The Global Cop-Out on Refugees

During the drafting of the 1951 Refugee Convention, a non-governmental observer – clearly frustrated by the difficulty of securing firm commitments to protect refugees – commented that:

decisions had at times given the impression that it was a conference for the protection of helpless sovereign states against the wicked refugee. The draft Convention had at times been in danger of appearing to the refugee like the menu at an expensive restaurant, with every course crossed out except, perhaps, the soup, and a footnote to the effect that even the soup might not be served in certain circumstances.¹

Despite Mr Rees’ pessimistic assessment, two of three key elements of a binding and powerful commitment to refugees were ultimately secured in the Refugee Convention. First, States agreed to a common definition of refugee status, which has largely withstood the test of time. Secondly, and equally importantly, they committed themselves to what remains an extraordinary catalogue of refugee rights – sensibly oriented to the economic empowerment of refugees, yet flexible enough to take real account of the circumstances of the States to which they flee. The major failing of the Convention, however, was the absence of agreement on a third key element: a common operational mechanism, in particular one that would ensure that protection burdens and responsibilities are fairly shared among States.²

Mr Rees’ unhappy restaurant menu metaphor would actually be more apt to describe the recently completed effort to respond to the missing third (operational) pillar of the Convention: the Global Compact on Refugees (Refugee Compact)³ and its companion Comprehensive Refugee Response Framework (CRRF).⁴ Despite the grand objective of moving beyond particularized duties to ‘provide a basis for predictable and equitable burden- and responsibility-sharing’⁵ among States, what we’ve been offered is very much a menu of possibly wonderful courses (we’re not sure, however, since the descriptions are vague). Indeed, this is not really a menu so much as an indication of items that might (or might not) be available on a given day. In fact, this is not really even a (quasi-) menu for a restaurant; it’s more about what might be offered in a special function dining hall that will only open if a truly large group of hungry people arrives (although we’re not sure how many have to show up before the chef and serving staff will come in to work). In short, this is not the menu for a restaurant that you’d want to count on when making plans to dine.

THE CHALLENGE

Persisting with the status quo ad hoc, State-by-State approach to implementing refugee protection obligations is not an option.⁶ Too often, refugees are forced to risk their lives in order to save their lives, with increasingly sophisticated barriers to access forcing them to undertake risky voyages and to rely on smugglers and even traffickers to reach safety.⁷ The safety on offer is in any event too often illusory. While long-term detention in camps is less common than in the past, it is still the reality for roughly a third of the world’s refugees,⁸ with most others left to struggle in urban slums with no real access to the rights that the Refugee Convention in principle requires.⁹ Worse still, more than 13 million refugees – two-thirds of the total number of refugees – have been waiting an average of two decades for a durable solution, with none in sight.¹⁰ Of these, fewer than one per cent are resettled in any given year.¹¹ In the result, just 10 – mostly very poor – countries now host more than 60 per cent of the world’s refugees,¹² with the entire developed world taking in only 15 per cent of those in need of asylum.¹³ And yet those same rich countries spend at least US$20 billion each year to fund their refugee reception efforts,¹⁴ more than four times the amount the United Nations (UN) refugee agency has available to meet the needs of the 85 per cent of refugees in poor countries.¹⁵ The result is a protection regime that is risky, chaotic, and debilitating, with resources grossly misallocated relative to needs, and which does not provide durable solutions for most refugees. If ever there were a case for a dramatic and fundamental reform, it is surely the current mess of a global refugee system.

Yet, the answer offered by UNHCR under the Refugee Compact process is decidedly ‘thin’. Rather than proposing, for example, a binding optional protocol to remedy the operational deficiencies of the Refugee Convention,¹⁶ the refugee agency has instead drafted a highly partial Compact, applying to undefined ‘large’ movements of refugees.¹⁷ If a situation is so defined, the only thing promised by the Compact is that its principles will serve as guideposts for a never-ending series of discussions.¹⁸ We will reinvent the proverbial wheel each and every time there is a ‘large’ movement of refugees, since every situation will require a new agreement,¹⁹ which will only ‘normally’²⁰ be based on the listed, incredibly vague principles. In truth, the clearest output of the Compact is that there will be lots and lots of meetings to chat about how best to respond to ‘large’ refugee movements: we will have a periodic Global Refugee Forum;²¹ high-level officials’ meetings between forums;²² meetings of national steering groups;²³ Support Platforms;²⁴ solidarity conferences;²⁵ and regional consultative mechanisms.²⁶ And these will be supported by ‘a multi-stakeholder and partnership approach’²⁷ comprising consultations with refugees and host communities, humanitarian and development actors, the UN system, local actors, networks of cities and municipalities, parliaments, faith-based actors, public–private partners, sports and cultural organizers, and yes – even a global academic network!²⁸ The Compact, in other words, is all about process – a bureaucrat’s dream perhaps, but nothing that comes even close to dependably addressing the operational deficits of the refugee regime.

