Restricting Emigration for Their Protection? Exit Controls and the Protection of (Women) Migrant Workers

Abuse and exploitation of migrant workers in Gulf States is common and well-documented, and women domestic workers are at special risk. Sending states – often relatively poorer South Asian states – are limited in the ways that they can protect the rights of their citizens when they are labouring abroad. One strategy that sending states have deployed is the adoption of “emigration bans” or “emigration conditions.” Emigration bans restrict citizens from taking up temporary labour market contracts, usually in specific states, but sometimes in general. “Emigration conditions” require would-be migrants to meet specific requirements in order to be permitted, by the sending state, to take up a labour market contract abroad. In this article, I examine whether it is morally permissible for source countries to prohibit migration to countries where they risk being exploited or abused. I examine the reasons states give to justify emigration bans and conditions: the “structured vulnerability” reason; the “gendered structured vulnerability” reason; and the “gendered paternalism” reason. Overall, I agree that the reasons motivating the bans and conditions are good ones – though I offer some criticism of the reason I describe as “gendered paternalism”. But, since there is only limited evidence of the effectiveness of bans and conditions in achieving substantive benefit for labour migrants, and on the contrary evidence of the real harm they can sometimes generate, I argue that, absent positive evidence of their success in achieving their objectives, they ought to be rejected in practice even if they are permissible in principle.


Introduction *
Abuse and exploitation of migrant workers in Gulf States is common and well-documented.Women migrant workers, largely labouring in domestic work, are at special risktheir labour is done in private spaces, with little oversight or access to protection when it is needed.Complicating the situation is that, as matter of law, domestic labour migrants are often entitled only to a small subset of protected rights in the host state.Sending statesoften relatively poorer South Asian statesare limited in the ways that they can protect the rights of their citizens when they are labouring abroad.One strategy that sending states have deployed, often in response to egregious instances of violence perpetrated against their citizens, is to impose "emigration bans" or "emigration conditions."Emigration bans restrict citizens from taking up temporary labour market contracts, usually in specific states, but sometimes in general."Emigration conditions" require would-be migrants to meet specific requirements in order to be permitted, by the sending state, to take up a labour market contract abroad.The purpose of such policies is, generally, protective, in two ways.Sending states aim to protect their citizens from exploitation and abuse in destination states, first by denying them the right to travel to those states for work, and second, by pressuring host states to adopt better worker protection policies in exchange for removing the ban or conditions.Given this context, the purpose of this article is to examine whether it is morally permissible for source countries of temporary labour migrants to prohibit migration to countries where they risk being exploited or abused.Why might this be a challenging question?
These policies pose two dilemmas, which I examine in this article.One dilemma is that the mechanism of securing protectionbans and conditions on emigrationappears to be in violation of a fundamental human right, the right to exit.Yet, while international human rights law certainly does protect a broad right to exit, sending states do have duties to protect the fundamental rights of their citizens, whether they are at home or abroad.However, the mechanisms for protecting citizens who are abroad are limited, often to consular and diplomatic protection, and these may well be insufficient to protect the rights of temporary labour migrants.So, one dilemma to resolve is whether selective destination bans can be justified, even though they appear to violate a basic human right.There is a second dilemma to consider, however, in order to complete a full assessment of whether such policies ought to be pursued.In particular, even if they are morally permissible, it matters whether they are likely to be effective at achieving better protection for their citizens; the evidence here, I suggest, is mixed at best.Their record at achieving substantial benefits for workplace protection is weak, and moreover their imposition often makes matters worse for would-be labour migrants, by encouraging them to migrate following irregular pathways.So, a second dilemma to resolve is whether the evidence of their limited effectiveness in practice undermines the case for adopting these bans and conditions in the first place.
To resolve these dilemmas, I proceed as follows.I begin with a brief account of the way that political theorists of migration have treated temporary labour migration in general, and suggest that the worries that motivate the adoption of emigration bans are common across objections to temporary labour migration.I then situate the bans and conditions on migrant labour in Gulf States in the more general political theoretic discussion of temporary labour migration.Next, I consider the possibility that such bans violate the right of individuals to exit their state.I explain, * This Working Paper is part of the 'Dilemmas' project at the Migration Policy Centre (MPC), European University Institute (EUI) https://migrationpolicycentre.eu/projects/dilemmas.It was prepared with the excellent research assistance of Engi Abou-El-Kheir and Madeleine Berry.I owe thanks to Sarah Song and the organizers of the 'Dilemmas' workshop for written comments on an earlier version of this article, as well as to the participants at the workshop for very helpful questions.
