Melbourne Journal of International Law
The Melbourne Journal of International Law was established with the aim of creating a forum to facilitate discussion and debate of international legal issues. International refugee law is one issue that has been the subject of increasing attention in recent years. In particular, questions have been raised regarding the ability of international refugee law to adequately protect the human rights of refugees.
In recognition of the importance of this issue, we have elected to publish the 25th Allen Hope Southey Memorial Lecture: ‘Why Refugee Law Still Matters’. This address was delivered at the University of Melbourne by Professor James C Hathaway, a leading authority on international refugee law. Professor Hathaway’s address challenges traditional notions of asylum and seeks to reconcile the tension between the human rights of refugees and the rights of states to control their borders.
It is our hope that the publication of this feature will inspire reflection on the role of international law as a practical remedy to balance the conflicting interests of states and refugees.
BRONWYN REDDAN, NATASHA SUNG AND ROB WALKER
Editors, Melbourne Journal of International Law
JAMES C HATHAWAY[*]
I am concerned that the singular importance of international refugee law is profoundly misunderstood. My more specific worry is that erroneous and competing claims by governments and the refugee advocacy community about the structure and purpose of refugee law threaten its continuing ability to play a truly unique human rights role at a time when no meaningful alternative is in sight.
In particular, governments of the developed world are now appropriating the language of burden-sharing in order to further an only mildly attenuated global apartheid regime under which most refugees not only remain in the less developed world, but remain there under conditions which are generally rights-abusive and often literally life-threatening. These states have distorted the true object and purpose of the Convention relating to the Status of Refugees (‘Refugee Convention’), erroneously suggesting that it sets only protection obligations of ‘last resort’ — that is, that refugees may be routinely sent away to any other state that will admit them without risk of return to their country of origin. Governments have further stigmatised refugees who arrive without pre-authorisation as ‘illegal’ despite the fact that the Refugee Convention itself requires otherwise. And perhaps most disingenuously, these same governments have sought to justify their harsh — and often illegal — treatment of refugees arriving at their territory on the grounds that such harshness is the necessary means to a more rational protection end, namely the reallocation of resources towards meeting the needs of the overwhelming majority of refugees located in the less developed world, with resettlement opportunities to be made available only to those said to be most acutely in need.
This rhetoric is largely a distortion of international refugee law.
First and most fundamentally, there is no duty whatsoever on a refugee to seek protection either in the first country where he or she arrives, or more generally within his or her region of origin. Nothing in either the Refugee Convention or in the so-called ‘soft law standards’ agreed to by the Member States of the Executive Committee of the United Nations High Commissioner’s Programme (‘UNHCR Executive Committee’) so requires. To the contrary, unless a refugee has already found protection elsewhere — ‘found’ meaning actually acquired (not just ‘could ask for’) and ‘protection’ meaning the whole bundle of protections guaranteed by treaty (not just not being removed — then present standards actually require deference to the wishes of the refugee regarding where to take his or her chances This does not require deference to the wishes of the refugee as to where they are to be admitted. But the essential structure of the Refugee Convention makes clear that even as states were given the power of status determination once reserved to the international oversight agency, refugees were to be afforded the prerogative to decide where to take their chances on status determination
Second, refugees who arrive at a state’s territory — indeed, refugees who come under a state party’s jurisdiction, including parts of the high seas over which a state has taken effective jurisdiction — are entitled to the benefit of the Refugee Convention. Efforts to resist such responsibility on the basis of the alleged illegality of arrival or presence are fundamentally at odds with the Refugee Convention’s requirement that refugees in flight from persecution are to be granted immunity from any penalties for illegal entry or presence. No state has ever been prepared to issue a visa for the purpose of travelling to its territory in order to make an asylum claim. And even if available, it would often be too risky or logistically challenging for a refugee to take advantage of such an option. The Refugee Convention therefore requires all state parties to treat refugees who make a clean breast of their illegal entry or presence as non-transgressors. This, in effect, creates a necessity-based exemption from the usual rules of migration control. It is completely inappropriate to stigmatise refugees arriving without visas as law breakers when a treaty we have freely signed provides exactly the contrary.
Third, it is increasingly claimed that the harshness meted out to the minority of refugees who arrive at the frontiers of developed countries can be justified as part of an overarching strategy to reallocate protection resources from wealthier asylum countries to the regions of the world where most refugees actually live. This strategy is justified not only on the grounds of the relative numbers of refugees in the developed world compared to the less developed world, but more fundamentally on the basis that refugees in the less developed world are — in the words of the Australian government — ‘most in need’. The reallocation of fiscal resources towards the less developed world is, as in the case of Australia, sometimes accompanied by a commitment to resettle some number of ‘those refugees most in need’ from the less developed world. Such a commitment is said by Australia to be possible ‘[d]ue to Australia’s success in dramatically reducing the number of illegal [refugee] arrivals over recent years’. In other words, it is claimed that a greater commitment to resettlement has been made possible by the savings realised from shutting down what the Australian government views as illegal, namely unauthorised, refugee arrivals.