Why were we presented with such a tepid response when the need for decisive action is so clear? Why a ‘thin’ approach to protection reform when something robust is so obviously what both refugees and the poorer States that receive most of them need and deserve?

UNHCR defends its minimalist effort on the grounds that the current political environment is simply not receptive to big picture reform.²⁹ It makes more sense, the agency suggests, to consolidate traditional standards in tandem with a voluntarist framework³⁰ that will at least get States talking about burden and responsibility sharing. To strive for more would be to risk complete failure.

In fairness to the agency, there is at least some academic support for its cautious approach. Michael Ignatieff, for example, has recently advocated a ‘thin’ approach to global justice on the grounds that there is no global normative consensus on what he terms the ‘one world perspective’:

As politics … the one-world perspective is not making much headway. States are no closer to a morally rational way of allocating responsibility for action on climate change. Countries still impose immigration quotas …

Citizens in most democracies believe that their own interests, democratically chosen, ought to prevail over the interests of people in other countries …

The consensus has to be thin because it has to be pluralist …³¹

Perhaps even more ominously, Harlan Cohen has argued that there is a life cycle to multilateralism, and that we may have reached the point at which the post-Second World War multilateral initiatives simply are less valued than they once were because they provide too little by way of concrete deliverables at the national level. This lack of enthusiasm for grand multilateralism may argue for more modest global undertakings:

Providing global public goods in a multipolar world may require smaller deals that can guarantee States specific, desirable club goods. International justice, for example, might have to move to the local level, take local interests more directly into account, and promise specific realizable benefits to the countries in question.³²

These views seem on first glance to align with UNHCR’s ‘thin’ reform agenda. But on a closer reading, there are caveats embedded within these general calls for a thinner version of multilateral human rights protection that actually explain why we should not rush to embrace a ‘thin’ approach to reform of the refugee protection regime.

Ignatieff’s reasoning, for example, is predicated on his view that what individuals share is not a language of the good or a global ethic, but rather a common desire in their specific local vernacular for a framework of expectations that makes their lives meaningful – what he calls ‘ordinary virtues’.³³ When we make moral decisions, we reason based on our sense of a relationship to the other – in Hannah Arendt’s words, ‘men, not Man, live on the earth and inhabit the world’.³⁴ So in speaking about asylum, Ignatieff writes:

From an international rights perspective, provided a stranger meets the criteria for protection set down in international law, there is no upward limit to the number of people citizens are required to receive into their community. From an ordinary-virtue perspective, this idea removes from a political community its very sovereignty … From the ordinary-virtue perspective, the claims of the citizen must trump the claims of the stranger or democratic self-determination has no meaning.³⁵

This is, of course, a view that is very confronting to those of us steeped in the language of universal human rights. But it is important to note where Ignatieff’s argument does not take us. He does not say that asylum cannot work under even an ordinary-virtues optic, but rather that what cannot work is the status quo approach to asylum: what is unworkable is the fact that it imposes unlimited and one-sided obligations on a given community based upon the simple fact of arrival. If, as Ignatieff argues, the most sensible role for international norms is to nudge ordinary virtues to enlarge the circle of moral concern,³⁶ that requires real, affirmative action on our part to defuse the sense that the provision of asylum is inattentive to the concerns of receiving communities – to engage and answer the perception that there is ‘no upper limit’ to the duty to provide asylum to those who arrive. And that is a challenge that we can and should meet.

Similarly, when Cohen questions the continuing appeal of multilateralism, he also insists that this is especially so when multilateralism ignores the importance of ensuring real benefits at the local level. His core argument is that ‘[w]hen material or political circumstances change and the costs of the agreement rise, the perceived benefits of staying in the agreement may be too low to keep a state in’.³⁷ Again, this is a challenge that we can and should meet in a practical and dependable way.

My point, then, is that even if you take the view – as UNHCR appears to do – that it is unwise simply to assume a strong commitment to global human rights or to multilateralism in general, then you should be advocating exactly the opposite of what the agency has proposed. If you believe that the commitment to global human rights is in retreat, then the answer is not a ‘thin’ version of protection under which we simply pay lip service to burden and responsibility sharing by setting up an endless loop of conversations. The answer is instead clearly and firmly to show how a dependable, managed model of sharing could meet the needs of all. And if you believe that the commitment to multilateralism is in decline, then the answer is not to ‘go thin’, but rather to show how a multilateral reform can be made to deliver at the local level – providing real benefits for real people in real communities that receive refugees. In short, the last thing we should be doing is proposing – as UNHCR does – an endless procession of voluntarist pledging conferences that may, or may not, deliver.

