’s enforcement tools includes ‘monetary penalties, temporary restraining orders to prevent the shipment of “hot goods”, injunctions to compel future compliance, and a prohibition against intimidating employees who complain’ (Weil 2009: 13). Auditors, by contrast, cannot even require companies to open locked drawers (also Esbenshade 2004, 2012, Allain et al. 2013, LeBaron and Lister 2015). Most private auditing mechanisms rarely, if ever, cover third-party labour intermediaries, whose key innovation is a blurring of the lines of accountability and liability between workers and employers, thus leaving the employers of sizable swathes of the US-working population removed from scrutiny. In short, and unsurprisingly, companies have designed private initiatives which leave the core components of their business models fully intact.

As has been well documented in the literature on the privatisation of governance (Cutler et al. 1999, Soederberg 2010, Büthe and Mattli 2011, Mayer and Phillips 2017), the turn towards corporate self-regulation has bolstered the power and authority of firms vis-à-vis the public, and especially, working populations. Firms have used this power and authority to maximise profit and design business models and organisations to facilitate future profits and growth. By withdrawing from regulatory functions, and facilitating the rise of private labour governance, the state has removed the protections it previously offered to workers, such as the guarantee that business models needed to comply with labour laws. This has led to a situation in which firms can exploit workers with relative impunity, and, where it becomes a feasible and profitable management practice (Crane 2013), configure their business models around unfree labour. For workers at the bottom end of the labour market, the conjunction of these state policies has heightened their vulnerability to these extreme forms of labour exploitation.

Conclusion

Our aim in this paper has been to call attention to a weak link in debates about unfree labour in the contemporary period, namely, the issue of the state. We have contended that states play a critical, multi-faceted role in creating the conditions in which unfree labour can flourish, and have brought together a range of perspectives and literatures to set out a framework for locating the state within a political economy of unfree labour, as a basis on which to call for a greater comparative research effort. We took the case of the United States as an empirical illustration, focusing on what we suggest are three of the key critical arenas in the US case: (a) the regulation of labour mobility; (b) labour market regulation; and (c) business regulation.

Each empirical case is, of course, specific, whether taken in isolation or as a representative of a particular model of political economy. The detail and focus of comparative research will inevitably need to capture complex forms of contingency. Table 1 sought to provide a schematic starting point for this comparative effort, drawing together the key direct and indirect roles that states play in fostering the conditions of vulnerability and conditions of profitability which combine to facilitate unfree labour. Not all of the elements outlined in that scheme apply in any single case, and the challenge for research is to combine big-picture theoretical work oriented to capturing the place of states in the political economy of unfree labour with the expansive comparative empirical work that is needed to underpin it. In each country, region, sector or model of capitalism, where the emphasis is placed will need to vary, but the point of our summary in Table 1 is that many of its elements will stretch across different contexts and thus provide the building blocks for a robust comparative and theoretical effort.

Let us briefly draw out some examples. To capture the US case, our focus on the regulation of mobility was oriented towards immigration policy. In some cases, such as the UK, continental European countries, Australia and others, this focus might fit well in view of the commonalities in the patterns of unfree labour that prevail among immigrant workers. In other countries, such as China, Brazil or India, the greater focus would need to fall on the state’s governance of mobility in an internal context, focusing on how particular dynamics of internal migration render workers vulnerable to extreme forms of labour exploitation. In other cases, such as in some South-east Asian and Middle Eastern states, the key issue might be the state’s role in promoting the ‘export’ of labour through organised migration schemes, and the vulnerabilities which are thereby created for migrant workers outside the state’s borders. Almost across the board, the focus on labour recruitment will be significant given the expansion of the labour contracting industry across the world.

Our focus on labour market regulation in the US context highlighted state governance strategies to achieve a core goal of ‘flexibility’ for business, in ways which have opened up possibilities for forced labour to become a feasible practice for some businesses. Such a focus travels well to other national contexts defined by neoliberal states and political economies – the liberal variety of capitalism, as it were – in which similar trends and processes have been amply in evidence. But it has wider resonance, capturing the restructuring of state-capital-labour relations across the vast majority of countries as they navigate the power relations encased within neoliberal globalisation. Core differences will be revealing: the focus in China would need to rest on the relationship between rising wages and the development strategy of ‘upgrading’ in higher value-added production, and labour shortages and the continued repression of workers’ rights and conditions. Our attention to prison labour in the US context is more specific, and would draw comparisons with China and some other countries, but would not be an issue in other contexts. However, it sheds light on the broader role of the state in underwriting or directly deploying unfree labour, as in the kafala system in the Middle East, and how this shapes the wider labour market and the place of other groups of workers within it.

