Human displacement, often accompanied by religious or ethnic persecution, is an old, arguably even ancient phenomenon. Indeed, the practice of expulsion created some notable communities of the early modern period: Jews from Spain who reinvigorated Ottoman Salonika after the Inquisition, Huguenot settlers from France who remade early modern London. States and empires generally regarded such mass dislocation with equanimity, as a common if not routine geopolitical phenomenon requiring no special response, until the late nineteenth century. It was not until the aftermath of the First World War that the global powers truly came to think of mass displacement as dangerous, a potentially serious challenge to the ethnically bounded nation-state on the one hand and a fractious European and American labor market on the other.
How could the Allied powers simultaneously prevent the displaced from causing mass political disruption (and potentially fueling a turn to Bolshevism) by flooding the labor market, and at the same time protect a fragile, nascent global system of ethnic nation-states under imperial sponsorship? The League of Nations, constructed to protect an imperial world order for the victors of the First World War, was charged as a first order of business with concocting some way of dispersing the war’s many refugees. It decided that placing refugees in low-wage employment in colonial settings could serve as a solution—an idea that would long outlast the League itself. This plan had a variety of incarnations through the interwar and post-1945 periods, including League proposals to release Russian, Armenian, and Assyrian refugees into the care of British-, French-, and American-connected corporate employers in Latin America and to put refugees to work developing the British- and French-occupied Middle East (inspired in part by the Zionist project of colonizing Palestine with European Jewish laborers). After the League collapsed in the late 1930s, its imperial practices vis-à-vis refugees served as inspiration and example for an evolving “international community” increasingly under American sponsorship: in the aftermath of another global conflict causing epic dislocation, the postwar International Refugee Organization promoted just such low-wage industrial employment in far-flung locations from Canada to India as a primary mode of dealing with refugees who refused to go home.
It was an approach that sat comfortably alongside the first attempts to codify the definition of refugee in international law and offer legal protection to those so brutally displaced in Europe’s great midcentury conflagration, a task accomplished to great fanfare with the signing of the Convention Relating to the Status of Refugees in 1951. Despite its narrow definition of refugees as only those displaced before 1951 and mostly in Europe, the Convention has been hailed ever since as a model of humanitarian vision. In fact, its intentions and consequences were even less universal, and less altruistic, than they appeared, excepting non-white displaced people in decolonial settings like partition-era South Asia as mere domestic refugees and reaffirming a separate, already-extant legal category for displaced Palestinians: both classifications that carried none of the legal, political, and social protections mooted for the subsequent definition of the presumably normative (at this time, mostly European) version. The countries of the Global North would henceforth be able to use their support of the Refugee Convention to illustrate their commitment to a universalist form of international humanitarianism, while denying its protections to nonwhite non-Europeans on the simple premise that they occupied different legal tracks of refugee. As the Cold War set in, this theoretical principle that refugees could be segregated into distinct legal categories received a new practical articulation on the ground; Western governments developed the notion of reserving political asylum mainly for white Europeans who could be plausibly depicted as victims of Communism, while declaring refugees from the decolonizing world eligible not for permanent asylum elsewhere but only for material help in situ. Increasingly, such help was offered via the mechanism of physical internment.
Aiming to contain refugees where they were, the United Nations eventually passed an addendum to the Convention that expanded its definition of refugee to include populations displaced outside Europe and after 1951—thus extending the UN refugee agency’s capacities for physical restraint of refugees everywhere while ostentatiously advertising the Western bloc’s commitment to humanitarian assistance and human rights. Around the same time, the refugee agency—the UN High Commissioner on Refugees, or UNHCR—began to advertise what it viewed as the three main “durable solutions” for refugees: resettlement in a third country; local integration where they had landed; or repatriation. As the Cold War dragged on, and especially as labor markets in the West became accustomed to both highly restrictive immigration regimes and global access to an apparently unlimited supply of extremely cheap and totally unprotected workers, officials gradually began to view repatriation as the most desirable option for the nonwhite refugees of decolonization who increasingly made up most of the world’s displaced people. “International co-operation and solidarity should be directed, first and foremost, in favor of the solution of return,” the UNHCR declared in 1985—even if nothing much had changed in the refugee-generating state. In practice, what this meant was that placement in third-country employment, the “solution” offered to so many Europeans in the aftermath of the Second World War, would now be reserved mainly for a small number of political dissenters from the Soviet system. Under this new banner of repatriation, African, Asian, and Middle Eastern refugees fleeing violence, famine, or occupation rather than demonstrable “persecution” (and therefore excluded from the protections of the Convention) would mainly be subject to mass internment pending some theoretical future return.