More generally, it is wrong to treat this thinnest of ‘thin’ reforms as a grandly titled ‘Global Compact on Refugees’ since it actually aspires to be only an ‘add-on’ for undefined ‘large’ movements of refugees – meaning that the current broken, yet massively expensive, system stays in place. By positing no shift away from the status quo, where most of the resources are spent to address the claims of a tiny number of refugees able to reach rich countries, UNHCR’s tack robs reform of the possibility of harvesting desperately needed resources that could provide a better system for all refugees, not just ‘large’ groups of refugees.

Put simply, if we are worried about the state of refugee protection in the world – and we should be – the best defence is a good offence. Make protection real, robust, and reliable – not ‘thin’ – by offering a vision that not only delivers more and more dependably for refugees, but which ends, once and for all, the ‘accidents of geography’ approach to the allocation of burdens and responsibilities, and is fundamentally and deeply embedded in delivering real benefits to real host communities. We cannot succeed by hunkering down and pretending that the challenges to the vitality of human rights and multilateralism can be papered over by adopting vague guiding principles coupled with never-ending talkfests.

A ROBUST ALTERNATIVE

So what would a courageous, ‘best defence is a good offence’, robust – not ‘thin’ – road map to address the current morass in the refugee protection system look like?

The first and most critical priority – ironically not even addressed in the Global Compact – is access to protection. While we ought to promote assisted entry wherever that is feasible, the non-negotiable baseline commitment must be that refugees be allowed to access the international protection system in whatever country they can reach. No more barriers to entry, no more politics of non-entrée.³⁸ While this may sound like ‘pie in the sky’, I believe it is a goal that can be attained if the place of a refugee’s arrival were to be only that – the point of entry into the international refugee system, with no domestic immigration consequences for that State. After all, why spend billions of dollars every year to deter something that has no consequences for you?

This links to a second plank of a robust model for reform. Upon arrival, internationally run, normally group-based refugee status assessment – quick and simple in the overwhelming majority of cases – would take place.³⁹ An international corps of decision makers would be dispatched to identify genuine refugees. The UN would then allocate and move refugees to receive ‘protection for the duration of risk’, normally in their region of origin in order to maximize functional and cultural compatibility, which are critical to refugees getting back on their feet, in at least the short-term. Against the backdrop of quotas for refugee protection, both destination State and refugee preferences would be factored into the assignment system, relying on sophisticated algorithms to generate speedy matches. A wonderful side benefit of such a system is that the illegitimate part of the smuggling market would simply dry up, as smugglers and traffickers would no longer be able to offer an immigration outcome for their services.

Thirdly, while refugees might be briefly detained – which is not precluded by the Refugee Convention⁴⁰ – during the short time required to verify their status in the country of first arrival, there would be no constraints on freedom of movement once they arrived in their assigned State of protection for the duration of risk – meaning no more closed camps. Rather, precisely in line with what the Convention requires,⁴¹ refugees would be allowed to get on with their lives: to set up businesses, work, educate their children. Social science research is clear that when freed to contribute, refugees are not only not burdens, but can actually be dynamic engines for development in their host communities.⁴²

Fourthly, we need to make asylum ‘doable’ for poorer States⁴³ – who would likely, under the proposed model, continue to offer the lion’s share of first-phase protection for the duration of risk. The current system of forcing poor countries to hope for charity to fund the costs of processing and receiving refugees must, and would, end.⁴⁴ Instead, the required funds would be guaranteed by the international agency – with funds raised under a common burden-sharing formula – contingent on respect for refugee rights by the receiving country.⁴⁵ At least as important, the system I propose would also provide economic start-up grants to refugees and to the communities that receive them – linking refugees to those host communities, and helping to generate the goodwill needed to make the system sustainable at the local level.⁴⁶

Fifthly, and most importantly, this new system would guarantee a true solution to refugeehood. On average, about a quarter of refugees are able to return to their home country after no more than five years abroad.⁴⁷ For those unable to go home, a system of truly empowering protection that links refugees to their host communities would make local integration a dramatically more viable option – providing an answer for perhaps another quarter.⁴⁸ But if neither repatriation nor local integration is viable at the five-year switch point,⁴⁹ the system I advocate would guarantee the remaining half of refugees access to resettlement – no more protracted refugee situations with lives on hold for 20 years or more.⁵⁰This critical job – now neglected, with only the paltry 100,000 or so resettlement places a year presently offered⁵¹ – would become the core job of States outside regions of origin. We would need about 1.7 million resettlement spots per annum – almost identical to the 1.65 million persons now arriving to make asylum claims in OECD countries.⁵² The main responsibility-sharing task of these extra-regional countries would therefore change from offering asylum to providing resettlement – a common, but differentiated, responsibility. A focus on resettlement commitments would allow the developed world to protect refugees in an orderly, managed way, even as it would meet a critical gap in the current system, which leaves millions of refugees in permanent purgatory.