here, the deep importance of the right to exit, and notice that emigration bans in particular restrict this right, though emigration conditions also make exiting more difficult for those who are subject to them.But, I suggest, since they are targeted bansthey typically ban travel to one specific country for one specific reason for a limited period of timethey are a limited constraint on the right to exit and thus may be permissible under the right conditions.To make this case, I consider how emigration bans are treated in the "brain drain" literature.Then, I consider the reasons given by states to justify emigration bans and conditions, reasons I have termed the "structured vulnerability" reason; the "gendered structured vulnerability" reason; and the "gendered paternalism" reason.Overall, I agree that the reasons motivating the bans and conditions are good onesthough I offer some criticism of the reasons I describe as "gendered paternalism".But, since there is only limited evidence of the effectiveness of bans and conditions in achieving substantive benefit for labour migrants, and on the contrary evidence of the real harm they can sometimes generate, I argue that, absent positive evidence of their success in achieving their objectives, they ought to be rejected in practice even if they are permissible in principle.

Temporary labour migration and emigration bans in political theory
Any moral evaluation of emigration bans and constraints requires situating them in the larger context of temporary foreign labour migration.In very general terms, temporary labour migration programs (TLMPs) invite migrants to take up short-term labour contracts, usually in industries facing acute labour shortages.Crucial is that the contracts are time-limited: when the work is complete, the migrant's visa expires and she must return home.In most cases, migrants who take up these opportunities are citizens of relatively poorer countries who are travelling to relatively wealthier countries.Although there is a good deal of variation among host countries, typically the rights of migrant workers are constrained, and often those that they do possess are not well-protected.Yet, migrants are often enthusiastic about the opportunities made available to them via TLMPs, since they offer access to (comparatively) better quality employment options; usually, migrants are leaving countries where there is high unemployment, or where the work opportunities that are available are poorly remunerated.They offer an additional benefit, which is that temporary labour migrants send a large proportion of their wages to their home countries in the form of remittances; these remittances, in addition to the skills and experience that migrants bring home with them when they return, support development across many Global South states (Gupta, Pattillo, and Wagh 2009;Catrinescu et al. 2009;Lim and Basnet 2017;Bedford et al. 2017).
When scholars are critical of TLMPs, the criticism typically focuses on the permissibility of restricting the rights of temporary labour migrants in their host country.Again, although there is a great deal of variation, labour migrants generally are not permitted to travel with, or later bring, their families and they are often denied the right to apply for permanent status in their host countries (no matter how long they have been labouring there).At a more basic level, labour migrants in some countries are restricted to one employer (or one industry); they may have access to only basic health care (and usually not other public services); they may not be permitted to join labour unions; and so on.Whereas some scholars (including me) suggest that these rights restrictions are sufficiently grave that, as far as is possible, temporary labour migration programs ought to be abandoned (Lenard and Straehle 2011;Lenard 2012), others argue that some of these rights restrictions (not the most severe ones) may be an acceptable "price" to be paid for securing access to the labour market in host countries for citizens of poorer states (Ruhs 2013).The challenge they pose is that, as is well-documented, these restrictions are often correlated with substantial exploitation of labour migrants (see in general Attas 2000;Lenard and Straehle 2010;Mayer 2005;Stilz 2010).So, at least normatively, political theorists typically operate by weighing the benefits to "remedying global wealth inequalities" against the rights restrictions to which labour migrants are subject, to offer views on the overall legitimacy of TLMPs on a global scale (see also Bauböck and Ruhs in this working paper series, 2021).For many, the exploitation to which migrant workers are subject is so great that they advocate for abandoning them (Lenard and Straehle 2011).But there is a complication that arises for those who make this argument, which is that as a matter of (unfortunate) historical fact, where opportunities for legalized movement are closed, migrants persist in moving, using irregular channels to take up irregular work opportunities (Massey and Lang 1989;Hahamovitch 2003).These migrants are even less protected from exploitation than labour migrants who take up TLMP opportunities in spite of the rights restrictions to which they are thereby subject.Such migrant workers are highly vulnerable because they cannot avail themselves of the protection of the state (which they must avoid in order to stay in the country and keep their job), and too often employers prove willing to exploit this vulnerability, by requiring migrant workers to labour for longer, for less pay, and under poor working conditions.The result is that, even if in principle there are normative problems with TLMPs that lead scholars to advocate for abandoning them entirely, there are similarly reasons to permit them to persist, all the while working to improve the working conditions for them, including advocating for expanding the set of rights to which they are entitled.I will return to this practical question, below, when I consider the legitimacy of emigration bans and constraints in practice.