This third component of recent refugee policy is somewhat more challenging to untangle than the first two components. There is certainly no doubt that the burdens and responsibilities of offering protection to refugees are today unfairly apportioned. More than 90 per cent of refugees remain in the less developed world, with some states — Jordan, Lebanon, Syria, Chad, Tanzania, Iran and Sierra Leone — hosting more than one refugee for every 100 citizens. In contrast, Australia’s refugee to citizen ratio is nearly 1:1400, the European Union’s roughly 1:2000, and Japan’s approaches 1:50 000. Not only is the less developed world hosting the overwhelming share of refugees, but it does so with a small fraction of the resources presently allocated to processing and assisting the tiny minority of refugees who reach richer states. In rough terms, less than US$1 per day is available to look after each one of the approximately 4.4 million refugees under direct UNHCR care in poorer states. Furthermore, not even that tiny budget is guaranteed, but has to be garnered each year from voluntary contributions to the UNHCR of a small number of wealthier countries — there is no formula-based funding arrangement. Meanwhile, developed states spend at least US$20 000 to process and meet the needs of each one of the tiny minority of refugees able to reach them. On average, then, the world now spends more than fifty times as much on a refugee arriving in the developed world as it does to protect a refugee who remains in the less developed world. This is clearly not even on the equitable Richter scale.
In such circumstances, it should come as no surprise that the living conditions of refugees in less developed host countries are often dire. In far too many cases rights abuse is rampant, and rationalised on the basis of extreme resource shortages. There is thus a very strong basis for the case being made by Australia and some other states that it makes sense more fairly to apportion resources in relation to relative need, and to require much more funding to be directed towards protecting refugees in the less developed world.
The reallocation, however, needs to be both much more significant and, most fundamentally, binding. Countries in regions of origin rightly protest that they cannot be expected to admit massive numbers of refugees — to whom they thus become legally obligated — on the basis of no more than discretionary grants which ebb and flow with the political, budgetary, and other preferences of wealthier governments. More fundamentally still, the rights of refugees in the less developed world are in no sense meaningfully vindicated by dollars sent to run UNHCR or other refugee camps, where rights abuse is common and opportunities for self-reliance usually nonexistent.
Indeed, the mandatory encampment policies pursued in much of the less developed world are themselves in flagrant breach of both the Refugee Convention and the International Covenant on Civil and Political Rights (‘ICCPR’). Yet those camps are increasingly sustained and funded by us in the name of vindicating refugee rights in regions of origin. If the transfer of resources is to be meaningful, it must absolutely be based on verifiable respect for refugee law obligations in the destination states. But to this point, we seem content to throw money at the less developed world as a sop to our consciences for the harsh treatment of refugees in our midst, even as we know that such resources do not dependably accrue to the real benefit of promoting the goals of refugee safety, autonomy, and self-reliance at the heart of the Refugee Convention.
Enhanced use of strategic resettlement is a similarly solid idea. But let us be frank: what is on offer today are numerically insignificant opportunities from only a fraction of the world’s developed governments. These do not even come close to meeting the need for residual resettlement of refugees who have no durable prospect of protection in the less developed world. Moreover — with only a few exceptions, for example women at risk and unaccompanied minors — it is not the case that we target resettlement efforts at ‘those refugees most in need’ as the official publicity would have it. We instead recruit those who we deem most suitable by reason of their education, work, or other abilities, and offer them the chance to leave the purgatory of refugee camps in Africa and elsewhere. While there may well be good reasons under the current system to seek out resettlement cases that we think are likely to be successful, it is disingenuous to market such efforts as fundamentally needs-based, much less as a needs-based alternative to asylum.
Finally, it must be said that as important as it is to protect refugees closer to their homes, there is simply no reason to stigmatise those who flee farther afield as ‘less deserving’ than those who remain in their own region. As courts have rightly insisted, there are often very good reasons why refugees do not seek protection in their own part of the world. Safety and security rank high on such a list, but the absence of fair status determination systems and the lack of access to any meaningful opportunity to re-establish themselves socially and economically are also valid concerns. Because we know that there is in fact no ‘protection’ worthy of the name — much less protection in line with legal obligations — being provided in most of the less developed world today, it is dishonest to stigmatise as ‘less needy’ those refugees who either have resources or who mortgage their future to traffickers in order to seek protection where they believe they will be treated fairly, where their children can learn, and where they are free to think and speak as they wish. Which one of us, confronted with the need to flee, would not make the same choice?
In sum, the official rhetoric surrounding refugee law reform is largely unhelpful. Refugees are not required to seek protection either in the first country they come to, or in their own region; nor are they are acting unlawfully by arriving to seek protection in a state which has not pre-authorised their presence. Moreover, as valuable as it is to promote the sharing of burdens and responsibilities on a more equitable basis, that goal is not served by a system of purely discretionary resettlement or fiscal transfers which promotes or sustains local responses — in particular, mandatory and long-term encampment — which are not themselves rights-regarding.
Now lest it be thought that blame for the failure to move the reform of international refugee law forward rests entirely at the feet of governments, I now wish to point to what I view as two elements of unhelpful intransigence on the part of the majority of the refugee advocacy community.
First, I believe it must be acknowledged that refugee law does not require states to admit refugees as permanent immigrants. As codified in international treaty law, refugee law simply guarantees the right of seriously at-risk persons to cross international borders, and be accommodated with guarantees of safety and dignity until and unless the threat in their home state is eradicated. Refugee law is therefore fundamentally a mechanism of human rights protection, not a mode of immigration. By erroneously insisting on an absolutist linkage between refugee status and a right of permanent immigration, advocates raise the stakes for governments. If every refugee admitted to an asylum country must be allowed to stay forever, it should come as no surprise that states feel an obligation to constrain their protective initiatives to accord with general immigration priorities.