A five-step plan of this kind would require no amendment of the Refugee Convention; it could be implemented via an optional protocol that would come into force as soon as a critical mass of 20 or 30 States were on board. The insurance-based, collectivized model of implementation would likely broaden the base for protection, as States that have stayed outside the Convention system would come inside, confident that they could do so without the risk of being overwhelmed – indeed, with the guarantee of financial and human support of other countries. This new mode of implementing the Refugee Convention, including the cost of assessing and relocating refugees to their host countries, and supporting those States, could easily be funded with less than the nearly US$20 billion spent each year by rich countries to run their domestic refugee systems⁵³ – systems that would no longer be required. When the cost of programmes that buy asylum space in Turkey,⁵⁴ in North Africa,⁵⁵ in offshore Pacific island States,⁵⁶ and elsewhere is factored in, the cost savings associated with the shift to a streamlined, common asylum system might well be dramatic.

COULD IT WORK?

So, could we really build this highway to true refugee protection? History shows that we can do everything I am proposing.

Starting in the 1920s, we let an international agency (the Nansen International Office for Refugees) make assessment of who qualified as a refugee, which States agreed to rely upon.⁵⁷ After the Second World War, the International Refugee Organization regularly distributed refugees from places of arrival to new homes where they could remake their lives – with its own fleet of ships, no less.⁵⁸ During the exodus of Vietnamese boat people in the 1970s, we set up an effective and efficient system to link States in the region with States far away – in which different States did different jobs to be sure that all refugees obtained protection somewhere.⁵⁹ That success was replicated to some extent in the ICARA⁶⁰ and CIREFCA⁶¹ systems in Africa and Latin America respectively.⁶² And most recently of all, just two years ago, Canadian officials did what they had once thought impossible: they retooled their traditional resettlement model so that 25,000 refugees could be resettled in Canada in just a few short months under a system that sacrificed nothing on security vetting, and which linked refugees to real people in their new communities to ensure acculturation and acceptance.⁶³ So, yes – while one would of course wish to pilot a common system sub-regionally and regionally before implementing a protocol at the global level, there is sound empirical evidence to support its practical viability.

The system I am proposing would provide a better place for everyone.

First and foremost, it would be better for refugees. Refugees would not have to put their lives on the line to get access to solid protection, since they would access the same protection regime with the same opportunities whether they walked across a land border or travelled for thousands of miles. Protection would really be the empowering, rights-regarding protection that the Refugee Convention calls for – not the purgatory of refugee camps or urban slums. And, most importantly, every refugee would get a durable solution within a reasonable period of time.

This would also be a much better system for the poorer countries that host 85 per cent of the world’s refugees, which would no longer need to beg for the charity of wealthier States, but would instead be guaranteed the funds needed to protect refugees. Those protection funds would go not just to refugees, but would also fund start-ups linking refugees to their host communities, so that everyone would benefit from the refugees’ presence. And regional host States would not face the possibility of indefinite hosting as a perverse punishment for keeping their doors open to refugees. To the contrary, refugees unable safely to go home or for whom local integration is not possible would be resettled to another State at the five-year cut-off point.

And, yes, the system would also be better for the developed world. The assignment mechanism would undermine the smuggling market and reduce the use of the refugee channel for economic migration. Industrialized States would have the time they need to vet security concerns before refugees were admitted into their territory under the resettlement scheme. And the focus on protection by resettlement would give wealthier countries a role that is an easier social ‘fit’ than is provisional or temporary presence.

In short, the shift I am proposing would be a win–win–win solution.

THE CHALLENGES

I am not, of course, suggesting that this is a ready-to-roll-out, easy-to-achieve option. It’s not. But then again, truly worthwhile undertakings rarely are. There are a number of practical and ethical challenges, but I believe these can be met.

First, there would need to be agreement on the formulae that would define both a fair sharing of (financial) burdens and (human) responsibilities which would ground the proposed system of common, but differentiated, responsibility. Building on rudimentary efforts by Grahl-Madsen⁶⁴ and the European Union,⁶⁵ Oxfam’s Sarnata Reynolds has recently modelled what is perhaps the most sophisticated and detailed option, merging consideration of GDP with rank on both the Human Development Index and the Fragile States Index.⁶⁶ While in contrast to Reynolds’ model I believe that different factors ought to define a State’s fair share for purposes of financial burden sharing and human responsibility sharing,⁶⁷ the point is that there are already viable starting points for discussion.