Emigration bans and conditions for temporary migration to the Gulf States
Let me first situate the specific case I'm considering in the terms set by this more general debate.The Gulf States receive thousands of migrant workers every year, most of them from ASEAN countries.Migrant workers take on many jobs, including in manufacturing, fishing, construction.Historically, migrant workers were mostly men, but women are making up an increasing proportion.In 2013, 44% of migrant workers, globally, were women (Migration Data Portal 2021); as of 2019, they made up roughly 30% of the migrant workers in the Gulf States (ILO 2021). 1 In the Gulf States, women migrant workers are largely working in domestic labour broadly understood and, in a smaller number of cases, in so-called entertainment industries, for example as dancers, singers, or masseuses.Just as described above, migrant workers choose to work abroad, on temporary visas, because there are relatively fewer options at home, and because the options abroad are relatively lucrative.In several countries, employment opportunities at home are further reduced for women, who are expected to stay out of the work force in order to care for family members, or simply because of (outdated) gendered expectations that formal employment is, or should be, restricted to men.
Although essential to the economies of many Gulf States, migrant workers labouring there both men and womenoften possess few rights, and labour under exploitative conditions (Shah 2006;Lenard 2014).Women's vulnerability to exploitation, both in domestic work and in "entertainment" work, is often heightened, for myriad reasons, including that in some cases domestic workers officially possess few rights (fewer than male migrants, that is), and typically domestic workers labour in the homes of their employers, giving employers tremendous power to control their work day (they often ask for more than the maximum permitted hours, refuse time off, and so on) (Islam 2016).Domestic workers are often young and uneducated, and do not speak the language of the host country, and so they have fewer internal resources on which to rely in the face of exploitative conditions as well.The incidence of sexual and physical abuse of domestic workers is high.2 Among sending states, it is well-known that migrant workers are often the victims of abuse and exploitation, and efforts of various kinds are made in attempts to protect them when they labour abroad.Emigration bans and emigration constraints are among the tools deployed by sending states in an attempt to protect their workers.Historically, men and women migrant workers have been subject to emigration bans, although women more often so.For example, Indonesia banned the migration of would-be male plantation workers in Malaysia in 2009, and in 2014 Cambodia banned men's migration into the fishing sector in Thailand.While these are not isolated examples, women are certainly subject to a wider number and range of emigration bans and, more often, to a wider range of conditions imposed by the sending country on their migration.So, to give just a few of many examples, Laos banned domestic labour migration in 2002 and again in 2013; Vietnam banned migration of "entertainment" workers in 2007; Indonesia banned domestic labour migrants to Malaysia in 2009, to the Middle East in 2011.Sometimes the bans are to specific countries and sometimes they apply to industries.Bans are distinct from emigration conditions, which are generally imposed only on women.In some cases, women must meet certain age requirements to take up temporary labour contracts, or get permission from their fathers or husbands to work.In others, they are not permitted to travel if they have young children at home.In what follows it will be important to consider the permissibility of bans separately from the permissibility of conditions.
Usually, emigration bans are implemented after a particularly grievous instance of abuse or exploitation becomes public, and origin states demand retribution of some kind.In an attempt, sometimes real and sometimes for show, to gain better protections for their workers, and to appear that they are doing something to defend their citizens, states adopt emigration bans until such time as the receiving state promises to do better at protecting the rights of labour migrants (Shivakoti, Henderson, and Withers under review).In most cases, the bans are relatively short-lived, and migration is again permittedin some cases, permissible migration begins after a Memorandum of Understanding (MOU) is signed between sending and receiving states, committing receiving states to better protect of migrant workers.In others, the demand for legal channels of migration is so high that sending states simply reverse the ban, after the furor has died down.

Brain drain restrictions versus temporary labour migration bans
Political theory discussions over the permissibility of emigration bans have, typically, focused on so-called "brain drain" situations.These are cases where high-skilled migrants from developing states migrate to take up opportunities offered to them in developed states (for example see Oberman 2013; Brock and Blake 2014).Many discussions of brain drain focus on the movement of medical professionals, both nurses and doctors, often educated at homestate expense.When defending exit restrictions in the context of a brain drain, scholars usually propose that such bans are permissible under various conditions, including that educated professionals "pay back" the cost of their education in some way.This pay back can be done either monetarily or by labouring for a specified amount of time in their home state (a common emigration ban is in the form of a 2-year compulsory service in the health sector requirement), especially where state-capacity is likely to be strained if state-educated medical professionals exit immediately in search of better opportunities.This way, the educating state reaps at least some benefit for their expense, or, at least, doesn't lose as a result of it.The general view is explained by Lea Ypi: "emigration may be legitimately restricted when allowing for the outflow of particular categories of emigrants leads to a reduction in the general welfare of the sending society" (Ypi 2008, 409).