Second, it is not the case that these human rights responsibilities need to be implemented unilaterally. As long as refugee rights are respected, the Refugee Convention actually encourages governments to collaborate in the sharing of burdens and responsibilities. More specifically, until a refugee is actually admitted to a state’s status determination procedure — at which point he or she becomes, in the language of the treaty, lawfully present — governments may lawfully assign their protection responsibilities to another country, even without the refugee’s consent. But — and this is a big but — any such assignment is legally contingent on meaningful verification by the sending state that the destination country will both recognise refugee status in accordance with international standards, and respect all Refugee Convention and other internationally binding rights. The refugee must moreover have access to a fair procedure under which to challenge the lawfulness of his or her proposed transfer. If, however, all of these criteria are met, responsibility sharing prior to lawful presence being established is legal.
Many non-governmental organisations and other advocates will resist the last two points I have made, just as many governments will resist the first set of baseline principles I posited. It is resistance of this kind which to my mind has made the contemporary debate about the reinvigoration of refugee law sterile. There is a near-universal misunderstanding of the authentic framework of international refugee law, leading to an assumption that it is unsuited to addressing contemporary refugee flows.
My view, in contrast, is that the normative foundation of the Refugee Convention remains sound and is a sufficient basis upon which to build a revitalised regime. I would therefore like to highlight some of the ideas which seem to me worth considering as a means of reinvigorating the mechanisms of international refugee law. In so doing, I want to be clear that what I am proposing is firmly grounded in existing treaty law. I do not propose disturbing the normative consensus about who qualifies for refugee status, nor the rights to which refugees are entitled. The challenge, in other words, is not to rewrite refugee law, but rather to take advantage of the flexibility which the extant body of law affords to retool it at an operational level.
Let me first be clear about why I believe that refugee law is worth saving.
There are those who resist meaningful refugee law reform on the grounds that the world should instead focus on fixing the human rights abuse and other problems that force refugees to flee in the first place. At an ideal level, this is of course a wonderful notion. But until and unless we reach the point where violence and other abuse can truly be stopped around the world, it would be completely unethical to abandon the palliative refugee law system designed to provide solid protection in a world where solutions to hatred are in regrettably short supply.
Despite all the talk about attacking ‘root causes’ of human rights abuse, and despite assertions regarding the attenuated nature of modern sovereign power, it regrettably remains the case that the international community can only make a real guarantee of rights to persons who are outside their own country. This notion that alienage is key to the making of real guarantees of protection is built into the definition of a refugee. A refugee is not just a fundamentally disfranchised human rights victim, but is, by definition, someone who has managed to get outside of his or her own country. Having left their country of origin, refugees are within the unconditional protective competence of the international community. As such, the special ethical responsibility towards refugees follows not just from the gravity of their predicament, but also from the fact that it is always possible to address their plight in ways that, sadly, we still cannot for those who remain inside their own country.
If and when the rhetoric about attacking ‘root causes’ is matched by reality, then few, if any, people will need or qualify for refugee status. But we are not there yet. Refugee law is thus a critical parallel track to intervention to promote human rights: it is the palliative mechanism that does what can be done for those to whom we have access until and unless the harm is stopped. There is no inconsistency between seeking to intervene and committing ourselves to palliation while those efforts are pursued.
Framing the point differently, and contrary to the underlying assumption of most contemporary thinking that ‘forced migration’ is a presumptively negative phenomenon, I believe that involuntary migration should be recognised as a positive option for persons faced with critical threats to human dignity inside their own state. There is a very real risk that a fixation with the avoidance of exodus, and with encouraging in-country intervention as an alternative to external protection, can blind us to the valuable human rights opportunities presented by involuntary migration. Current thinking about forced movement fails to recognise that exile may in fact be the ‘least bad’ option from the perspective of protection.
The particular beauty of refugee law is that it not only commits states to do what they can for the victims of serious human rights abuse who are accessible to them, but it does so in a way that is reconcilable to their core interests. Refugee law has been able to entrench a critical humanitarian exception to the usual rules of border control because it is reconcilable to the self-interest of states. More specifically, the practicality of refugee law follows from the need of governments committed to border control for a mechanism to respond to the arrival of involuntary migrants that is reconcilable to hard realities, as well as to states’ own basic legal and political values.
In terms of hard realities, history makes it clear that at least a minority of any truly at-risk population will find a means to flee to relative safety in another country. And at least some number — because they are better resourced, better informed, more ingenious, or simply more desperate — will travel far from their own country to seek protection. In a highly interdependent world in which impermeable barriers to migration normally impose unacceptably high costs, it is in the self-interest of states to establish an efficient means to accommodate inflows which are, in practical terms, unstoppable. The alternative would be to concede that border control is essentially an unviable project. For states, the real value of refugee law is that it accommodates the claims of those whose arrival cannot be dependably stopped, even as it vindicates the exclusionary norm in relation to other would-be entrants.
Second and equally important — at least in states legally committed to basic democratic values — is that the mechanism for responding to involuntary migration needs to be structured in a way that respects the duty of governments to act fairly. Refugee law meets this test by defining those to whom protection is due on the basis of agreed human rights standards, in particular, respect for the core norm of non-discrimination. Refugee law therefore rejects arbitrariness and holds states accountable to honour a decision to flee motivated by the very values said to be fundamental and inherent in all persons. It is a principled means by which to pursue the practical end of migration control.
By embracing, channelling, and legitimating essentially unstoppable flows, refugee law sustains and validates the protectionist norm. In this sense, refugee law functions as a sluice gate in the dam of immigration control, raised or lowered by states to avoid rampant flooding or the cracking under pressure of the dam itself. Like a sluice gate, refugee law is an imperfect but practical mechanism: it can minimise the potential for inundation, but its functioning requires tolerance of some unplanned inflow. Refugee law has historically been valued by states because it is a politically and socially acceptable way to maximise border control in the face of recurrent involuntary migration.