By the last years of the Cold War, then, the longstanding internationalist idea of making use of refugees as laborers for the global expansion of imperial capitalism seemed to have gone the way of the dodo, with “solutions” of internment and repatriation replacing the older solution of low-wage, far-flung refugee employment. But even as international policymakers sought to defend the borders and the labor markets of the Global North by keeping refugees interned in place, they repeatedly hearkened back to the possibilities of treating refugees as workers. Eventually, the temptations of using refugee labor would prove too great to ignore altogether.
To Live Like a Refugee
After 1979, with the diminution of refugee flows from communist Eastern Europe and the ever-increasing numbers of decolonial refugees in Asia and Africa, the Cold War-derived premises for even a small number of resettlements were weakening. The UNHCR, at least, viewed this development as an opportunity to double down on return as the ultimate answer for mass displacement, and to abandon once and for all the work schemes of an earlier era. Repatriation was replacing employment as the institutional answer to the problem of mass displacement.
The League of Nations, constructed to protect an imperial world order for the victors of the First World War, was charged with concocting some way of dispersing the war’s many refugees.
There was no question that this turn to repatriation, voluntary or otherwise, would present substantial philosophical and political challenges: for a key element of the refugee regime encapsulated in the 1951 Convention Relating to the Status of Refugees was the principle of “non-refoulement,” or the right of the refugee not to be returned to the threatening country. But it was part of a long-term shift in the UNHCR’s move away from resettlement of any kind. In the early 1990s, UNHCR officials began to circulate the term “preventive protection”—referring not to the promotion of asylum, which the High Commissioner coolly noted in 1992 “remains an essential protection mechanism, but [is] not always the appropriate long-term solution,” but strategies “to attenuate the causes of departure and to reduce or contain cross-border movements or internal displacements.” The doctrine of preventive protection, in other words, emphasized the immobilization of refugees and made clear that the contemporary UNHCR’s primary responsibility was not to the refugee but to the state.
This approach followed national trends in migrant policy in Europe, where “reintegration” (read: expulsion) schemes were increasingly the order of the day. The Netherlands put in place a return scheme for migrants from Turkey, Morocco, and Tunisia in 1974; France followed in 1977, and Germany in 1983. These programs essentially represented an invitation to self-deportation, offering a legal alternative not to staying on but to forcible removal; in the French case, it explicitly indicated applicability to migrants without papers and to those who had already been served with an “invitation to leave the territory.” Now the UNHCR was committing to follow these state-level examples, proactively declaring that its role was not primarily to assist refugees but to prevent their emergence, reduce their numbers, and minimize the disruptions their presence caused to the emerging neoliberal system. If pleas and coercion failed to move refugees back to their home countries, the UNHCR would enact what it called “source control”: ensuring not that refugees would be able to seek asylum elsewhere but that they could be minimally protected and contained in situ, with a view to some eventual future repatriation.
The long-established practices of internment and restriction so ingrained in the UNHCR and its parallel organizations now became both more overt and more exhaustive, in conjunction with the development of ever-more-sophisticated methods of mobility restriction at the national level. In Bosnia in the early 1990s, the UNHCR fielded dozens of offices and thousands of personnel to enforce repatriation and containment, on the principle that, as one UNHCR working group put it, it was better to “bring safety to the people, rather than bring the people to safety.” In some notable instances, UN-linked “humanitarians” carrying out refugee containment-turned-repatriation operations deployed state militaries—not infrequently the very forces that had helped to produce refugee flows in the first place. It was American army personnel who organized the mass transfer of Kurdish refugees fleeing Saddam Hussein’s brutalities in Iraq during the American invasion of the country in 1991, a development followed by the declaration of American, British, and French-enforced “no-fly zones” to simultaneously “protect” and contain Kurdish refugees trying to escape. (The United States and the UN agreed to avoid the appearance of impropriety by not flying American and UN flags simultaneously over military-run refugee camps.) In the Balkans, UNHCR offices provided home bases for military crews running airlifts to Sarajevo.