Our discussion of the relationship between states and firms in creating the conditions for unfree labour is fertile ground for further comparative work, supplementing an already fruitful body of literature on public/private governance. In the US, we documented a decline of state-based labour standards enforcement in favour of private governance, which finds parallels as a global trend. But it is not uniform and is continually evolving: in Brazil, for instance, a much more active – and activist – role for the state in governing labour standards developed over the 1990s and 2000s, but the political climate has more recently swung away from this particular mode of governance. In other parts of the world, the weak bargaining power of states vis-à-vis TNCs and other private actors has held state-based enforcement at bay, or brought out its systematic retraction in order to create a climate deemed favourable to investment, particularly where the investment is premised on low-cost labour and low regulatory standards.

In short, the comparative ground is fertile, and such an effort to understand the place of the state in underwriting unfree labour in its diverse forms, in diverse political economies, and in diverse geographical and social settings would pay dividends for our understanding of how these forms of labour exploitation flourish. Our final point is to underline the challenge that we have sought to pose here to existing debates, in which it is frequently presumed that unfree labour flourishes in the absence of the state, where regulation and enforcement are low or non-existent. An absence of regulation or the presence of a state with low capacity or political will for enforcement may well be connected with the incidence of unfree labour, and those conditions are worthy of investigation in themselves. Yet, powerful and active states with high levels of institutional capacity, such as the US state, are also critical to laying down the conditions in which unfree labour flourishes. They represent a core dimension of the root causes of the problem, not simply a potential source of regulatory solutions to it. States, in short, cannot be written out of the political economy of unfree labour.

Acknowledgements

We thank the three anonymous reviewers for NPE for their time, effort and very helpful comments on an earlier draft of this article. Our thanks are also due to the Liu Institute for Global Issues at the University of British Columbia for generously hosting Nicola Phillips as a Liu Visiting Fellow, during which our collaboration leading to this article began.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Genevieve LeBaron http://orcid.org/0000-0003-0083-6126

Nicola Phillips http://orcid.org/0000-0001-9588-976X

Notes on contributors

Genevieve LeBaron is Senior Lecturer in the Department of Politics at the University of Sheffield, Co-Chair of the Yale University Modern Slavery Working Group, and a UK ESRC Future Research Leaders Fellow (2016-2019).

Nicola Phillips is Professor of Political Economy in the Department of Political Economy and Vice-President and Vice-Principal (Education) at King's College London.

Additional information

Funding

This work was supported by Economic and Social Research Council: [Grant Number ES/N001192/1].

Notes

1 It is important to acknowledge, however, that the empirical basis of much recent Marxist scholarship on unfree labour has been peasant and ‘deproletarianised’ communities involving rural workers in the developing world. The invisibility of the state within these debates may be explicable in that states and state actors (via labour market policy, policing, and even the law) were not always directly involved in people’s daily lives and labour market circulation.

2 The only available national estimate for prevalence of ‘modern slavery’ in the US comes from the Global Slavery Index produced by an NGO; however, the Index’s accuracy has been repeatedly questioned by commentators and academics who have raised compelling criticisms about its methodology and credibility of data. Several organizations – including the US Department of Justice – are currently working to develop stronger prevalence estimates for forced labour within the United States.

3 Author’s interviews, Department of Labor Wage and Hour Division and various NGOs, Washington, DC, March 2012.

4 Prison labour is defined as a form of ‘involuntary servitude’ in the 13th Amendment to the US Constitution and its use by private businesses is illegal in most countries, including the 178 countries that have ratified the 1930 ILO Forced Labour Convention. The US has not ratified this convention.

5 See, for instance, Colorado Correctional Industries (https://www.coloradoci.com/) and California Prison Industry Authority (http://pia.ca.gov/) websites.