In structuring the compromise between the interests of refugees to be protected and of states to control their borders, the drafters of the Refugee Convention were clear that states receiving refugees retain the sovereign discretion to decide which refugees will be allowed permanently to integrate. While refugee law imposes a strict duty to refrain from any actions which might infringe a refugee’s ability to access real protection, it mandates entry only to the extent truly required by what amounts to a claim of necessity. It requires that host communities receive refugees in dignity, but does not compel asylum countries permanently to redefine the nature of their own political community.
This does not mean, however, that states are free to conceive so-called ‘temporary protection’ policies in whatever way they wish. Many Refugee Convention rights — including, for example, the right not to be detained other than during a strictly provisional assessment of identity and risk — inhere provisionally long before refugee status is formally recognised by a state party. Even more critically, the Refugee Convention does not allow governments to assert jurisdiction over refugees on the high seas without simultaneously assuming responsibility for their protection. Perhaps most obviously of all, neither refugee law nor international law more generally allow a state to avoid its freely assumed refugee law obligations by the disingenuous manoeuvre of purporting to declare any portion of its territory to be non-territory for refugee law purposes. Tactics of this sort are not only legally unviable, but are simply unworthy of states committed to human rights, and more generally to the rule of law.
How, then, should we reinvigorate the international refugee law regime in a way that states continue to see it as reconcilable to their self-interests, in particular their migration control objectives, yet which does not compromise the right of refugees to access true protection?
First, we should acknowledge that there is no legal prohibition on limiting refugee status and the rights attached to it to the duration of risk in the country of origin. That is, refugee protection may lawfully be conceived as a human rights remedy, not an immigration path. There will, of course, be some cases that call for immediate integration — unaccompanied minor refugees and the severely traumatised being perhaps the clearest examples. Humanitarian principles and psychosocial research moreover suggest the need to make a commitment to the conversion of refugee status to a more permanent status after roughly five years abroad. But even with these limitations, it remains that a conservative estimate of 50 per cent of global asylum capacity could be regenerated by the routine implementation of refugee law as a system of rights-regarding protection for the duration of risk rather than as a system automatically leading to immediate and permanent immigration. This, it seems to me, is a good thing.
Second, assuming the various procedural caveats already noted, we should acknowledge that governments may allocate the responsibility to protect refugees between and among themselves. However, this must be a genuine rights-regarding allocation of responsibility, not a simple dumping of refugees abroad on the amorphous ground that they will be admitted and protected from expulsion in the destination country. In particular, this operational flexibility may not under any circumstance override the core commitments to protection embodied in the Refugee Convention. This means that governments must allow all who wish to claim protection as refugees access to their territory, at least pending an assignment of responsibility, and it means that refugees arriving may not be stigmatised as unlawful entrants. It also means that account must be
taken — both in the site of arrival and in any potential state to which protective responsibility is to be assigned — of the full requirements of refugee and other international human rights law, not just of the ability to secure entry and to be protected against refoulement
The process of sharing-out the fiscal burdens and human responsibilities of refugee protection should not be unilaterally engineered. Nor should it be imposed by a coalition of wealthy states just because they are able to drive a hard bargain with poorer countries. I propose instead two cornerstones for a principled and meaningful system to share burdens and responsibilities.
First, we should move away from a system of unilateral, state-by-state implementation of refugee law towards a system of ‘common but differentiated responsibility’. States now increasingly make binding commitments to a wide variety of regional and other organisations, based on common interests such as free trade, security, the environment, economic development, and shared heritage. The impetus for states to share refugee protection responsibilities should come from an appreciation that cooperation offers them a form of collective insurance when they, or states with which they have close ties, are faced with a significant refugee influx. It is only by ensuring the broad distribution of the responsibility of physical protection, and the reliable availability of fiscal support, that states will feel able to remain open to the arrival of refugees.
More specifically, on the financial (burden-sharing) side, a globalised system of common but differentiated responsibilities should encompass a major, binding, and practically enforceable obligation — not just rhetoric — to shift protection dollars to the places where most protection needs to happen. Some developed governments have advocated this shift at a principled level. However, no developed government has taken the initiative to design and formalise a truly binding system of resource redistribution that would dedicate at least the same quantum of funds now spent on refugee protection in wealthier countries to the poorer regions where most refugees are today, and will remain. Binding fiscal burden-sharing must replace the fickleness and unpredictability of the current charity-based regime if governments of the less developed world are to have the confidence they need to remain open to the arrival of refugees.
In terms of the human dimension (responsibility-sharing), we should be open to different states taking on different kinds of responsibilities. Apart from a common duty of all states to provide first asylum, there is no reason to expect every country to play an identical refugee protection role. It is important to recognise that there are very real differences in the manner in which different countries can best contribute to the successful implementation of a more collectivised system of refugee protection. For example, some governments will be amenable to sharing the responsibility of providing protection for the first several years after flight. Others will be willing permanently to resettle those refugees whose special needs require immediate permanent integration, or who cannot safely return home within a few years after flight.
The precise allocation of burdens and responsibilities should be flexible, but should operate against a foundational principle that even the significant assumption of fiscal burdens cannot justify withdrawing from human protective responsibilities. While the provision of major fiscal support should clearly be factored into the equation, every state party should be required to play a meaningful role in not just the sharing of burdens — that is, financial and related costs — but also in the assumption of human responsibility for refugee protection. Clear criteria, established in advance and supervised by a revamped UNHCR, should guide this process of distributing responsibilities and burdens among states.