Repatriation thus continued to beckon as the best of all possible “solutions.” Increasingly, such operations did not even pretend to be voluntary. In 1996, the UNHCR officially allowed that it would be willing to “contexualise standards” for voluntary repatriation—making clear, in other words, that there were circumstances under which the agency would sign off on forced return. Withdrawal of social services or basic welfare provisions became an especially common way of compelling people to opt for returns deemed voluntary by agents and states. In 2010, for instance, a British repatriation program for families withdrew social services from unaccompanied Afghan children in the UK as a way of forcing their “voluntary” removal to Kabul. (In a notable display of circular logic, the UNHCR also adopted the principle that refugee return could itself be taken as evidence of an improvement of country conditions.) As with its increasingly active involvement in the legal process of determining refugee eligibility, the organization presented removal schemes as a necessary way of shoring up the legitimacy and efficacy of the broader asylum system to ensure state-level cooperation: removal was a defensible aspect of its mandate to protect “genuine” refugees from the shadows cast, and the resources wasted, by “illegitimate” ones. Eventually, the UNHCR was openly offering its services to assist states as varied as Switzerland, Tanzania, and Turkey in deporting refugees whose asylum cases had been rejected for one reason or another, including unaccompanied children. Ironically, this deportation-centric approach would turn out to coexist quite comfortably with remaking refugees as a migrant labor force who might in some circumstances be permitted to stay on as workers.
Forced or coerced repatriation proved easier to implement in conjunction with a new legal status for refugees: that of “temporary protection,” which turned refugees into an (even) more precarious category of migrant with fewer protections and rights. As a legal concept, temporary protection had no prior basis in international law. Nevertheless, it quickly became a central premise first of European and then of international refugee policy for the opportunity it presented to keep refuge-seekers in a more or less permanent state of legal limbo—protected to some degree from forcible refoulement, but without access to local integration or permanent legal residency, employment, or citizenship and therefore more easily encouraged to return home. (As the International Organization on Migration euphemistically put it, “The limited duration of temporary protected status has proven to be an effective inducement to voluntary return.”) And it had another benefit: even when it did not successfully effect return, it could remake refugees as unprotected temporary labor with few to no legal or political guarantees.
The doctrine of temporary protection received its first detailed expression in the context of the Bosnian War of the early 1990s. A UNHCR document from 1994 outlined the rationale for the policy, arguing that “Temporary protection arrangements offer a means of providing protection to those who need it while continuing to emphasize repatriation as the preferred solution.” This novel legal category provided an alternative to categorizing the refugees streaming out of the former Yugoslavia as “convention refugees,” fleeing solely on the basis of persecution and therefore eligible for the convention’s more extensive set of guarantees. The reason for the establishment of temporary protection was a straightforward unwillingness on the part of other European countries to accept them as refugees eligible for asylum. Its primary benefit was the understanding that it could be terminated at any time, unlike the rights guaranteed by the 1951 Convention, which assured certain refugees of a permanent set of rights to asylum.
The UNHCR itself likewise came to understand temporary protection as a potential path to suspending the guarantees of the Convention at least provisionally, in ways that might be convenient for receiving states. In 2012, it held a roundtable on the question of temporary protection to explore just such a possibility. A discussion paper outlined the legal specifics for deploying the concept of temporary protection to relieve states from their legal obligations to refugees, holding, among other explanations, that “temporary suspension or derogation of the Convention is permitted because of the impact of the movement on the stability and security of the receiving State.” As per usual, the UNHCR was following on extant state approaches: the concept of temporary protection had already became the basis of an enormous number of subcategories in various European nations’ migration law, most of which explicitly foreclosed the possibility of a path to naturalization or other forms of permanent residency.
The application of these novel categories did not, of course, actually prevent refugees from entering a country and working there; indeed, in some ways they hearkened back to the pre-1945 refugee regime. Temporary protection and its related categories had the practical effect of remaking refugees as unprotected transitory labor, joining the ranks of undocumented or under-documented migrants working on sufferance in a sphere from which they would be perpetually subject to expulsion—a situation broadly viewed as beneficial for employers and states alike. It was in this context that the UNHCR itself began to traffic in the emerging concepts and practices of “managed migration,” a euphemism for the conflation of refugee and migrant status and a broad reduction in the rights of mobile workers across different legal categories. In 2005, following a long series of fora intended to produce a new internationalist policy surrounding questions of global migration, the UNHCR signed off on what it called an “International Agenda for Migration Management.” This agreement advocated for an expansion of legal channels of migration through multilateral agreements among states, on the basis of neoliberal schemes of economic development paying special attention to questions of labor. “Regulated migration,” it noted calmly, “can help ensure the availability of labour when the host country requires it. . . . A key element to achieve a more humane and orderly movement of people is a systematic, managed approach to migration, including consideration of different categories,” including refugees, who would be regarded as simply a variety of migrant and considered within the same framework.
