Customary International LawSymposium

The principle of responsibility-sharing in refugee protection

An emerging norm of customary international law

In December 2018, the Global Compact on Refugees was adopted. Especially over the last year, its drafting and negotiations could appear in odd contrast to the surrounding world, in which conditions for seeking asylum continued to harshen. Is the compact a step towards countering these conditions, or mere window-dressing without much effect on refugees’ rights? The compact is non-binding and one widely shared view is that all will depend on how states build on it (see analyses here and here). Yet something has happened already. Over the course of the last two years, the principle of responsibility-sharing has been endorsed repeatedly and can today be considered an emerging norm of customary international law (CIL). This norm is limited in content, as it does not prescribe any concrete modalities of responsibility-sharing, but it is nonetheless relevant because it precludes states’ complete withdrawal from participating in the common task of refugee protection. In this post, I trace the recognition of the principle and discuss its status in light of the current theoretical debate about CIL.

Non-refoulement – and its problems

The Global Compact on Refugees, alongside the Global Compact on Safe and Orderly Migration, is the result of negotiations over the period of more than two years, which began in the wake of the European ‘refugee crisis’ of 2015 and led, first, to the adoption of the New York Declaration in September 2016. What prompted the process was the recognition of profound problems in the current system, which is centered on the rule of non-refoulement. Non-refoulement prohibits states from returning a refugee “to the frontiers of territories where his life or freedom would be threatened,” as in the formulation of Article 33, para. 1 of the 1951 Geneva Refugee Convention. It is widely recognized as a norm of customary international law. The exact scope of the rule under CIL is disputed, yet regardless of this question, the centrality of non-refoulement has shaped the current system of international protection.

The obligation that the rule of non-refoulement imposes on states is, on the one hand, far-reaching, and, on the other hand, very narrow. Since the determination of refugee status is declaratory in nature, non-refoulement obliges states vis-à-vis all persons at the border who are potentially refugees and must therefore be admitted on a preliminary basis to assess their claim. In that sense, it is far-reaching. At the same time, non-refoulement leaves a large margin for practices directed at hindering asylum-seekers from reaching the territory. States, to differing degrees and in different forms, engage in such practices of deterrence. These are formally not in breach of international refugee law, but they effectively undermine its purpose.

These practices result, first, in a concentration of refugees in regions neighboring the states of origin, often countries with limited capacity and a diminishing willingness to shoulder the responsibility alone. Second, they lead to the ever-increasing involvement of humanitarian organizations in assisting refugees. Without other solutions, the emergency relief that these organizations provide turns into prolonged situations of humanitarian government, in which refugees’ survival is ensured while they remain excluded from genuine membership in a society. A system centered on the principle of non-refoulement, without any rules about the cooperation of states and their participation in the task of refugee protection, is susceptible to becoming dysfunctional.

The endorsement of a principle of responsibility-sharing

While these dysfunctionalities have been obvious for many years, 2015 marked a moment in which states acknowledged that the system urgently needed reform in order not to collapse. The process that began focused on responsibility-sharing in refugee protection as a core problem and key goal of reform. In March 2016, a High-Level Meeting on Global Responsibility Sharing through Pathways for Admission of Syrian Refugees took place. In late June 2016, a draft was circulated for a Compact on Responsibility Sharing for Refugees, which in para. 6 stresses “the centrality of the principle of responsibility-sharing in ensuring effective refugee protection.” While states could not agree on the adoption of such compact at the UN GA meeting in September, they unanimously adopted the New York Declaration for Refugees and Migrants. In its para. 68, the Declaration states: “We underline the centrality of international cooperation to the refugee protection regime. … To address the needs of refugees and receiving States, we commit to a more equitable sharing of the burden and responsibility for hosting and supporting the world’s refugees, while taking account of existing contributions and the differing capacities and resources among States.” In para. 1 of Annex I on the Comprehensive Refugee Response Framework, the New York Declaration further asserts the “principles of international cooperation and on burden- and responsibility-sharing”.

The New York Declaration foresaw that two separate compacts should be created. While the Global Compact on Safe and Orderly Migration is also of relevance for refugees, the Global Compact on Refugees was where the considerations on responsibility-sharing mainly continued. Unlike the prior statements on a principle of responsibility-sharing, the Global Compact on Refugees was oriented at agreeing on concrete modalities and provisions. The zero draft labelled the first section of the Program of Action “Principal modalities for burden- and responsibility-sharing,” the first draft called it “Mechanisms for burden- and responsibility-sharing,” and the final version speaks of “Arrangements for burden- and responsibility-sharing.”

The focus on responsibility-sharing thus continued throughout the negotiations. The modalities remained rather vague in the version ultimately adopted, which can be a reason for criticism. Yet this criticism does not mitigate the role that responsibility-sharing occupied in the process as a principle. Especially in the absence of agreement about concrete rules, it is noteworthy that states continued to emphasize the centrality of responsibility-sharing. The Global Compact on Refugees was adopted by a vast majority of states, with Hungary and the United States being the only states voting against it. Through this continuous endorsement, the recognition in the New York Declaration and the endeavors in the Global Compact to flesh out more detailed modalities, the principle of responsibility-sharing has successively solidified.

