There are now more than 60 million displaced people in the world. Of these nearly 20 million are refugees or asylum seekers, whose prospects are governed by the United Nations Convention on Refugees of 1951, designed in the aftermath of WWII to protect people unable to return to their country of origin. Refugees are distinguished from other groups of people, such as so-called economic migrants, who seek entry to another state and refugee status is determined by an investigative process—it is not an assumed status of people seeking asylum.
Most arguments about refugees start from humanitarian premises. Advocates for refugees appeal to the moral obligation to protect the rights or welfare of desperately needy people who are fleeing persecution. Most discussions of refugees end in dismay and frustration, observing how humanitarian imperatives seem to weigh so little in the deliberations of states that ultimately determine policies towards refugees.
Indeed, as Matthew Gibney has noted, "A kind of schizophrenia seems to pervade Western responses to asylum seekers and refugees; great importance is attached to the principle of asylum but enormous efforts are made to ensure that refugees (and others with less pressing claims) never reach the territory of the state where they could receive its protection."1 We take it for granted that politicians will continue to discount the humanitarian imperative of addressing the needs of refugees relative to the interests of their political constituencies. States will continue to guard jealously their sovereignty, so they are unlikely to create a new and improved United Nations Convention on Refugees, or support the emergence of a well-functioning global authority that binds states to take a fair share of refugees. Moreover, instances of individual states voluntarily absorbing large number of refugees, as Germany did in 2016, will remain few and far between. Are there feasible ways to improve the treatment of refugees? We shall argue that there are: progress is indeed possible. The schizophrenia that Gibney refers to stems in large measure, we argue, from perverse incentives that states face under the current system. There is scope for reforms that could benefit refugees and states that are signatory to the Convention. These reforms would ensure more efficient use of the resources provided by states to address the plight of refugees. These reforms could therefore address the plight of refugees more consistently and effectively, without requiring substantial additional sacrifices by states.
We propose that refugees should acquire citizenship in a virtual state constituted by recognized refugee status. That citizenship would entitle them to a package of economic and political rights supported by contributions from the international community. Refugees would be supported through services provided by a host state, but it would not be that host state, but the international community more generally, that underwrites the cost of this provision. Our proposal is to delink three stages in the process by which durable solutions are found for refugees, beginning with processing to determine status, continuing to hosting while durable solutions are determined, and finally to provision of durable solutions. The proposal assumes that states would continue to be willing to make financial contributions to the cause of refugees, but they would retain their rights to determine how many refugees to host and how many to resettle. States could come to view this proposal as ultimately in their interests as well as responding more credibly to humanitarian imperatives. This proposal is not a panacea that could guarantee the timely repatriation or settlement of all refugees. But it holds promise for providing more consistent and humane treatment of refugees, particularly for the often protracted periods in which they seek asylum and then await longer term solutions.
Why is the current system inefficient? Consider first the currently recognized legal obligations towards refugees. According to the Convention (Art. 33 (1), the principle obligation of states granting asylum to those determined to have refugee status, is non-refoulement. This is a negative obligation of states not to return the refugee to the country that she has fled, fearing persecution. Acquiring refugee status does not, however, trigger a positive duty on the part of the state in which refugee status is granted to provide them with a pathway to permanent residency or citizenship, nor does it create such an obligation for any other signatory. For it to be otherwise, states would have to cede some of their sovereign discretion in determining who can remain and settle on their territory. The overall aim of the refugee system is of course to provide longer term protections (commonly referred to as 'durable solutions') for those seeking asylum, but no particular state has an obligation to ensure a durable solution for any particular refugee. Durable solutions include repatriation, integration in the country in which refugee status is initially granted, or resettlement in a third state. Nothing in the Convention specifies which of these solutions must be adopted, or the specific responsibilities of the states in which asylum is initially granted. Refugee integration and resettlement are entirely voluntary schemes, and for reasons noted above, this seems unlikely to change.
Repatriation is often infeasible. Repatriation will only be a solution, rather than a violation of non-refoulement, when conditions in a refugee's home country have changed sufficiently that their rights will be protected, rather than violated, by their government. Despite the challenges involved in repatriation, an overlooked fact is that a majority of refugees (90 percent) have ultimately returned to their country of origin in recent years.2
What about resettlement? Recent statistics show that the UNHCR recommended 134,000 cases for resettlement in 2016.3 Even assuming only 10 percent of refugees seek resettlement, given the repatriation option, the pathway is long and complicated and only a small minority will be resettled quickly.
Integration is not a solution on which refugees can rely. Most refugees are in poor countries close to the conflicts from which they fled. Often the countries where refugees arrive are also riven by political conflict and cannot secure the lives of their own citizens. Ethiopia has 736,000 refugees, Kenya and Uganda over 500,000, Lebanon over 1 million, and Turkey now has 2.5 million refugees resulting from conflict in the Middle East. The capacity of such countries, which may themselves be poorly resourced or unstable, to integrate refugees or even secure their welfare is limited.