Of course, not all political theorists defend the right of would-be sending states to erect emigration bans; some argue that the costs to individual liberty, of refusing the right to exit in search of better employment opportunities (especially where employment opportunities at home are poor and poorly remunerated), are too high to countenance (Sager 2014).Others propose that the correct strategy is to require receiving countries to engage in some forms of compensatory behaviour towards sending states, for example by repaying some of the cost of educating the migrating professional.One example of such a proposal is "the Bhagwati tax", a special tax on the income that high-skilled workers earn in host states, which is redirected towards developing countries (Bhagwati 1976).
This debate, over whether the right to exit of educated professionals can permissibly be denied, does not translate well to the case under discussion here, however.Certainly, some of the dynamics are similar.In both cases, the sending states are relatively poorer and have little power that they can realistically exert on destination states to extract compensation for their losses.As well, in both cases, employment opportunities in the sending state are such that migrants have strong incentives to leave in search of better ones, and yet migrants are denied the right to seek, at least temporarily, better employment.But, the reasons for the emigration bans are quite different in both cases.In the brain drain case, the reasons centre on the costs to sending states of losing educated (largely medical) professionals; the emphasis on costs of the brain drain also allows for an easy solution, in the form of having such professionals pay back the cost of their education.However, in the case under discussion here, the reasons have a paternalist flavour, that I will consider in some more detail in section 4, and centre on the potential harm to migrants, and importance of attempting to protect them from such harm.The benefits of the restrictions in the latter case are, at least in principle, to the workers whose rights may, as a result of a successful negotiation to re-introduce migrant labour flows, be better protected.
In both cases, the bans on emigration violate a basic human right, namely, the right to exit one's state (Whelan 1981).One major purpose of the right to exit (as opposed to the broader right to move) is protective: citizens must be able to escape oppressive and predatory governments, if their lives or basic security are at risk.But the right to exit in general, rather than in search of security, is nearly as important; it is one key right that individuals require in order to live an autonomous life, and can only be curtailed permissibly under exceptional circumstances.As described for example in the International Covenant on Civil and Political Rights, "Everyone shall be free to leave any country, including his own."The constraints on this right are limited: "The above-mentioned rights shall not be subject to any restrictions except those which are provided by law, are necessary to protect national security, public order (ordre public), public health or morals or the rights and freedoms of others, and are consistent with the other rights recognized in the present Covenant" (OHCHR 1976, article 12).
While the right to flee from persecution is violated by general exit bans, such as that in North Korea, it is not violated by selective exit bans that prohibit migration to specific destinations.Moreover, in at least one sense, the emigration bans imposed on would-be migrant workers are typically less restrictive than are emigration bans in the brain drain cases.In most cases, they apply only to migration to a particular country, and only with respect to taking up particular forms of employment.Would-be migrant workers, at least in principle, remain free to travel to other countries, and to take up other work; indeed, they remain free also at least in principle to travel to those countries targeted by labour emigration bans, for other reasons, for example to visit friends or family.
The difficulty of course is that the women migrant workers who are the focus of this article have very few options in the first place: they are often taking up migrant labour opportunities because they have difficulty finding employment at homenot only because the employment sector is less robust, which is also true in brain drain cases, but also because of gendered norms that restrict their perceived employability (this latter is not treated as a major consideration in the literature on brain drain, typically).Women migrant labourers from ASEAN states mostly choose domestic labour because they are not trained to take on other options, and gendered norms keep them out of employment in other common areas of migrant labour, including manufacturing and construction.Moreover, citizens from relatively poorer ASEAN countries may have trouble accessing visas to most countries; even where they are in principle permitted to travel elsewhere, they will struggle to meet increasingly demanding visa requirements (including especially those that demand proof of financial viability).So, the emigration bans imposed by ASEAN countries, on women domestic labour migrants may, comparatively, impose more substantial freedom restrictions on them, compared to the emigration bans imposed on skilled workers in brain drain situations.In neither case do emigration bans violate the narrowly conceived right to exit as a right to flee from persecution or violence, but in both cases coercive action by the state restricts the options that would-be migrants might otherwise have available to them.

Justifications for emigration bans imposed on women domestic workers
In what follows, I examine the normative plausibility of three central reasons given to justify emigration bans and conditions.All three of these reasons are paternalist, in the sense that they are at base justifications for bans that aim at protecting the rights of their citizens.Their controversy stems at least in part from the fact that, while state paternalism is often justified domestically to protect citizens from certain harms, the permissibly of deploying paternalist arguments to restrict exit, even partially, is normatively suspect.As I noted earlier, international human rights covenants protect a broad right to exit, and no provision within those covenants is made for restricting it on the basis of "paternalist" reasons.The three reasons are: the reason from structured vulnerability; the reason from gendered structured vulnerability; and the reason from gendered paternalism.The first two reasons are attentive to the way in which temporary labour migration schemes are typically structured, noticing that they can be structured to do better to protect migrant workers.The second acknowledges the specific ways in which the structure of such schemes render women especially vulnerable.The third emphasizes the allegedly appropriate social norms to which women must adhere, and their allegedly inferior capabilities, even as they accept migration opportunities.