Beyond the shift to embrace a globalised system to share refugee protection burdens and responsibilities, the second critical goal must be to reach agreement on a meaningful system to oversee the common but differentiated responsibility and resource transfer regimes. It would otherwise be far too easy for the developed world simply to ‘buy out’ most of its protection obligations, and for less developed host states to profit from the presence of refugees without honestly ensuring their rights. The approach on the ground must be rooted in the central importance of ensuring refugee autonomy and self-reliance, precisely in line with the rights regime established by the Refugee Convention.
The approach to reform I have outlined not only requires no reform of either the existing Refugee Convention or its companion Protocol, but draws both its strength and foundational commitments from these treaties. The critical difference between what I am proposing here and most of what is on the table today is that the variants of orderliness, structure, and systematisation now under discussion do not begin from, nor are they primarily directed to, the provision of meaningful protection to refugees. Current reform initiatives seem instead primarily designed to achieve state-based migration management goals.
The more comprehensive model I advocate is clearly attentive to state interests — but, I must add, to the interests of all states, not just the powerful minority. More fundamentally, however, it is anchored in a recognition that most refugees today do not enjoy the rights which refugee law formally guarantees them — either because they have no access to a safe state, or the safe state to which they are able to travel is either not really safe or forces them to languish without hope of regaining any meaningful measure of autonomy in their lives. That is the reality that must end.
It is, in my view, our responsibility to be open to pragmatic reforms that will collectivise and share burdens and responsibilities to provide the human rights remedy at the heart of refugee law, even as we must never agree to sacrifice the entitlements of those who cannot safely remain at home. Refugee law is a powerful sign of solidarity with the world’s most severely at-risk people. It is the only international human rights remedy which can be engaged directly and immediately by at-risk persons themselves. Most important of all, it is a fundamentally practical remedy which can be reconciled to the most basic interests of states. It is, in sum, a uniquely valuable asset which must never be allowed to atrophy.
[*] James E and Sarah A Degan Professor of Law, and Director of the Program in Refugee and Asylum Law, University of Michigan Law School. These remarks were delivered at the 25th Annual Allen Hope Southey Memorial Lecture at the University of Melbourne Law School on 22 February 2007.
 Opened for signature 28 July 1951, 189 UNTS 150 (entered into force 22 April 1954). The Refugee Convention has been complemented by the Protocol relating to the Status of Refugees, opened for signature 31 January 1966, 606 UNTS 267 (entered into force 4 October 1967) (‘Protocol’).
 See, eg, UNHCR Regional Office Canberra, ‘Australia’s Protection Obligations: Only a Last Resort?’ (Discussion Paper No 1/2004, UNHCR, 2004) 3.
 As the Guardian newspaper conceded after complaint about its use of the ‘bogus asylum-seeker’ label:
This term [‘illegal asylum seeker’] is always incorrect. It cannot be illegal to seek asylum since everyone has the fundamental human right to request asylum under international law. The term ‘bogus asylum-seeker’ is also inaccurate and misleading as it pre-judges the outcome of an asylum application — rather like describing a defendant as entering a ‘bogus plea of innocence’ during a trial.
Ian Mayes, ‘Seeking Asylum without Prejudice’, Guardian (London, UK) 24 April 2004, 24, citing MediaWise, Reporting Asylum Seekers and Refugee Issues: A Resource by and for Journalists (2004).
 Refugee Convention, above n 1, art 31.
 See generally James C Hathaway, The Law of Refugee Status (1991) 46–7. ‘First country of arrival rules’ have nonetheless become a pervasive feature of the refugee protection regimes of developed countries: see James C Hathaway, The Rights of Refugees under International Law (2005) 293–6. But state practice which is at odds with formal treaty-based obligations cannot in and of itself give rise to new law superseding those treaty obligations: at 68–74.
 The most relevant provision of the Refugee Convention, above n 1, is art 1(E), which excludes from refugee status persons ‘recognized by the competent authorities of the country in which [they have] taken residence as having the rights and obligations which are attached to the possession of the nationality of that country’. But even this provision only excludes persons who are de facto nationals from refugee status, not those who might have sought protection elsewhere: see Hathaway, The Law of Refugee Status, above n 5, 211–14.
 The UNHCR Executive Committee has stated that:
Refugees and asylum-seekers, who have found protection in a particular country, should normally not move from that country in an irregular manner in order to find durable solutions elsewhere but should take advantage of durable solutions available in that country.
UNHCR Executive Committee, 40th sess, Problem of Refugees and Asylum-Seekers Who Move in an Irregular Manner from a Country in Which They Had Already Found Protection, Conclusion on International Protection No 58 (XL), UN Doc A/AC.96/737 (19 October 1989) [e] (emphases added).
 The High Court of Australia seems recently to have moved cautiously towards recognising this requirement of international refugee law. In NAGV and NAGW of 2002 v Minister for Immigration and Multicultural and Indigenous Affairs  HCA 6; (2005) 222 CLR 161, the Court determined that
a perusal of the [Refugee] Convention shows that, Art 33 apart, there is a range of requirements imposed upon Contracting States with respect to refugees some of which can fairly be characterised as ‘protection obligations’. Free access to courts of law (Art 16(1)), temporary admission to refugee seamen (Art 11), and the measure of religious freedom provided by Art 4 are examples: at .
Regrettably, the High Court gave no sense of the reasons that led it to characterise only these particular rights as limits on removal. And to be frank, if the right to access the courts is correctly deemed a ‘protection obligation’, why not the cognate right to administrative assistance? And if religious freedom is logically understood to be a ‘protection obligation’, then why is the same not true of freedom of association, or of movement?