Secondly, assuming agreement on the formulae, there needs to be a strong central actor able and willing to administer the quota and assignment programme. But who would this be? Even if States could be persuaded that there is little, if any, benefit to State-by-State management of what is fundamentally a transnational concern, the UN’s refugee agency (UNHCR) has to date shown remarkably little interest in assuming such a role. Indeed, Betts and Collier argue that UNHCR is ‘not currently equipped’⁶⁸ to manage a new refugee regime as its institutional strengths ‘are no longer the primary skills needed to ensure refugee protection in the twenty-first century’.⁶⁹ Yet, despite all my concerns about how the agency has evolved, I wonder – particularly with the International Organization for Migration nipping at its heels as the new ‘UN migration agency’ – whether an enlightened High Commissioner might not be persuaded that it actually is in the agency’s long-term interest to refocus on its core mandate and to show its ability to provide a real, practical, and efficient answer to one of the most pressing political challenges of our time. If UNHCR were to offer a real answer that meaningfully and dependably addressed States’ concerns, wouldn’t that be a smart institutional move?

Thirdly, and perhaps most fundamentally, a critical shift of this kind – even one not involving formal amendment of the Refugee Convention – requires a champion. The dearth of political leadership on refugee protection at the present moment is, however, extreme. There have of course been moments when leaders have risen to the challenge of protection: Canada’s Justin Trudeau, Germany’s Angela Merkel, Jordan’s King Abdullah, and Tanzania’s Jakaya Kiwete come to mind. An effort to bridge the leadership gap – a ‘World Refugee Council’ co-chaired by former leaders from Canada, Germany, Pakistan, and Tanzania – seemed to hold promise, but has thus far offered only vague directions to guide reform.⁷⁰ But here, too, success is possible. Is it really too late, for example, to light a fire under the World Refugee Council to persuade it to deliver a bold and principled vision of reform? And with that ammunition, couldn’t a quadrumvirate of Abdullah, Kiwete, Merkel, and Trudeau – and perhaps others – be the torchbearer for a broader political commitment? After all, these leaders have already hitched their political stars to doing right by refugees.

Beyond these practical questions, the key ethical question is whether it is inherently unseemly to establish a system that would essentially distribute refugees – initially to a country in which they would receive protection for the duration of risk and, if residual resettlement is required, to permanent home countries. This critique is usually framed as directed to the ‘commodification’ of refugees – treating human beings as though they were bales of wheat to be shipped around the world.⁷¹ I don’t shy away from this concern, but do invoke utilitarianism in my defence. While it may be suboptimal to structure asylum opportunities, surely this is less bad than the result produced by the current system under which massive resources are now expended on the 15 per cent of refugees able to reach the developed world – disproportionately young, male, and mobile – while comparatively derisory resources are made available to the 85 per cent of refugees who remain closer to home. In particular, the fact that some 13 million refugees in the global South are in protracted refugee situations – long-term indeterminacy, with no solution in sight – argues strongly for the importance of sacrificing some of the relative privilege of refugees able to reach asylum in wealthier States in order to do right by the massive majority of refugees consigned to dramatically more difficult circumstances. Put simply, the equal moral worth of all refugees requires that we pay as much attention to those we don’t see as to those who are in our midst.

Furthermore, the managed system I am proposing need not – indeed, should not – sacrifice a commitment to refugee autonomy. Even if the utilitarian claim holds – as I believe it does – an ethical responsibility-sharing system can and should be structured in a way that maximizes the agency of refugees, albeit within an overarching system of assignment. So long as this argument is made on behalf of all refugees – not just those who make it to the developed world – I think that incorporation of a system for ‘preference matching’ in the assignment system has real merit.⁷² The importance of maximizing refugee agency is also a matter of pragmatism since, as Joris Schapendonk has recently noted, ‘[i]f you deprive refugees completely of choice, we will have a flourishing industry specialized in “reintroducing” agency, consisting of … brokering service, smugglers and traffickers’.⁷³

I do not offer this alternative approach as any sort of perfect plan. But recalling Voltaire’s insistence that ‘the perfect is the enemy of good’,⁷⁴ I think the real question should be whether a proposal for a global and managed system of refugee protection is a better place to focus than what is on offer under the Refugee Compact – a shameful sop that doesn’t eliminate barriers to access; doesn’t dependably get refugees to a place of protection; doesn’t ensure dignified and empowering protection for the duration of risk; doesn’t require meaningful burden and responsibility sharing; and doesn’t guarantee solutions either for refugees or for their host communities.

I think we need to call out this ‘Compact’ for what it really is – a ‘cop-out’. We should be clear that we do not need a Compact ‘on’ refugees, in which refugees are simply the object, not the subject, of the agreement. It is high time for a reform that puts refugees – all refugees, wherever located – first, and which recognizes that keeping a multilateral commitment to refugee rights alive requires not caution, but rather courage.

Distinguished Visiting Professor of International Refugee Law, University of Amsterdam. This article is based on the Keynote Address to the International Association for the Study of Forced Migration (University of Macedonia, Thessaloniki, Greece, July 2018).

Footnotes

1

Statement of Mr Rees of the Standing Council of Voluntary Agencies, UN doc A/CONF/2/SR.19 (26 November 1951) 4–5.