The reason from structured vulnerability, offered by states, in favour of emigration bans emphasizes the vulnerability that migrant workers face as a result of the way in which temporary labour migration schemes are typically structured, i.e., in ways that permit or even demand rights restrictions of certain kinds.On this view, emigration bans are justified because they protect their citizens from the abuse and exploitation that TLMPs permit and sometimes encourage.As evidence of their protective intent, emigration bans are typically adopted in response to especially egregious instances of abuse or violence.To give just some examples of precipitating events, the Philippines adopted emigration bans in 1995 and 2008, to Singapore and Kuwait respectively, after their citizens were executed (in Singapore for the alleged murder of a child in the domestic worker's care) or murdered.It adopted a ban (2009) preventing emigration to Malaysia after several reports of abuse against its domestic workers.Nepal adopted an emigration ban in 1998 after a citizen committed suicide, citing the ongoing physical violence and rape by her employer in Saudi Arabia.Both Indonesia (2011) and Sri Lanka (2013) adopted bans on emigration to Saudi Arabia, after their citizens were executed by the state, both because of alleged crimes they committed (Shivakoti 2020).In defending labour migration bans to Kuwait, President Rodrigo Duterte cited the importance of treating "Filipinos as human beings" (Human Rights Watch 2018). 3Similarly, the Nepali government in justifying the imposition of bans, emphasized its role in protecting Nepali domestic workers from abuse at the hands of their employer (Shivakoti 2020, 29).Shivakoti quotes a senior official in the Nepali government acknowledging the possibility that recent bans may violate citizens' mobility rights, but says nevertheless "the government looks after the safety of its citizens" (Shivakoti 2020, 29).
The point of emigration bans is not, however, to restrict access to particular labour markets in perpetuity.Rather, it is to pressure host states to adopt better labour protection laws for migrants.The hope is that by restricting the flow of migration to them, destination states will be pressured to offer more protections in exchange for re-opening the flow of migrants.For example, when Indonesia adopted an emigration ban to Malaysia in 2009, the government "emphasized how this action reflected not only the problem of violent abuse against domestic workers, but also the need for better protections and rates of pay for these workers more generally" (Elias 2013, 395).An MOU promising better worker protections was signed in 2011, to restart migration from Indonesia to Malaysia (Shivakoti, Henderson, and Withers under review).
The reason from gendered structured vulnerability likewise focuses on the specific dangers posed to migrant workers by the structure of labour migration schemes, and emphasizes that women are especially rendered vulnerable by them.From this perspective, and in justifying emigration bans in terms of their protective role, the relatively higher vulnerability of women domestic workers is emphasized, and for good reason.For one thing, as I noted earlier, domestic workers often have fewer rights than migrant workers in generaland where their rights are outlined in principle, in practice they are often violated.For example, research suggests that over half of domestic workers labouring in Asian countries are exempt from "working hours legislation" and just about half have no legal entitlement to days off (Islam 2016).Legally and illegally, domestic workers are often denied the right to leave their employer's home; their passports are often confiscated; they are often denied payment or severely underpaid in relation to the number of hours they work.All of these rights violations, some of which are permitted as a matter of policy, others of which transpire because of the lack of legal enforcement of work contracts by external authorities, mean that women domestic workers are at severe risk of abuse and exploitation by their employers (International Organization for Migration 2009, 54).They are accepted in large part because of the way in which domestic work is devalued: "Domestic work specifically is hidden in private work spaces and is undervalued, seen as work that women "naturally" do, and thus not requiring any skill or commanding a living wage" (Napier-Moore 2017, 10).
That is to say, although below I will offer some critical comments on the gendered nature of emigration bans, it is important to observe at the outset that the additional vulnerability faced by domestic workersboth because destination states refuse to protect their rights on par with other migrant workers (who themselves are often the victims of abuse and exploitation) and because the nature of domestic work is such that those who carry it out are at greater risk of abuseis a matter of genuine concern.Attention to their vulnerability, and the adoption of gendered legislation as a result, is not straightforwardly a manifestation of discrimination against women.The heightened danger that women face (compared to other migrant workers) is genuine, and the difficulty of protecting women migrant workers in domestic spaces (compared to the space in which most other migrant workers labour) is severe, and wellacknowledged around the world (Macklin 1994;Stasiulis and Bakan 1997).