Refugees who seek asylum in wealthier countries (typically far from the country of the refugees' origin) are often supported under a variety of special visa conditions whose aim is to secure the minimum welfare of refugees while blocking a path to citizenship in the host country. Australia, for example, processes some classes of refugees offshore in purpose- built offshore detention centers because onshore processing creates a pathway to residency and citizenship, which is perceived as a burden. These arrangements are economically costly, have proven poor at protecting the basic rights of refugees, and have become politically pricey as well. Australia spends $400,000 per year per refugee detaining refugees offshore to avoid their integration, repatriation, or resettlement in Australia.
The 1951 Refugee Convention mandates that signatory states grant travel and work rights to those they recognize as refugees,4 and measures like offshore processing are in effect attempts to circumvent compliance with these obligations. Consequently, many millions of people across the world live in limbo, unable to return to their home country or to migrate to a new country unless their application for resettlement is accepted by a host country.
The principle of non-refoulement is an essential protection. But because this strict obligation is combined with the unspecified obligations of states to find durable solutions along with the requirement that they will provide travel and work rights to those on their territory classified as refugees, the Convention creates perverse incentives. Countries know that, if they process refugees, they will have an obligation not to return them, yet no one else will have a clear obligation to help either support them where they are granted asylum, or offer them other forms of support elsewhere. Consequently, states tend to devise ad hoc policies which balance the minimal obligations of the convention against local political and economic pressures.
For example, most refugees in Australia are entitled to the same benefits as permanent residents.5 This is a relatively substantial allocation, and refugees who settle in Australia have good prospects for robust protection of their rights. But there is also great inefficiency in the Australian system. This derives from the offshore processing system which is designed to deter refugees from attempting to arrive in Australia in the first place. As a method of protecting refugee welfare this system is spectacularly inefficient. At present, the total cost of offshore refugee processing is 1 billion per year, a fraction of which is dedicated to refugee welfare.6 Much is devoted to supporting administration and infrastructure. In 2014, there were 2432 people detained offshore awaiting processing. On average more than 60 percent of people held offshore are determined to be refugees.
In contrast, the UNHCR's total spending on refugees in South East Asia in 2015 was US$157 million allocated for the support of 200,000 refugees and another 2 million asylum seekers and displaced persons. Unlike Australia's Nauru detention center, most of this money goes directly to support the welfare of refugees. For example of the 37 million spent in Thailand on refugees, 36 million is estimated to be allocated to direct welfare support.7
The current patchwork of policies might appear to be a "best worst" solution but there is an alternative which does not require states to sacrifice sovereignty over borders or citizenship or require them to change their visa and immigration rules. It should be clear, when we look at the total amount of money spent across UNHCR government and NGOs per capita on refugees' welfare, that it is not an efficient system. If the burden of supporting refugees is shared more efficiently host countries might reduce or remove some of the restrictions on refugees' entitlements. At the very least it would reduce the incentive to quarantine asylum seekers and refugees and speed up processing times.
How would this proposal work? Asylum seekers could be quickly processed and, if deemed to have refugee status, be provided with a refugee passport. This passport would guarantee them economic support and a range of rights in the countries that host them.
It must be made clear that the countries in which refugees are processed need not be the countries in which they are hosted, and it must also not be assumed that the countries that host people on refugee passports are expected to provide them with a pathway to citizenship, though they may of course do so8. This removes the strong incentive of countries to prevent refugees from reaching their territory. Second, the cost of the system must be underwritten through an equitable system of burden sharing. Hosting refugees would remain voluntary, and to increase the numbers of refugees that a country would accept would require severing the connection between hosting refugees and underwriting the economic support that they may need. For the system to work, the level of support for refugees would need to be such that they were perceived an economic asset rather than a perceived burden. If the investment refugees currently attract was provided in a way that allowed them to spend it on housing, medicine, education, and other services in the host country, hosting would become a more attractive proposition to a broader range of countries.
Under this proposal, refugees could apply for resettlement in any country while being supported in an initial host country (which, as indicated, need not be the country of arrival). For example, Sudanese refugees could be processed in Kenya and hosted in Cambodia, while applying for resettlement anywhere in the world.
The cost of supporting this program would be broadly shared by all signatories to the convention. The details would have to be worked out, and there are a variety of forms it might take, but obligations would be based roughly on their ability to pay.