The following year High Commissioner António Guterres proposed a new plan of action for dealing with “Refugee Protection and Mixed Migration,” including volunteering UNHCR staff to train border guards to differentiate between migrants and refugees and forcibly deport the former. This was not, in fact, an entirely new idea. The UNHCR had already dipped a toe into “migration management” in the Middle East, particularly in Turkey, where its officials had decades of experience in “refugee status determination” (that is, deciding who qualified as a refugee and who did not). Beginning in 1994, the agency began to offer training sessions to Turkish officials on the legalities of asylum and refugeedom as a first step in constructing a UNHCR-vetted internal asylum regimen there. This development was designed to allow the European Union to declare it a “safe third country” where refugees were theoretically able to seek protection—a legal basis for denying the claims of any asylum applicant in a European state who had first passed through Turkey. Now, backed by Guterres’s plan, the UNHCR began to run not only refugee camps and “reception centers” in Turkey but also deportation facilities, an action it defended as necessary to protect positive public opinion vis-à-vis the rights of “genuine” refugees.
Such schemes of delimiting access to the refugee regime served not so much to exile asylum seekers altogether as to remake them as itinerant undocumented labor in their new host country—always vulnerable to deportation but more likely to hang on as cheap, replaceable migrant labor without substantive rights. The very proliferation of these categories of temporary protection further served to create a new pool of workers whose presence and work permissions were short-term and often unclear—a kind of deliberately produced obscurity, designed to offer maximum leeway in either direction to the governing state.
Special Exploitation Zones
Remaking refugees as unprotected migrants was not the only possibility for their forcible integration into a global labor market. In the context of the Syrian civil war, the Middle East became the site of a new experiment in combining physical restraint with industrial labor: the placement of refugees in menial work within enclosed, monitored, walled zones.
What some saw as a “‘win-win’ outcome that suits both Northern donors and Southern hosts” was, in fact, an agreement involving a total disregard for the interests of refugees themselves.
The idea of “Special Economic Zones” was first mooted in the late 1950s, at roughly the same moment as the installation of the modern refugee regime. Although it had several antecedents, most observers agree that the first modern SEZ was located at Shannon Airport in Ireland, which had previously served as a necessary fuel stop for transatlantic air traffic. After the invention of the long-haul jet deprived the airport of its main reason for being, local planners tried to reinvigorate the space as a “customs-free” industrial zone. Companies would be exempt from national and local taxes and would receive incentives for the purchase of machinery and for worker training. The idea caught on quickly, producing some other early SEZs that were mainly “export processing zones,” physically enclosed areas featuring stringent customs controls, in which most products were destined for export. These zones began to advertise themselves as a venue for the legal deployment of cheap, impermanent labor. During the 1960s and 1970s, the United States and Japan jointly undertook to create SEZs in South Korea and other parts of Asia as a way of expanding Japan’s role as an American trading partner and offering, as one advocate put it, “an alternative strategy to the guest worker programmes pursued in countries such as West Germany. . . . Japanese zones would allow Japanese firms to manufacture using cheaper labour without allowing migrants to settle in Japan.” By the end of the decade there were SEZs in China, Hong Kong, Singapore, Taiwan, South Korea, Malaysia, the Philippines, Mauritius, and the Dominican Republic.
The model of the SEZ took hold in precisely the same economic atmosphere that had diminished the idea of refugee employment schemes: one of increased urban migration and widespread underemployment, ever-more-granular production strategies allowing for the use of exchangeable unskilled labor across different locations, and new technologies of transport and communication that could easily connect different zones of production. The United Nations, in the form of the UN Industrial Development Organization (UNIDO), actively promoted the construction of special zones as a valuable path to foreign investment and rapid development, touting the benefits of “low cost labour and freedom from industrial unrest.” When faced with the accusation that the zones exploited workers, the UN threw its weight behind employers. “Some zones may outlaw unions and ban strikes,” a 1994 UNIDO handbook on the zones conceded, but “within zones, factories are new, lighting conditions are good and canteen and medical facilities are often available. . . . If workers were not better off being ‘exploited’ by EPZ firms, the latter would find it impossible to hire them.” Still, a presentable public face was important. Acknowledging that SEZs had to have enforceable borders to function properly, UNIDO warned that “the appearance of the fence should be aesthetically acceptable—no barbed wire.”