The protective reasoning that justifies emigration bans is also deployed to justify emigration conditions.Emigration conditions are just as they sound: conditions that workers must meet in order to be eligible, from their home country's perspective, to take up temporary labour migration contracts (they thus contrast with immigration conditions, imposed by destination states on would-be migrants).In the case of domestic workers, rather than adopt outright bans, a range of conditions are adopted, and I wish to consider them in two clusters here.One cluster of conditions are those that are imposed on would-be domestic workers, on the idea that if they are met, women will be better able to advocate for themselves and/or less likely to be the victims of abuse.The most common condition here is an age-requirementthe Philippines requires its domestic workers to be at least 23 years old; Cambodia and Indonesia require its domestic workers to be 21 (Thimothy and Sasikumar 2012, 27).In each of these countries, women labour migrants entering other sectors may do so at 18.The purpose of age requirements is to ensure that women labour migrants have attained a certain degree of selfconfidence and maturity that, optimistically, gives them the internal resources they need to advocate for themselves and, if necessary, to exit difficult employment situations to search for external support.The age requirement is adopted in the face of fears, in particular, that younger women are more vulnerable to sexual abuse, and at relatively higher risk of being forced into sex work or human trafficking schemes from which they cannot, or do not known how to, escape.Again here, the reality that women domestic workers face a particular set of challenges is important to acknowledge, especially given the lack of respect given to domestic labour in host states.This first cluster of conditions focuses on the unique vulnerability of women domestic workers and at least in principle aims at ensuring that women who take up these opportunities are able to protect themselves given the labour migration context.It is thus another form of reason from gendered structural vulnerability.
A second cluster of conditions is more explicitly infused with patriarchal norms about the allegedly proper place of women in society; these conditions are justified with respect to a form of gendered paternalism.These conditions emphasize the role of women as mothers, and their subservience to the men in their lives.They include, for example, the requirement that women have written permission from their parents or husbands to gain or renew passports or to take up domestic labour opportunities abroad.Women migrant workers from Sri Lanka and Indonesia are subject to this requirement, for example (Shivakoti,Henderson,and Withers under review,10), and the Nepali government has recently proposed adopting such a regulation as well, citing the need to curb the trafficking of Nepali women (Khadka 2021).In others, women can only travel if their children are over the age of 2 and sometimes 5 (Shivakoti, Henderson, and Withers under review).On this latter condition, the implication is that women should travel for domestic labour only once they have fulfilled their reproductive roles at home.This moralized understanding of the role of women at "home" is reproduced in other emigration bans that also impact women disproportionately.For example, Laos (2002, revised in 2013) banned migrant labour that is "contrary to Lao customs and traditions", which include cleaning, domestic work and sex work.Vietnam (2007) bans migrant labourers from taking up work as "dancers, singers, or masseuses in restaurants, hotels, or entertainment entities" (Napier-Moore 2017, 17-18).Although these bans were not formally gendered, they only applied to industries in which women predominate.
These conditions are especially troubling to liberal democrats who press against the gendered norms that force women into (nearly exclusive) caregiving roles, subservient to the men in their lives.Already in the countries from which women migrants come, there exists stigmatization of working women, some of whom may need to work in order to sustain their family's livelihood, alongside "stubborn expectations" which require "that women should, irrespective of their paid work commitments or their spouses' lack thereof, perform the bulk of unpaid care work and particularly childcare" (Shivakoti,Henderson,and Withers under review,10).These emigration conditions, in effect, incorporate "patriarchal familial ideology … into state policy" (Shivakoti,Henderson,and Withers under review,10).This state policy often translates into strong expectations, and sometimes coercively imposed ones, that women send their wages back home.
In this way, women migrants are represented as 'economic heroes' or 'martyrs' carrying out a feminine familial duty as 'good' mothers, daughters and wives whose remittances 'save' their countries, communities and households (Shivakoti,Henderson,and Withers under review,11).
All of the reasonsstructured vulnerability, gendered structured vulnerability, and gendered paternalismgiven in favour of emigration bans and conditions for women migrant workers are focused on protection in some sense.Their mode of operation is, fundamentally, to restrict the right to exit of women migrant workers, citing the importance of rights protection in host states; in so doing, whatever the reason cited by the state that imposes such bans and restrictions, women migrant workers are denied the opportunity to migrate to take advantage of certain temporary labour migration contracts.The first reason focuses simply on the role a state can play in ensuring that its citizens basic human rights are protected in their place of employment, both by preventing its citizens from taking up the most dangerous of opportunities and by pressuring host states to commit to better protection of migrant workers.The second reason focuses directly on the special risks faced by women in domestic labour environments.The third reason, which is also attentive to the special risks faced by women, is infused with patriarchal norms of gender inequality that treat women as children who are unable to protect themselves, and whose duties are primarily to their family.The language defending the second and third reasons are often blurred, but normatively it is important to keep them distinctwomen really do face unique challenges labouring in their employer's homes, and any attempt to protect them in their workspaces will need to be attentive to these challenges.The welldocumented abuse of domestic workers even in countries that are generally more committed to gender egalitarianism suggests this attention is key, even if exacerbated in states where women's subservience to men is generally accepted, in principle as well as in practice.
