World Refugee Day is an opportunity to focus attention on one of the most pressing refugee law and policy challenges: almost 9 out of 10 refugees are hosted in developing countries. Better distribution of responsibility for seekers of international protection remains urgently needed. Policymakers across the ideological and geographic spectrum—including the United States under both the Biden and Trump administrations, Colombia, the European Union, the African Union, and the United Nations—have all invoked versions of responsibility sharing in their rhetoric, albeit in service of diverging policy objectives.
In a new empirical and analytical paper, forthcoming in the California Law Review, we identify, describe, and rate new responsibility sharing arrangements across the world. Some arrangements are progressive, if second-best policy solutions, that transfer seekers of international protection to more affluent, safer, more institutionally competent states. Others are regressive, third-best mechanisms, that in fact constitute responsibility dumping.
Our metric builds on existing assessments of responsibility sharing in key respects. To start, we move from the realm of political theory and wishful thinking to assess current practices. We also shift the perspective from obligations to individual asylum seekers to obligations host states have toward one another. Most existing assessments of responsibility sharing arrangements, including influential 2013 UNHCR guidance, have demanded compliance only with a legal minimum anchored in the non-refoulement principle—the obligation not to turn asylum seekers back to danger. Our analytic lens is much broader in comparison. What’s more, refugee scholars, advocates, and the news media have largely focused on the proliferation of regressive arrangements that externalize responsibility for asylum seekers, including widely criticized Trump administration policies. We instead argue that scholars, advocates, and others should also recognize and encourage the diffusion of progressive models, even when these have many shortcomings. We identify the EU migration solidarity mechanism as one progressive model that should inspire imitation.
We place various arrangements on a scale from progressive to regressive based on four parameters: (1) hosting commitments, (2) monetary component or equivalent non-monetary assistance and whether they reflect actual asylum seeker hosting costs, (3) bilateral versus multilateral arrangements, and (3) the existence of legally binding implementing instruments. First safe country arrangements, that is, ones that divert asylum seekers to the first “safe” country of transit, are frequently regressive under this logic because of geographic realities. Frontier countries bordering major producers of asylum seekers tend to be developing countries themselves suffering significant hardship.
One example of a regressive model is the Trump administration’s Asylum Cooperative Agreements (ACAs) with Guatemala, Honduras, and El Salvador. The agreements allowed the United States to transfer asylum seekers to one of those three countries instead of processing their asylum requests. Now terminated by the Biden administration, the agreements have been widely condemned on the ground that the United States failed to guarantee a safe and adequate environment for transferred migrants. We find reasons to condemn the agreements beyond safety concerns. The agreements are highly regressive under our framework because they shift seekers of international protection to less safe, less institutionally competent, less affluent countries without adequate offsetting U.S. contributions. Those governments were effectively coerced by withholding U.S. aid. This example is therefore a striking example of responsibility dumping.
By contrast, the evolving EU Solidarity Mechanism is a highly progressive model. In response to the 2015 migration crisis, the EU adopted measures to transfer a total of 160,000 asylum seekers from the most heavily impacted member states—Greece and Italy—to the other EU members. The measures imposed binding quotas for the number of asylum seekers each state was required to accept, calculated according to an allocation formula based on objective parameters such as member state size and GDP. Despite uneven implementation and political backlash, 34,705 asylum seekers were relocated in this framework out of an eventual net target of 98,256. The EU model expands on U.S. efforts to relocate refugees from Indochina in the 1980s by addressing the much larger and more diverse category of asylum seekers—individuals seeking international protection who have yet to be granted refugee status. Separately, Germany famously voluntarily opened its doors to one million asylum seekers.
The EU has subsequently negotiated more permanent reform of its migration solidarity mechanism, expanding on the Dublin Regulations that determine which member state would be responsible for processing an asylum request (typically, the member state of first irregular entry). This effort produced the New Pact proposal in late 2020 and its comprehensive migration regulation. The proposal leaves greater leeway for member states to choose how to meet their allocated share of responsibility. Instead of binding hosting quotas, the New Pact allows member states to opt for monetary or equivalent contributions. Nonetheless, the New Pact solidarity mechanism is still based on a binding allocation formula imposing quantifiable obligations and grounded in objective parameters.
We argue that the EU model shows what is possible in the realm of responsibility sharing for asylum seekers. This mechanism goes well beyond other relatively progressive mechanisms like the Global Compact on Refugees, endorsed by the United Nations General Assembly in 2018. Instead of soliciting voluntary contributions from states—if you will, passing around the collective collection plate—the EU solidarity mechanism is a multilateral, binding, legal instrument that imposes measurable obligations on states under the New Pact proposal. The various versions of the obligation allocation formula consist of objective rather than politically negotiated criteria, which benefits the less powerful member states by isolating the allocation method from the contingencies of any given migration crisis. While the EU solidarity mechanism currently sits on one end of the spectrum, other examples we discuss in the research paper include progressive elements as well.
Our analytical framework aims to inform responsibility sharing norm development and guide domestic policymaking. As we begin to consider concrete norm development in this area, it is essential that commentary and reflection on these issues be grounded not just in theory but also in evidence about different responsibility sharing arrangements.
In the meantime, it is wise to pay close attention to responsibility sharing norm diffusion. We have seen an increase in the incidence of regressive policies such as flawed “safe” third country agreements and tactics for externalizing responsibility for international protection seekers by various states. Proliferation of such regressive arrangements should be discouraged, while diffusion of progressive arrangements like the EU solidarity mechanism across different jurisdictions should be recognized as such and encouraged. While the EU is institutionally unique, it can inspire other responsibility sharing arrangements, especially at the regional and subregional level.