The content of the principle of responsibility-sharing

What is the content of this principle of responsibility-sharing, absent more specific rules on how responsibility is to be shared? The principle precludes a complete withdrawal of states from the task of providing for refugees, and it asserts that the contribution each state has to make is dependent on the overall task. Since states recognizing jointly that an international system of refugee protection only functions with a sound level of participation by all, practices that are directed at avoiding any participation are in violation of the principle of responsibility-sharing. It thus opposes what Peter Sutherland has described as a system of “responsibility by proximity,” in which only the states to where refugees arrive first are obliged to host them, while all other states can reject any share in the task.

This is, obviously, a very limited content. The principle does not say anything about how states can contribute: can financial support of hosting states be sufficient, or does responsibility-sharing require also to let a certain number of refugees enter the country? And the principle does not offer guidance on how responsibility is to be distributed.  Does geographical proximity constitute a factor in which states should bear the main responsibility? What role does a state’s financial or other capacity play?

It is the nature of a principle that it does not regulate details but captures an agreement on the general direction and about the ground on which one argues. The principle of responsibility-sharing nevertheless reflects an important development from the mere mention of international cooperation in the preamble of the 1951 Refugee Convention. It reflects the recognition that the regime of international protection cannot be based on non-refoulement alone, since this has allowed many states to hide behind a formal commitment to refugee protection while evading their share in the responsibility for it. The principle of responsibility-sharing does not counter this situation at once, but it offers a legal reference point for calling out such evading practices and negotiating in specific situations what a more equitable sharing of responsibility requires.

Theories of customary international law

What is the status of this principle of responsibility-sharing, which has been endorsed by states over the course of the last years? My proposition is to understand it as an emerging norm of customary international law. Clearly, this claim depends on the theory one subscribes to regarding CIL. The classic understanding required state practice and opinio juris as combined elements for a rule to be considered part of CIL. However, when these elements are deemed fulfilled does not follow a consistent methodology, as Stefan Talmon has pointed out with respect to decisions of the International Court of Justice, or Jean d’Aspremont has described with respect to positions taken by the International Law Commission. This allows for a dominance of Western states in what is considered part of custom, both in terms of state practice and of opinio juris, as B.S. Chimni argues. (His remarkable article last year was followed by a noteworthy symposium in AJIL Unbound.) Brian Lepard stresses that the assessment what constitutes a rule of CIL never takes place in an ethical vacuum, and argues for a new normative theory that places opinio juris at the center.

Without delving into the details of the debate here, two central insights can be gained from these newer approaches. The first is that the structure of the norm in question matters. In particular, we must ask, who benefits from the rule? Are states incentivized to agree on a rule to coordinate their behavior, or does it reflect a commitment towards individual rights? The second is that the idea of an objective assessment of CIL in detachment from normative convictions is a mirage. Since every determination of CIL takes place against the background of normative convictions, these ideas should be spelled out and CIL more explicitly oriented at the pursuit of more just global order.

The field of refugee law is designed to safeguard the rights of individuals who by definition are not represented by any state. The incentive for states to agree on rules and respect them in practice is low. Accordingly, where states nonetheless show an explicit and continuous endorsement of a norm relating to refugees, this should be weighted in favor of the norm’s status of CIL. Unlike rules that states recognize to promote their own immediate self-interest, the breaking of a rule in refugee protection cannot be considered to exclude its status of CIL as such (as is also the case for the rule of non-refoulement). Much state practice does indeed support the principle of responsibility-sharing, alongside the pledges to mend the current deficiencies. States commit to financial support and to resettling refugees and have done so for many years, despite the recent regrettable decline in some states’ commitments.


States’ obligations towards refugees have always been a strained field. Refugees are the exceptional case in which states have an obligation towards individuals for no other reason than their need and their presence at the border. That this “one cosmopolitan law” has gained an influential position and, as the principle of non-refoulement, has been recognized as customary international law, defies all probabilities. This recognition did, unfortunately, never mean a universal respect of the principle – but that a rule is violated does not as such contradict its status as a rule. States, despite acts of undermining and sometimes outright violating international refugee law, cling to the core idea that the person who is forced to flee her home deserves protection elsewhere.

Refugee law remains a strained field also today. It is not difficult to list examples of refugee rights being violated, and a rhetoric of threat and fear being employed to deter migrants and seal borders. Yet despite these adverse conditions, and partly because of them, states have shown over the last years efforts to repair aspects of the international refugee protection system. Without glossing over the disappointments in the resulting Global Compact on Refugees, the focus to improve the responsibility-sharing in refugee protection must be acknowledged as an important step. While specific modes and rules of responsibility-sharing remain to be developed and negotiated in particular cases, a general principle of responsibility-sharing can be considered an emerging norm of CIL. Such claim builds on newer approaches that have introduced a critical and ethical perspective on CIL and sought to de-colonize the methods of determining it. The principle of responsibility-sharing has been endorsed with view to its central role for a functioning system of international protection. It is limited in its content yet offers a much-needed legal reference point for countering the system’s erosion.


Dana Schmalz is a visiting scholar at the Zolberg Institute on Migration and Mobility at The New School, New York.


Cite as: Dana Schmalz, “The principle of responsibility-sharing in refugee protection – an emerging norm of customary international law”, Völkerrechtsblog, 6 March 2019, doi: 10.17176/20190306-132332-0.

ISSN 2510-2567
Previous post

Trading Blows

Next post

Religious Freedom and Customary International Law



No Comment

Leave a reply

Your email address will not be published. Required fields are marked *