 According to the UNHCR Executive Committee:
The intentions of the asylum-seeker as regards the country in which he wishes to request asylum should as far as possible be taken into account; … Regard should be had to the concept that asylum should not be refused solely on the ground that it could be sought from another state. Where, however, it appears that a person, before requesting asylum, already has a connection or close links with another State, he may if it appears fair and reasonable be called upon first to request asylum from that State.
UNHCR Executive Committee, 30th sess, Refugees Without an Asylum Country, Conclusion on International Protection No 15 (XXX), UN Doc A/AC.96/572 (16 October 1979) [(h)(iii)], [(iv)] (emphases added). Additional flexibility of this kind is authorised only where necessary to avoid a ‘refugee in orbit’ situation: see UNHCR Executive Committee, 44th sess, General Conclusion on International Protection, Conclusion on International Protection No 71 (XLIV), UN Doc A/AC.96/821 (12 October 1993) [k]; and where there is genuine regional harmonisation of national policies ‘to ensure that persons who are in need of international protection actually receive it’: UNHCR Executive Committee, 45th sess, General Conclusion on International Protection, Conclusion on International Protection No 74 (XLV), UN Doc A/AC.96/572 (11 October 1994) [p].
 James C Hathaway, ‘A Reconsideration of the Underlying Premise of Refugee Law’ (1990) 31 Harvard International Law Journal 129, 165–7.
 Hathaway, The Rights of Refugees under International Law, above n 5, 160–73. See also UNHCR Executive Committee, 54th sess, Conclusion on Protection Safeguards in Interception Measures, Conclusion on International Protection No 97 (LIV), UN Doc A/AC.96/987 (10 October 2003) [(a)(i)].
 Refugee Convention, above n 1, art 31(1):
The Contracting States shall not impose penalties, on account of their illegal entry or presence, on refugees who, coming directly from a territory where their life or freedom was threatened in the sense of article 1, enter or are present in their territory without authorization, provided they present themselves without delay to the authorities and show good cause for their illegal entry or presence.
 Ibid. See generally Hathaway, The Rights of Refugees under International Law, above n 5, 386–405.
 See, eg, Department of Immigration, Multicultural and Indigenous Affairs, ‘The Principle of Effective Protection in the Australian Domestic Legal and Policy Context’ (Discussion Paper No 1/2004, UNHCR, 2004).
 Ibid 6.
 ‘Australia is increasing the size of the Humanitarian Program to 13,000 places and within it, the size of the refugee category from 4000 to 6000 places in 2004–05’: ibid 5.
 As of 31 December 2005, only about 800 000 of the world’s 12 million refugees were present in the developed world (including Europe; Canada and the US; Australia, New Zealand, and Japan): US Committee for Refugees and Immigrants, World Refugee Survey 2006 (2006) 4–5, available from <http://www.refugees.org> at 18 May 2007.
 Ibid 14.
 At the beginning of 2006, the UNHCR was providing direct assistance to some 4.4 million refugees present in poorer countries: UNHCR, 2005 Global Refugee Trends (2006) 3, available from <http://www.unhcr.org/statistics.html> at 18 May 2007. The UNHCR’s combined budget and refugee-oriented supplementary programmes totals only US$1.243 billion: UNHCR, UNHCR 2007 Financial Overview (2007) <http://www.unhcr.org/partners/PARTNERS/45f027512.pdf> at 18 May 2007. Even if every dollar were spent on refugee assistance (that is, ignoring all administrative and other costs) this would equate to only about US$0.78 per day per refugee assisted. No comprehensive statistics are available to document the amount spent by less developed countries hosting refugees without direct UNHCR support, though it is likely that such expenditures per capita are far below those of UNHCR-assisted refugees.
 See UNHCR, ‘Part I: Identifying Needs and Funding Programmes’, UNHCR Global Appeal 2007 (2007)  <http://www.unhcr.org/static/publ/ga2007/ga2007toc.htm> at 18 May 2007.
 Jenny Bedlington, ‘Creating Shared Solutions to Refugee Protection: An Agenda for the International Community’ (Speech delivered at the Advanced Study Center of the International Institute, University of Michigan, Michigan, US, 14 April 2004):
Developed States — with their legally sophisticated asylum systems and administrative and judicial review frameworks — ie the US, Canada, Western Europe and Australia and NZ, spend at an absolute minimum 10 billion USD on the approximately one million asylum-seekers who have reached their territory, of which only about 20% are refugees. This is an annual cost of 20–25,000 USD per
asylum-seeker, or put another way, about 100,000 USD for each recognised refugee.
Jenny Bedlington, a former First Assistant Secretary in the Refugee and Humanitarian Division of the Australian Department of Immigration and Multicultural Affairs, was at the time of this speech a Consultant to the Inter-Governmental Consultations on Asylum, Refugee and Migration Policies, a consortium of developed world states, which produced and authorised her to release these statistics. More recent statistics have not, however, been provided to the public. Over the last three years, asylum arrivals in developed countries have decreased without a proportionate reduction in the administrative infrastructure in place to address their claims. It is therefore likely that the present cost of addressing the claims and needs of refugee claimants in the developed world exceeds the 2004 statistics cited here.
 See generally UNHCR, The State of the World’s Refugees: Human Displacement in the New Millennium (2006).