That said, what distinguishes the third form of reasoning from the second is that while the second reason recognizes women's vulnerability, the third treats them as inevitably bound to uphold patriarchal familial relations which rely on their subservience.The second reason is attentive to the fact that one can, at the same time, be both vulnerable and an agent in one's own life, and thus that the state can act to protect those who are vulnerable without undermining their agency, and indeed by offering them supports to better advocate for themselves, which is different from assuming they do not possess the capacity to exercise it (Straehle 2017).The third reason, however, infused as it is with expectations for the proper behaviour and place of women builds into the constraints the requirement that women, effectively, conform to such expectations.It thus operates mainly to constrain women, without necessarily offering the concomitant protection benefits.
That is to say, given the very grave harm that stems from restricting exit, even if limited in various ways, it seems essential to their justification that they offer protection benefits.So, it seems justified that sending states do what they can to protect the labour rights of migrant workers, given the limited options available to them to do so, and also that they should be attentive to the specific risks faced by women migrant workers.But, when their actions serve to reinforce gendered norms that persist in constraining women's agency and violate their human rights (rather than offer substantial labour market protection), they cannot be justified.

Do emigration bans/conditions work? (And how should this influence our moral thinking?)
The adoption of migration bans by sending states is an attempt by relatively less powerful states to bring relatively more powerful states to the negotiation table.The idea is that if destination states struggle to fill jobs that had previously been filled by migrant workers, they will be motivated to offer concessionsin the form of better worker protectionsto citizens of sending states, in order to allow for travel to resume.
The challenge sending states face in protecting their citizens abroad is well-noted in the literature.In general, sending states have few options by which to protect their citizens abroad, beyond offering diplomatic and consular services, and such services do not seem able to offer substantial protection to their citizens.This challenge is compounded by the relatively weak political position that sending states find themselves in: if the flow of migrant workers from one state is slowed or closed, then destination states will often respond by recruiting from alternative states where the flows main open.The result is generally that sending states that adopt bans no longer benefit from the labour markets in sending states (remittances are reduced), and secondarily, domestic labour migrants (who often chose this route out of desperation in the first place) choose in any case to migrate, but following unofficial channels.In other words, evidence suggests that where legal opportunities for migration close, women respond by travelling irregularlyoften with the support of recruitersand find themselves in even more exploitative situations, and without state protection of any kind (Napier-Moore 2017).As Napier-Moore explains: "Under employers' threats of informing authorities of a workers' irregular status, migrant workers become undemanding and uncomplaining, fearing arrest and deportation.Unbalanced employer-employee relationships create the conditions for forced labour, human trafficking, and other workplace violations.Blame for abuses is placed on individual employers" (Napier-Moore 2017, 51).Where protection is required, they turn to their initial recruiters whose incentives are to ensure that women stay with the employers who have paid their fee rather than to sanction employers who engage in abuse towards domestic workers.Some states (for example Singapore) even issue legal work permits to domestic workers who manage to gain entry irregularly, and in violation of a ban imposed by their sending states (Napier-Moore 2017, 26-27).
The pattern I have just described is well-observed, but it is not absolute.There are cases where the imposition of migration bans has brought destination states to the negotiating table.The Philippines has been especially successful here, since its citizens are in demand in many destination states for their linguistic skills and comparatively higher levels of education.In 1988, the Philippines banned its citizens from taking up domestic work abroad, after widespread reports of abuse.In response, 16 states were willing to sign MOUs outlining a more substantial commitment to protecting the rights of domestic workers, though at least one scholar suggests that the states that were most attractive to Filipina domestic workers were not among those who were willing to sign MOUs (Gonzalez 1996).In 1995, the Philippines adopted a ban preventing their citizens from migrating to Singapore, and there the result was a 1996 agreement between the Philippines and the recruitment agencies that placed Filipina domestic workers in Singaporean homes.Recruitment agencies that failed to protect workers whom they had placed risked losing the "accreditation" that allowed them access to domestic workers.The guidelines specified that Filipina domestic workers were entitled to "specified days off and set holiday leave, the employee's right to keep her passport, and the exclusion of car washing and body massage from the domestic helper's responsibilities" (Gonzalez 1996, 170).In 2012, an MOU was signed between the Philippines and Saudi Arabia to guarantee a minimum wage to domestic workers and better labour protections (Shivakoti,Henderson,and Withers under review,15).Similarly, Indonesia banned women from travelling to Malaysia in 2009, and the two countries signed an MOU to re-open the flow in 2011.The MOU outlined that Indonesian domestic workers were permitted to keep their passports, were entitled to regular days of rest, and to contact their embassies when needed (Shivakoti, Henderson, and Withers under review;Elias 2013, 395).Cambodia banned its citizens from migrating to Malaysia in 2011, and the two countries signed an MOU in 2015 aimed at strengthening worker protections (Holliday 2012;Channyda and Cuddy 2015).Indonesia banned its citizens from taking up domestic work in Saudi Arabia in 2011; an MOU was signed in 2014 (Osman 2014), and then again in 2018, to outline a wide range of worker protections to which migrant workers would be entitled as a result (Arab News 2018).