SEZs in their first decades often hired migrant labor but did not specifically target refugees, whose international protections often rendered them less desirable in the eyes of employers. But the Syrian civil war that began in 2011, and the massive refugee crisis it sparked, prompted new consideration of using SEZs for the exploitation and control of refugees and their labor. In 2016, the Jordan Compact, a three-way agreement between the Jordanian monarchy, the European Union, and the UN, proposed a new kind of international deal vis-à-vis refugee labor. The Jordanian government would issue two hundred thousand work permits to Syrian refugees and provide schooling for Syrian refugee children in its territory in return for $700 million in grants and $1.9 billion in concessional loans. In addition, goods made in eighteen Special Economic Zones by Jordanian companies employing at least 15 percent Syrians would be eligible for a number of trade concessions: the relaxation of the EU’s rule of origin restrictions, tax exemptions, subsidized transport and electricity, less restrictive licensing and permitting procedures, and substantial administrative assistance. (After two years, the requirement would rise to 25 percent.)
Sweetening the pot even further, the World Bank gave Jordan access to $1.4 billion of credit at rates normally only available to low-income countries. Outside the scope of the scheme itself—but clearly intended to further prop up the Jordanian government’s capacity to keep refugees within its own borders—the European Commission also routed some €584 million into Jordan to support refugee efforts there. In addition to this funding, refugee aid from other donors in Europe, the United States, and elsewhere rose from just over $3 billion in 2016 to more than $4 billion in 2019. Jordan has now received more refugee-related aid than any other country in the world.
The Jordan Compact married two longstanding strands in international refugee policy: a commitment to physically containing refugees in situ as far as possible from the Global North, and an equally longstanding commitment to exploiting refugees as cheap labor for global industrial production. Two of its original architects, the refugee scholars Alexander Betts and Paul Collier, perfectly encapsulated the Compact’s intentions when they described their goal as a “‘win-win’ outcome that suits both Northern donors and Southern hosts”—that is, an agreement involving a total disregard for the interests of refugees themselves. Unsurprisingly, it aroused considerable refugee resistance. Refugees did not want to travel to work for low wages in poor conditions in the special zones, which were mainly located far from the urban centers where most of their families and other contacts lived. Transport to the zones was difficult or nonexistent; rent there was high (in some cases, higher than wages); and childcare was generally unavailable on site, making it difficult for women to participate in the scheme. Further, the SEZs forbade refugees from participating in a number of sectors, including virtually all professional work; the scheme was intended to enforce refugee participation in menial labor related to manufacturing and exclude them from any other kind of employment irrespective of their own prior qualifications. Echoing the much earlier refugee worker schemes of the interwar and WWII periods, refugee employment was to be deployed solely for mass industrial or agricultural production, and refugees would be permitted only to serve as the most menial kind of laborers.
Ultimately, this kind of externally supported development broadly failed to capture the interest of either investors or industries. By 2018, the scheme had attracted only eleven companies, and few of those proved capable of actually producing goods for European export. The uptake of these refugee employment incentives, on the part of both refugees themselves and hiring companies, was so modest that the EU eventually scrapped its initial requirement that would have raised the Syrian refugee percentage to 25 percent of the workforce, reverting to its original more modest aim of 15 percent. The Jordanian government, too, shifted its attention away from the SEZs, setting up new employment centers in the Zaatari camp in 2017 and Azraq in 2018 that issued refugee work permits that could be used outside the special zones, though most remained tied to a specific employer. The compact is theoretically still in effect, but it appears to have largely failed in its effort to produce a tightly confined pool of cheap laborers for the expansion of industrial manufacturing in Jordan. In language reflecting the persistent twentieth-century determination to regard refugees as mere cogs in the machine of industrial capitalism, a 2019 Financial Times essay declared that the essential problem of the SEZ was one of profitability. “SEZs have the tools to pilot programmes that would be difficult to start from scratch across a country,” its author declared, “but such schemes must be sound from a business perspective.”