Countries would retain their sovereign rights to determine the number of refugees they are willing to host temporarily or integrate more permanently. The flipside of this is that the system would only work if enough states would find it in their interest to support it. We've already noted that many states would be more likely to host refugees if the potential costs for doing so were offset. Today very few middle-income countries host refugees. But the resources for this would have to come from somewhere, so the question then is why would wealthy states pay into a system that provides support to refugees in other countries? A few points are worth raising. First, we've noted that many countries already spend a great deal on refugees and the current system is also politically costly. Countries like Australia routinely receive sharp criticism for their treatment of asylum seekers.9 Moreover, under the proposed scheme, there would be greater consistency and regularity. This would limit the non-economic reputational and political costs of the status quo. And of course there are moral reasons to care about offering protection for refugees. Sharing equitably in a scheme that offers protection to a larger number of refugees who are distributed more evenly throughout the world would enable states to uphold their commitment to the principle of asylum without fear that they would consequently bear enormous financial burdens.10
Signatories to the convention (and other willing parties) would contribute to a fund whose aim is twofold: to assure a basic level of welfare and to secure, where possible, for refugees the same non-migratory rights (and obligations such as taxation) as citizens in countries in which they reside. Where possible the two goals can be jointly realised. The basic aim is to allow refugees to live in and contribute to a community in which they reside rather than being quarantined. This again seems consonant with the aims of international refugee law.11
Of course some forms of detention are initially necessary: people seeking asylum after a crisis need emergency housing and support and need to have their status determined, but the problem created by prevailing arrangements is a lack of urgency in processing their claims and finding durable solutions. Host countries use emergency solutions as a form of indefinite quarantine to deny access to the community and/or a path to citizenship. For example the Daadab camps in Kenya function as a way to quarantine Sudanese refugees. If the costs of such access are mitigated and separated from the pathway to citizenship, countries might be less reluctant to temporarily host refugees.
Granting refugees with this package of rights would provide a strong incentive to countries to remove them from physical economic and social quarantine and place them as near as possible on the same footing as local citizens while awaiting a durable solution.
We foresee objections to this idea from wealthier countries who see themselves as magnets for refugees and are worried about the economic and political costs. This proposal, however, should not increase costs. If it increases refugee intake above the quota provided for in a country's visa program then costs are met. However, given differing marginal cost of refugees' support and the preference of most refugees to stay close to home (Syrian refugees in Turkey or Lebanon, for example) it is more likely that most of the costs of maintaining refuges will arise in countries where marginal costs of supporting refugees are much lower.
Another more troubling worry is that this proposal in effect may create a class of people living indefinitely on a refugee passport unable to repatriate or resettle. Here we need to recognize, however, that refugees typically live in conditions far worse than would be available under the scheme. The UNHCR defines a protracted refugee situation as "one in which 25,000 or more refugees from the same nationality have been in exile for five or more years in a given asylum country". Even given this restricted definition the scope of the problem is large. In 2015 6.7 million refugees (41 percent of those under UNHCR's mandate) were in a protracted situation by the end of the year. These refugees were living in 27 host countries, constituting 32 protracted situations. The UNHCR focuses on situations, not individuals. Noting that "the average duration of the 32 protracted refugee situations at the end of 2015 is estimated at about 26 years, most of these (23) have lasted for more than 20 years.12
So while this proposal would not provide a guarantee that refugees would be resettled or repatriated in a timely fashion, and would not be able to guard against the prospect of a refugee being without a durable solution for a long period of time, it would prevent protracted situations in which large numbers of people persist in very poor living conditions. This stands as a significant improvement, and one that may prove consistent with the interests and ambitions of countries throughout the world.
1Matthew J. Gibney, The Ethics and Politics of Asylum: Liberal Democracy and the Response to Refugees, Cambridge University Press, Cambridge, 2004, p.2
2See Megan Bradley, 'Rethinking refugeehood: statelessness, repatriation, and refugee agency', Review of International Studies, 2014, 40(1), pp. 101-123.
4Refugee Convention, art. 17(1)
5For a summary of the welfare burden created by welfare refugees in Australia is here. http://theconversation.com/factcheck-qanda-do-refugees-cost-australia-100m-a-year-in-welfare-with-an-unemployment-rate-of-97-54395.
6See the report by the Refugee Council of Australia. 'At what cost? The true cost of Australia's refugee policies', http://www.refugeecouncil.org.au/getfacts/seekingsafety/asylum/offshore-processing/at-what-cost-unicef-save/
8This does not seem at odds with international law regarding refugees. See James C. Hathaway, The Rights of Refugees under International Law, Cambridge University Press, Cambridge, 2005.
10We are not the first to stress the importance of burden sharing in meeting the claims of refugees. See Peter Schuck, "Refugee burden-sharing: A modest proposal," Yale Journal of International Law 22 (1997), pp. 243-297, and James C. Hathaway, and R. Alexander Neves, "Making International Refugee Law Relevant Again: A Proposal for Collectivized and Solution Oriented Protection," Harvard Human Rights Journal 10 (1997), pp. 115-212, for alternative proposals regarding the form this might take.
11See Hathaway, The Rights of Refugees under International Law, esp. chapter 6.
12UNHCR Report. Global Trends. Forced Displacement in 2015. p 20