Evidence of the effectiveness of such MOU's on achieving better worker protection is difficult to adjudicate.On the one hand, the fact of MOU signings gives some attention to the importance of workplace protections, and on the other, reports of abuse persist even after they are signed and migration reopens (Holliday 2012;Channyda and Cuddy 2015).In other words, the signing of MOU's does not reliably translate into rights protection, especially if the destination state does not commit to adopting appropriate enforcement mechanisms but they do bring attention to key issues associated with worker protections.Were it the case that such bans were never successful in achieving a desired outcome, then it would be clear that on balance, the restrictions on the right to exit would not be justified.On the contrary, they would be impermissible, not only because rights restrictions are permissible only with adequate justification, but also because in this case they operate in practice to pressure women into irregular migration routes which are even more dangerous.However, even though they limit women's right to exit, because of their occasional success in securing better outcomes for women migrant workers, they cannot be rejected outright as impermissible.More assessment of the impact of signed MOUs is needed to form a firm view, here.

Conclusion
In their contribution to this working paper series, Rainer Bauböck and Martin Ruhs propose a multi-pronged, global strategy for protecting the rights of temporary labour migrants.They rightly argue that it would be preferable to build legitimate concerns about the protection of vulnerable migrants into TLMPs directly, by negotiating the terms in a way that fairly represents the interests not only of host states, but also sending states and migrants.It goes without saying that if such an effort were successful, then the emigrations and conditions I consider here would be neither necessary nor legitimate mechanisms by which to protect migrant workers, even if there are contexts where they have been effective so far.However, that is not yet the case, and so in this article, my objective has been to examine the normative merits of emigration bans and conditions on their own terms, first and foremost, and second, to assess how an assessment of their merits is affected by evidence of their limited effectiveness at achieving their objectives.Emigration bans sacrifice the right to exit of some citizens, even if only for a limited period of time and for certain destinations, in an effort to protect them from abuse, and to gain concessions from host states, in the form of a greater commitment to worker protections.Emigration conditions do not deny exit rights for specific destinations, but constrain the exercise of such rights, sometimes to the point of making exit de facto impossible for certain groups, most often women.
My examination suggests that states do have good reasons to adopt emigration bans and conditions, even where their imposition is explicitly gendered in ways that restrict the movement of women specifically.These reasons are fundamentally protective, in the sense that they aim at improving working conditions for vulnerable migrants, who are too often subject to violence and abuse at the hands of their employers.Domestically, states adopt policies that aim at protecting citizens from violence and abuse, and doing so is typically agreed to be justified on the grounds that such protection is the state's job.In the case considered here, a worry is that the protective policies adopted appear to violate the right to exit as it is described in international human rights law, which as it is articulated does not permit its violation even for protective reasons.But, as I have attempted to show, there may be reasons to permit the violation of the right to exitonly temporarily, and in a limited wayin the case I have considered here.Sending states are generally operating as the less powerful negotiating partner, not only because destination states have other options with respect to from where they recruit migrant workers, but also because the combination of remittances and reduced strain on struggling employment sectors, are of significant benefit to sending states.In many cases, moreover, in addition to violating the right of citizens to exit, migration bans harm sending states economically; they can therefore reasonably be interpreted as pursued only as a last resort.
However, on balance, the harms that are generated by emigration bans and conditions, which leave women with fewer opportunities to earn a livelihood, and which often drive them to migrate irregularly with even less protection, press against adopting them overall.It is for these reasons that the International Labour Organization, too, recommends against them (Napier-Moore 2017).Were evidence of their effectiveness proffered, however, my normative assessment here would shift.That is to say, if evidence were proffered that emigration bans operate to create incentives for host states to do better at protecting migrant workers' rights, then my view would be thatgiven that they are limited restrictions on the right to exit and given the limited options that sending states have for protecting their citizens as they labour abroad, they are worthwhile, and normatively defensible, policy tools to adopt.