2

The best that could be achieved was a recognition in principle that ‘the grant of asylum may place unduly heavy burdens on certain countries, and that a satisfactory solution … cannot be achieved without international cooperation’: Convention relating to the Status of Refugees (adopted 28 July 1951, entered into force 22 April 1954) 189 UNTS 137 (Refugee Convention) preambular para 4.

3

Global Compact on Refugees, Final Draft (26 June 2018) <http://www.unhcr.org/en-us/events/conferences/5b3295167/official-version-final-draft-global-compact-refugees.html>.

4

Comprehensive Refugee Response Framework, Annex I to the New York Declaration for Refugees and Migrants, UN doc A/RES/71/1 (3 October 2016). ‘Part II of the global compact is the Comprehensive Refugee Response Framework as adopted by the United Nations General Assembly … This constitutes an integral part of the global compact’: Refugee Compact, para 10.

5

Refugee Compact, para 3.

6

The CRRF itself recognizes that ‘[t]he scale and nature of refugee displacement today requires us to act in a comprehensive and predictable manner in large-scale refugee movements’: CRRF, para 1. Why this is so in relation to ‘large-scale movements’ but not otherwise is, however, not explained.

7

T Gammeltoft-Hansen and N Nyberg Sorensen (eds), The Migration Industry and the Commercialization of International Migration (Routledge 2013).

8

United Nations High Commissioner for Refugees (UNHCR), ‘Global Trends: Forced Displacement in 2017’ (2018) (‘UNHCR, ‘Global Trends’) Table 5.

9

Baher Kamal, ‘Now 1 in 2 World’s Refugees Live in Urban Areas’ (Inter Press Service, 19 May 2016), reporting that ‘more than 1 in 2 of all the world’s refugees live in slums or in informal settlements and on the fringes of cities, in overcrowded neighbourhoods and in areas prone to flooding, sanitation hazards and diseases’.

10

UNHCR, ‘Global Trends’ (n 8) 22. Indeed, more than 3 million refugees have been in protracted refugee situations for nearly 40 years. These numbers actually understate the problem, since UNHCR sets a minimum 25,000 refugee threshold for a ‘protracted situation’ – meaning, for example, that Eritreans in Israel and Malians in Burkina Faso were not counted in 2017.

11

A total of 102,800 refugees were resettled in 2017: UNHCR, ‘Global Trends’ (n 8) 8.

12

UNHCR, ‘Global Trends’ (n 8) 18. Among the top 10 receiving countries are four of the least developed States in the world – Bangladesh, Ethiopia, Sudan, and Uganda.

13

UNHCR, ‘Global Trends’ (n 8) 15.

14

This is based on an average cost of US$11,500 per refugee (OECD, ‘Who Bears the Cost of Integrating Refugees’ (January 2017) 13 Migration Policy Debates 1) and an average of 1,650,000 refugee claims made in developed countries in 2015–16 (OECD, ‘Key Statistics on Migration in OECD Countries’ (2018)). This is likely a very conservative estimate since while States aim to process refugee claims in the first year, backlogs and appeals mean that procedures and hence support costs may extend into a second or subsequent year: European Council on Refugees and Exiles, ‘The Length of Asylum Procedures in Europe’ (2016). Indeed, extrapolating from data on costs in Germany, it has been suggested that ‘the world spends approximately $75bn a year on the 10% of refugees who moved to developed regions’: A Betts and P Collier, Refuge: Transforming a Broken Regime (Allen Lane 2017) 129.

15

The agency’s 2018 refugee programme budget for all of Africa, the Middle East and North Africa, Asia and the Pacific, and the Americas totalled US$4.352 billion: UNHCR, ‘Global Focus: UNHCR Operations Worldwide’ <reporting,unhcr.org/financial> accessed 21 September 2018.

16

‘The global compact is not legally binding’: Refugee Compact, para 4.

17

This response ‘will be developed and initiated by the Office of the United Nations High Commissioner for Refugees … for each situation involving large movements of refugees’: CRRF, para 2. ‘[T]he CRRF relates specifically to large refugee situations …’: Refugee Compact, para 12, although such movements ‘are not necessarily homogenous’.

18

Refugee Compact, paras 17–30.

19

CRRF, para 3.

20

ibid para 4.

21

Refugee Compact, paras 17–19.

22

ibid para 19.

23

ibid para 20.

24

ibid para 22.

25

ibid para 27.

26

ibid paras 28–30.

27

ibid para 33.

28

ibid paras 34–44.

29

‘It is important to bear in mind what you put up for discussion at the intergovernmental level with 193 UN member states, because at the end of the day we will need to present a consensus document. It’s a question of strategy … I know that the current state of affairs is such that we wish there would be more resettlement, better family reunification, and more funding. The global compact is an aspiration that will yield better results over time’: V Türk, Assistant High Commissioner for Protection, quoted in Charlotte Alfred, ‘UN Official: Refugee Compact Will Meet Fear and Ignorance with Facts’ (Refugees Deeply, 1 March 2018).