 This position has been articulated by R Margabandhu, the Indian representative to the UNHCR Executive Committee:
Today, neither the duty to receive refugees nor the real costs associated with their arrival are fairly apportioned across the world. Distribution of State responsibility is based on accidents of geography and the relative ability of States to control their borders. The entire system survives tenuously on undependable funding and vague promises of cooperation. A large number of developing countries engaged in a desperate struggle to provide their own citizens with the basic necessities of life find themselves crippled by the enormous burden that mass exoduses impose on them. [To] date, the international response to refugee crises has been mostly on an ‘ad hoc’ basis. We believe that ... we must pave the way for equitable and consistent responses to all refugee situations in all parts of the world. Developing countries who put at risk their fragile environment, economy and social fabric to provide refuge to millions are in reality the largest ‘donors’ to the refugee cause.
R Margabandhu, Member of Parliament, India, ‘Questions relating to Refugees and Displaced Persons and Humanitarian Questions’ (Statement to UNHCR Executive Committee, 12 November 1998) <http://www.un.int/india/ind237.htm> at 18 May 2007.
 See, eg, Guglielmo Verdirame and Barbara Harrell-Bond, Rights in Exile: Janus-Faced Humanitarianism (2005) 15–16.
 Above n 1, art 26.
 Opened for signature 16 December 1966, 999 UNTS 171, art 12 (entered into force 23 March 1976). See also Hathaway, The Rights of Refugees under International Law, above n 5, 695–719.
 See, eg, UNHCR Executive Committee Standing Committee, 33rd mtg, Local Integration and Self Reliance, UN Doc EC/55/SC/CRP.15 (2 June 2005) :
the 1951 Convention gives refugees a solid basis on which they can progressively restore the social and economic independence needed to get on with their lives. Host States should make every effort to assure to refugees the rights envisaged under the 1951 Convention, particularly those rights which relate to income generation. This includes the right to freedom of movement enabling refugees to market their goods and access the labour market. Equally important is refugees’ ability to access education, health care and other social services where available.
 In 2005, UNHCR reported that it requested resettlement for only 46 000 refugees worldwide, and that 38 500 were in fact resettled in 23 destination countries: UNHCR, Measuring Protection by Numbers: 2005 (2006) 19 <http://www.unhcr.org/publ/PUBL/4579701b2.pdf> at 18 May 2007.
 See, eg, Department of Immigration, Multicultural and Indigenous Affairs, above n 15, 5, in which the Department claims to be ‘focussing on providing resettlement opportunities for those refugees most in need’.
 This has received tacit recognition by the UNHCR which has stated that:
UNHCR aims to ensure predictability and global consistency in the application of resettlement criteria according to identified needs and priorities. However, the country of resettlement makes the decisions concerning who to admit on the basis of national policies and requirements.
UNHCR, Measuring Protection by Numbers: 2005, above n 30, 19.
 Hathaway, The Law of Refugee Status, above n 5, 46–50.
 See Hathaway, The Rights of Refugees under International Law, above n 5, 977–90.
 Refugee Convention, above n 1, art 1(C)(5).
 See, eg, Deborah Anker, Joan Fitzpatrick and Andrew Shacknove, ‘Crisis and Cure: A Reply to Hathaway/Neve and Schuck’ (1998) 11 Harvard Human Rights Journal 295, 302–3.
 See generally Hathaway, The Rights of Refugees under International Law, above n 5, 322–4. See above Part II(B).
 Hathaway, The Rights of Refugees under International Law, above n 5, 175–8. See especially Rajendran v Minister for Immigration and Multicultural Affairs (1998) 86 FCR 526.
 Hathaway, The Rights of Refugees under International Law, above n 5, 663–4.
 Refugee Convention, above n 1, art 33(1). See also Hathaway, The Rights of Refugees under International Law, above n 5, 322–33.
 ICCPR, above n 28, art 13. See also Manfred Nowak, UN Covenant on Civil and Political Rights: CCPR Commentary (1993) 228.
 See Fourth Colloquium on Challenges in International Refugee Law, ‘The Michigan Guidelines on Protection Elsewhere’ (2007) 28 Michigan Journal of International Law 1 (forthcoming).
 See generally James C Hathaway (ed), Reconceiving International Refugee Law (1997); James C Hathaway and R Alexander Neve, ‘Making International Refugee Law Relevant Again: A Proposal for Collectivized and Solution-Oriented Protection’ (1997) 10 Harvard Human Rights Journal 115.
 See, eg, Sadako Ogata, ‘Foreword’ in UNHCR (ed), The State of the World’s Refugees 1995: In Search of Solutions (1995) 8–9:
The limitations of … traditional solutions, coupled with the growing scale of the refugee problem and the changing nature of the international political and economic order, have prompted UNHCR to develop a new approach to the question of human displacement. This approach is proactive and preventive, rather than reactive. … UNHCR’s work is becoming more and more linked with a wide range of UN efforts, from political negotiations, peacekeeping operations, economic and social development, to the defence of human rights and environmental protection.
There is, however, reason for some optimism that the new High Commissioner’s insistence on a rededication of the UNHCR to the primacy of protection will reverse this trend, at least in part. The agency’s current Mission Statement, for example, attempts to link efforts to address ‘refugee problems’ and pursue ‘solutions’ with the advancement of refugee rights:
UNHCR is mandated by the United Nations to lead and coordinate international action for the worldwide protection of refugees and the resolution of refugee problems. UNHCR’s primary purpose is to safeguard the rights and well-being of refugees. In its efforts to achieve this objective, UNHCR strives to ensure that everyone can exercise the right to seek asylum and find safe refuge in another state, and to return home voluntarily. By assisting refugees to return to their own country or to settle permanently in another country, UNHCR also seeks lasting solutions to their plight.