The failures of the Jordan Compact illustrate a striking truth about the contemporary labor market: that Syrian refugees, vulnerable and exposed as they were, nevertheless represented a kind of privileged subcategory of migrant with more protections and greater demands than other possible sources of labor. Since the 1970s, Jordan, like so many other developing nations, has encouraged the use of migrant labor in its manufacturing sector, particularly from Egypt and South Asia. Jordanian garment production, one of the country’s most significant export sectors, is legally permitted to recruit three-quarters of its workforce from abroad and to exempt migrants from labor law protections that apply to Jordanians (for instance, eight-hour workdays). Syrian refugees, with their local connections, higher wage requirements, expectation of better working conditions, and connections to international organizations offering attention and support, represent an unattractive prospect for Jordanian employers accustomed to a migrant workforce enjoying no such protections. As JGATE, the trade organization representing Jordan’s garment industry, told an investigator in 2016, “Migrant employees are the CORE of this sector due to their higher efficiency and productivity, skill levels, willingness to work overtime, lower turnover rate . . . [and] cannot be REPLACED by Jordanians nor Syrians.” To the extent that the industry was willing to formally employ Syrian refugees, then, it was only as a replacement for the 25 percent of workers who were Jordanian—not as an aspect of the extant migrant labor force.
Ghosts in the Machine
Even as internationalist principles and practices of containment and repatriation were painfully hammered out, then, the idea that refugee labor should somehow be deployed in support of industrial capitalism has continued to exert an apparently irresistible pull on policymakers. Within this new post-Cold War context of mass confinement, two main ideas for incorporating refugees as workers emerged. The first was a discursive one: deliberately blurring the categories of refugee and migrant through the advent of a new legal category of temporary protection, forcing refugees into an informal labor market by stripping away their formal protections under the Convention-era rights regime. The UNHCR participated vigorously in this scheme by volunteering to police claims on refugee status, and training various country nationals to do the same. The second approach was more experimental: the idea that refugees could be deployed as laborers within an institutional confinement. The subsequent advent of the refugee-linked SEZ, in which the displaced could simultaneously be detained and exploited, represented a new vision for how to wring labor out of refugees without releasing them into protected national, regional, or global labor markets.
The idea that refugee labor should somehow be deployed in support of industrial capitalism has continued to exert an apparent irresistible pull on policymakers.
In 2019, reflecting this shift toward once again trying to make use of refugee labor, the UNHCR officially abandoned its longstanding commitment to its three basic “durable solutions” and added two other possible aims: “other local solutions” and “complementary pathways.” The vagueness of such schemes offered host states maximum flexibility in dealing with the refugees who come through their territories, making use of UNHCR bureaucracy and documentation to support a range of options from informal labor to deportation. “They don’t have work permits,” one shaweesh (an intermediary serving as a labor contractor for refugees) in Lebanon told an interviewer in 2021 regarding Syrian refugees.
But there’s a tacit oral understanding with the state that in Lebanon, in the Bekaa area in particular, landowners are in need of workers and that the shaweesh provides these workers. . . . With the outbreak of the Syrian revolution and the flock of refugees, the state now knows how many workers I supervise and how many people live in the camp through the Army Intelligence, the State Security, the municipality, the Information Branch. . . . I have to inform the intelligence when anyone leaves my camp.
Rohingyan refugees in Malaysia operate in the food service sector in much the same way. Known as “ghost” labor, they work in markets as vendors and producers under the “protection” of organized crime syndicates with active connections to local police and authorities—a situation that gives workers a semi-legal presence (often based on their possession of UNHCR refugee cards) but also premises their continued employment and indeed presence on an uncomplaining silence. “I also understand and accept that as a migrant, I have less freedom to say or do what I want, even if I am in the right,” one such refugee told an interviewer in Kuala Lumpur. “Arguing achieves nothing productive and only puts my employment and source of income at risk.”
In other words, the UNHCR has come to collaborate with host states to produce deliberately vague legal and political formulations encompassing any number of forms of long-term legally unprotected residence-cum-employment in countries of transit. These are almost always located in the Global South and increasingly feature circumstances of refugee confinement or surveillance or both. A carefully maintained ambiguity around the legal status of refugee workers allows them to remain in place, often occupying long-term positions or returning to the same employees on a seasonal basis year after year, working under observation from a position of political and economic vulnerability that sharply limits their capacities to make demands either of the state or of their employers—bringing them closer and closer to the experiences of undocumented labor migrants. After all this time, then, it would appear that the hoary old idea of putting refugees to work as cheap, expendable, mobile workers in a globalized industrial economy has made a remarkable—even a triumphant—comeback.
Excerpted with permission from Human Capital: A History of Putting Refugees to Work by Laura Robson. Courtesy of Verso Books.