30

‘The global compact … will be operationalized through voluntary contributions to achieve collective outcomes and progress towards its objectives’: Refugee Compact, para 4. The operational language is also vague; for example, ‘States …, as appropriate, would …’ provide support for immediate and ongoing needs: CRRF, para 6. Both the ‘as appropriate’ qualification and conditional ‘would’ (not ‘will’) language make clear that firm commitments are not envisaged.

31

M Ignatieff, The Ordinary Virtues: Moral Order in a Divided World (Harvard University Press 2017) 47.

32

HG Cohen, ‘Multilateralism’s Life Cycle’ (2018) 112 American Journal of International Law 47, 65.

33

Ignatieff (n 31) 202.

34

H Arendt, The Human Condition (2nd edn, University of Chicago Press 1998) 7 (emphasis added).

35

Ignatieff (n 31) 210–11.

36

ibid 214.

37

Cohen (n 32) 65.

38

See generally T Gammeltoft-Hansen and JC Hathaway, ‘Non-Refoulement in a World of Cooperative Deterrence’ (2015) 53 Columbia Journal of Transnational Law 235.

39

The Compact calls only for voluntary contributions to ‘preparation for large refugee movements’ and for an ‘asylum capacity support group’ that would enable ‘standby arrangements and sharing of good practices between States’: Refugee Compact, paras 52, 62. No shift from State-by-State status assessment mechanisms is envisaged: Refugee Compact, para 61.

40

Refugee Convention, art 31(2).

41

See generally JC Hathaway, The Rights of Refugees under International Law (Cambridge University Press 2005); A Zimmermann (ed), The 1951 Convention Relating to the Status of Refugees and Its 1967 Protocol: A Commentary (Oxford University Press 2011).

42

A Betts and others, Refugee Economies: Forced Displacement and Development (Oxford University Press 2017).

43

The Compact persists with the tradition of pure voluntarism, suggesting that ‘[t]he global compact can help attract support to ensure that refugees and their host communities are not left behind …’: Refugee Compact, para 64. More specific promises are similarly vague; eg ‘[d]epending on the context, host countries may seek support from the international community as a whole to address the accommodation and environmental impacts of large numbers of refugees’: Refugee Compact, para 78.

44

The Refugee Compact recognizes that ‘it is also important to support countries [which] elect to resolve a refugee situation locally’, but provides only vaguely that ‘the international community as a whole will, in close cooperation with national authorities of host countries, contribute resources and expertise to assist with the development of a strategic framework for local integration’: Refugee Compact, paras 97, 99.

45

The CRRF disturbingly does not insist on full respect by receiving States for refugee rights, but is content to provide that ‘[h]ost States, bearing in mind their capacities and international legal obligations … where appropriate, … would … [p]rovide legal stay to those seeking and in need of international protection as refugees’: CRRF, para 13 (emphasis added).

46

In contrast, under the CRRF there is only a commitment to ‘[c]onsider establishing development funding mechanisms’ for [receiving] countries: CRRF, para 6(d).

47

The statistical analysis is necessarily approximate. In the timeframe 1998–2007, an average of 1.14 million refugees returned to their home country each year: M Bradley, Refugee Repatriation: Justice, Responsibility and Redress (Cambridge University Press 2013) 3. But over both a longer (1974–2013) and a more recent (2003–13) timeframe, the average number of returns per annum has been in the range of 650,000–700,000: UNHCR, Statistical Yearbook 2013 (2015) 49–50. The average of 675,000 is in fact roughly the number of returns for the most recent year about which we have data (2017), in which 667,400 returns were reported: UNHCR, ‘Global Trends’ (n 8) 8. This amounts to 40% of the number of new asylum applications in 2017 and 15% of the total number of newly displaced refugees that year: at 2.

48

For example, a 2014 US Department of State analysis showed that an empowering approach to protection in Cameroon, Tanzania, and Zambia led to more offers of local integration of refugees by those countries: US Department of State, ‘Evaluation of Local Integration: Final Report. Evaluating the Effectiveness of Humanitarian Engagement and Programming in Promoting Local Integration of Refugees in Zambia, Tanzania, and Cameroon’, doc No SAWMMA13F2592 (2014).

49

The cut-off reflects the point at which the psychosocial need for permanence compels a durable solution: see JC Hathaway and RA Neve, ‘Making International Refugee Law Relevant Again: A Proposal for Collectivized and Solution-Oriented Protection’ (1997) 10 Harvard Human Rights Journal 115, 182–84; M Domanski, ‘Insights from Experience: A Background Paper on Temporary Protection’ in JC Hathaway (ed), Reconceiving International Refugee Law 22 (Brill 1997).