UNHCR, ‘UNHCR Mission Statement’, UNHCR Global Report 2005 (2006) 3 <http://www.unhcr.org/publ/PUBL/4492676f3.pdf> at 18 May 2007.
 See, eg, Gil Loescher and James Milner, ‘The Missing Link: The Need for Comprehensive Engagement in Regions of Refugee Origin’ (2003) 79 International Affairs 595.
 Nils Geissler, ‘The International Protection of Internally Displaced Persons’ (1999) 11 International Journal of Refugee Law 451, 457.
 Refugee Convention, above n 1, art 1(A)(2).
 James C Hathaway, ‘Is Refugee Status Really Elitist? An Answer to the Ethical Challenge’ in Jean-Yves Carlier and Dirk Vanheule (eds), Europe and Refugees: A Challenge? (1997) 79, 87–8.
 James C Hathaway, ‘New Directions to Avoid Hard Problems: The Distortion of the Palliative Role of Refugee Protection’ (1995) 8 Journal of Refugee Studies 288.
 Arthur Helton, The Price of Indifference: Refugees and Humanitarian Action in the New Century (2002). See also Susan Martin, ‘Making the UN Work: Forced Migration and Institutional Reform’ (2004) 17 Journal of Refugee Studies 301.
 See generally Aristide R Zolberg, Astri Suhrke and Sergio Aguayo, Escape from Violence: Conflict and the Refugee Crisis in the Developing World (1989).
 Even political realists such as Morgenthau acknowledge that there is a moral significance to political action, and specifically that there is a ‘tension between the moral command and the requirements of successful political action’: Hans J Morgenthau, Politics Among Nations: The Struggle for Power and Peace (1st ed, 1948) 10.
 The nexus (‘for reasons of’) clause requires that the risk of being persecuted be for reasons of one of five grounds, the common linkage among the five grounds being non-discrimination norms: Islam v Secretary of State for the Home Department  UKHL 20;  2 AC 629. Courts have also turned to international human rights standards to interpret other key definitional questions, such as the gravity of risk required for persecution: see, eg, Canada (Attorney-General) v Ward  2 SCR 689; and the notion of a failure of state protection: Minister for Immigration and Multicultural Affairs v Respondents S152/2003  HCA 18; (2004) 222 CLR 1.
 See generally Hathaway, ‘A Reconsideration of the Underlying Premise of Refugee Law’, above n 10.
 See, eg, the declaration of the French representative to the Conference of Plenipotentiaries, Mr Rochefort:
[France] was quite prepared to continue to assist such refugees so long as assistance was necessary. But if their country of origin reverted to a democratic régime, the obligation to assist them should not fall perforce upon the French Government. ... France had merely said that she did not wish to be under an obligation to continue to provide assistance to refugees who could seek the protection of their country of origin.
Conference of Plenipotentiaries on the Status of Refugees and Stateless Persons, Summary Record of the Twenty-Eighth Meeting, General Assembly, UN Doc A/CONF.2/SR.28 (28 November 1951) 13–14.
 Refugee Convention, above n 1, arts 1(A), 33.
 See Hathaway, The Rights of Refugees under International Law, above n 5, 300–1.
 Refugee Convention, above n 1, art 31(2).
 Hathaway, The Rights of Refugees under International Law, above n 5, 156–86.
 Ibid 335–42. See also UNHCR, Advisory Opinion on the Extraterritorial Application of Non-Refoulement Obligations under the 1951 Convention Relating to the Status of Refugees and its 1967 Protocol  (26 January 2007) <http://www.unhcr.org/cgi-bin/texis/vtx/rsd/rsddocview.pdf?tbl=RSDLEGAL & id=45f17a1a4> at 18 May 2007:
It is UNHCR’s position ... that a State is bound by its obligation under Article 33(1) of the 1951 Convention not to return refugees to a risk of persecution wherever it exercises effective jurisdiction. As with non-refoulement obligations under international human rights law, the decisive criterion is not whether such persons are on the State’s territory, but rather, whether they come within the effective control and authority of that State.
 Hathaway, The Rights of Refugees under International Law, above n 5, 321–2.
 Refugee Convention, above n 1, arts 1(C)(5), 1(C)(6).
 See Hathaway and Neve, above n 43, 180.
 Manuel Angel Castillo and James C Hathaway, ‘Temporary Protection’ in James C Hathaway (ed), Reconceiving International Refugee Law (2007) 1, 17–18.
 Even during the Cold War years, when the causes of refugee flight were arguably more intractable, large-scale repatriations took place within five years of the commencement of mass influx in about one-half of cases: see generally Sadruddin Aga Khan, Special Rapporteur of the UN Human Rights Commission, Study on Human Rights and Massive Exoduses, UN Doc E/CN.4/1503 (31 December 1981). Contemporary evidence is even more encouraging on this point. As the UNHCR notes, ‘[t]he past four years (2002–2006) have seen unprecedented levels of voluntary repatriation with more than 6 million refugees able to return home’: UNHCR, Measuring Protection by Numbers: 2005, above n 30, 17.
 See Fourth Colloquium on Challenges in International Refugee Law, above n 42.
 See Hathaway and Neve, above n 43, 201–2.
 Ibid 207–9.
 Ibid 209.
 Ibid 202–7.
 See Hathaway, The Rights of Refugees under International Law, above n 5, 992–1002.
 The range of current reform efforts is summarised in James C Hathaway, ‘Book Review: Rethinking Refugee Law by Niraj Nathwani’ (2004) 98 American Journal of International Law 616.