50

In contrast, the Refugee Compact provides only that ‘[c]ontributions [to resettlement] will be sought from States’ and that ‘[a]s a complement to resettlement, other pathways for the admission of persons with international protection needs’ will be promoted: Refugee Compact, paras 91, 94. Suggesting no intent to dramatically expand resettlement opportunities, the CRRF sets as a goal only that ‘States aim to provide resettlement places and other legal pathways on a scale that would enable the annual resettlement needs identified by the Office of the United Nations High Commissioner for Refugees to be met’: CRRF, para 16. UNHCR has, however, never identified annual resettlement needs beyond a very small fraction of refugees in protracted refugee situations.

51

See note 11 above.

52

In 2015–16, 1.65 million asylum claims were made in OECD States: OECD (2017) (n 14). This equates to roughly 50% of the total number of new refugees in that same timeframe (3.4 million per annum): UNHCR, ‘Global Trends: Forced Displacement in 2016’ (2017) 2 – used here as a rough surrogate for the estimated 50% of refugees who would need resettlement each year. The backlog of refugees in protracted situations will of course also require a solution, perhaps staggered over a decade so as not to undermine the efficacy of the reformulated protection system.

53

See note 14 above.

54

The European Union agreed to pay Turkey nearly US$7 billion to keep Syrian and other refugees from arriving in Europe: ‘EU and Turkey Reach Refugee Deal’ (Politico, 20 March 2016).

55

For example, the European Commission has offered nearly US$7,000 per refugee to North African States willing to house them in ‘secure centres in their territory’: ‘EU Offers to Pay Countries €6,000 per Person to Take in Migrants Rescued from Mediterranean’ The Independent (24 July 2018).

56

Australia budgeted more than US$500 million in 2017–18 to operate its offshore processing system: C Higgins and Z Tishler, ‘Factsheet: The Cost of Australia’s Asylum Policy: A Guide to Sources’ (Andrew & Renata Kaldor Centre for International Refugee Law, 2017).

57

See J Hope Simpson, Refugees: Preliminary Report of a Survey (Oxford University Press 1938).

58

See L Holborn, The International Refugee Organization: A Specialized Agency of the United Nations (Oxford University Press 1956).

59

See W Robinson, Terms of Refuge: The Indochinese Exodus and the International Response (Zed Books 1998).

60

See R Gorman, ‘Beyond ICARA II: Implementing Refugee-Related Development Assistance’ (1986) 20 International Migration Review 283.

61

See A Aguilar Zinser, ‘CIREFCA: The Promises and Reality of the International Conference on Central American Refugees’ (Center for Immigration Policy and Refugee Assistance, Georgetown University 1991).

62

See P Mathew and T Harley, Refugees, Regionalism and Responsibility (Edward Elgar Publishing 2016).

63

‘Canada Meets Target to Resettle 25,000 Syrian Refugees’ The Guardian (29 February 2016). The cost of resettlement was about US$12,000 per refugee: L Sevunts, ‘Bringing 25,000 Syrian Refugees to Canada Cost $385M’ (Radio Canada International, 22 November 2016).

64

A Grahl-Madsen, ‘Refugees and Refugee Law in a World of Transition’ [1982] Michigan Yearbook of International Legal Studies 65, 74.

65

See European Commission, ‘European Agenda on Migration’ (13 May 2015), outlining the basis for the relocation mechanism approved by the European Parliament on 17 September 2015 <http://europa.eu/rapid/press-release_STATEMENT-15-5664_en.htm>.

66

S Reynolds, ‘Building a Lifeline: A New Global Platform and Responsibility Sharing Model for the Global Compact on Refugees’ (December 2017) <https://zolberginstitute.org/publication/building-a-lifeline-a-proposed-global-platform-and-responsibility-sharing-model-for-the-global-compact-on-refugees/>.

67

See Hathaway and Neve (n 49) 204–07.

68

Betts and Collier (n 14) 6.

69

ibid 38.

70

World Refugee Council, Transforming the Global Refugee System: Solidarity, Humanity and Accountability, Interim Report (Centre for International Governance Innovation, 17 April 2018).

71

See eg BS Chimni, ‘The Past, Present, and Future of International Law: A Critical Third World Approach’ (2007) 8 Melbourne Journal of International Law 499.

72

See eg J Fernandez-Huertas Moraga and H Rapoport, ‘Tradable Immigration Quotas’, CESifo Working Paper No 4087 (2013); W Jones and A Teytelboym, ‘Choices, Preferences and Priorities in a Matching System for Refugees’ (2016) 51 Forced Migration Review 80; K Bansak and others, ‘Improving Refugee Integration through Data-Driven Algorithmic Assignment’ (2018) 359 Science 325.

73

J Schapendonk, ‘The Geography of a Global Refugee System’ (2018) 9 Asiel & Migrantenrecht 64.

74

Voltaire, Dictionnaire philosophique (first published 1764).