Executive Summary

 

 

 

 

Introduction

As displaced people and other migrants[1] increasingly move out of the conflict-ridden and less-developed regions of their displacement and into relatively rich and stable regions of the world, the countries of destination are increasingly working to contain and even stem the migration flow before the tide of those migrating reaches their shores. One of the cruel ironies in recent years is that countries that have developed rights-sensitive standards and procedures for assessing protection claims of asylum-seekers within their jurisdictions have simultaneously established barriers that prevent asylum-seekers from setting foot on their territories or otherwise triggering protection obligations. Consequently, asylum-seekers who would otherwise have been able to avail themselves of asylum procedurals, social support, and decent reception conditions are often consigned to countries of first arrival or transit that have comparatively less capacity to ensure rights and process claims in accordance with international standards.

Humans migrate for a host of reasons –risk of harm or death; family reunification; or in search of new opportunities – and these factors can evolve en route. Recent estimates place the total number of migrants world-wide at an all-time high and rising – with the latest estimates above 244 million (UNFPA 2002). The number of the forcibly displaced – refugees, asylum-seekers, and internally-displaced persons – are also at historical highs. The overwhelming majority of the approximately 63.5 million people “of concern” to the United Nations High Commissioner for Refugees (UNHCR) at the close of 2015 have remained close to their places of original displacement (more than 38 million of them remain internally displaced in their home countries; 86 percent of the world’s refugees, 13.9 million, are staying in the developing world) (UNHCR 2016c). But hundreds of thousands have moved or are seeking to move beyond nearby countries of first arrival and knocking on the doors—and sometimes pushing them open or climbing over and around them—of industrialized countries, principally member states of the European Union and the United States, as well as Australia, though its deterrent measures have very effectively kept new arrivals from entering in recent years.[2]

The world’s attention has recently been seized by the drama of asylum-seekers and migrants on overcrowded boats and at newly erected fences on the frontiers that divide the world’s “haves” from the “have-nots.” From London to Canberra, political leaders have suggested that people smugglers were responsible for enticing migrants to make their journey or, worse, that human traffickers were forcing or tricking them to board boats or take dangerous overland routes (Frelick 2015).[3] Politicians argued that the “tide” could be “stemmed” by cracking down on smuggling and trafficking crime syndicates and increasing deterrence to dissuade people from migrating. Some gave voice to the worst fears—that terrorists might pose as refugees. Others suggested that many of those at the gates were seeking jobs rather than asylum and were not in need or deserving of rights protections at all. Finally, particularly in Europe, many said that even refugees fleeing conflict and persecution could and should have found asylum in countries closer to home, and were moving onward not to seek protection but rather only in search of a higher quality of life.

This essay acknowledges that key industrialized destination States, and at least the United States and most of the EU member states, have generally been providing protection to the people on their territories with legitimate claims for international protection, but a parallel development has been the development of their toolbox for preventing migrants and asylum seekers from reaching their territories and triggering the state’s international obligations or for removing them quickly without properly examining their refugee claims.

Part I of this article will suggest a working definition of externalization of migration controls and describe some of the ways that this phenomenon works in practice. Part II will briefly discuss the human rights of migrants, and asylum-seekers in particular, and how the externalization of migrant controls affects these rights in practice, including the specific issue of the international responsibility of destination States that pursue migration-control externalization. Part III will highlight a number of the most troubling externalization strategies used by Australia, the European Union, and the United States. Part IV will present some recommendations that could strengthen protection of human rights in the context of externalization of migration management.

Part I: What is Externalization of Migration Controls?

Externalization of migration controls describes extraterritorial State actions to prevent migrants, including asylum-seekers, from entering the legal jurisdictions or territories of destination countries or regions or of making them legally inadmissible without individually considering the merits of their protection claims. These actions include unilateral, bilateral, and multilateral State engagement (see Gammeltoft-Hansen 2011; see also, Crépeau 2013), as well as the enlistment of private actors. These can include direct interdiction and preventive policies,[4] as well as more indirect actions, such as the provision of support for or assistance to security or migration management practices in and by third countries.[5]

As migration policy has become an increasingly politicized issue, externalization is often deceptively framed as either or both a security imperative and a life-saving humanitarian endeavor rather than simply as a strategy of migration containment and control.[6] Control of migration flows is cast as an effort to prevent “illegal” (or irregular) immigration or to protect migrants from the dangers of the journey.[7] Extraterritorial actions to manage migration flows are also increasingly linked to the ineffectiveness (and politicization) of national or regional migration policies. Over time, the phenomenon has expanded to commonly include the systematic enlistment of third countries in preventing migrants, including asylum-seekers, from entering destination states. Third, States enlisted in the prevention of onward movement of migrants and asylum-seekers are—at least implicitly—encouraged to prevent migrants and asylum-seekers from entering their territories or to apprehend and return them (Haddad 2008, 199).[8] Much of this public engagement takes place in the context of transnational crime-control efforts by States that is directed at seeking to counter human trafficking or people smuggling (and often conflates the two) (Plaut, 2015). A prominent example is law enforcement or military assistance designed to stop the flow of illicit materials, such as weapons and drugs, which may have the additional effect of sealing borders (both for exit and entry), encouraging push-backs, increasing apprehensions and/or reducing access to protection mechanisms in the context of apprehension and deportation practices.[9]

Externalization occurs through formalized migration policies and visa regimes, through bilateral and multilateral policy initiatives between States, as well as through ad hoc policies and practices. Externalization policies and practices may explicitly seek to prevent the entry of migrants into a destination State or have only an indirect impact on migration.

The rhetoric that provides the humanitarian rationale for externalization focuses not only on preventing migrants and asylum seekers from embarking on what is characterized as a dangerous journey, but is also framed as an exercise in capacity-building for countries of origin, countries of first arrival, and transit countries (IFRC, 2015). While it may be questionable whether such capacity-building actions seek to improve rights protections in countries of origin as a humanitarian end in itself or rather to address the “root causes” of international migration as a means of containing irregular migration flows, the building of capacity for developing rule of law, respect for human rights, conflict resolution, good governance, and humane quality of life all represent positive aspects of externalization (see Ogata 1995; see also, Haddad 2005, 202-02. 204). While recognizing that successfully addressing root causes should enable more people to choose to remain in their homes or in countries of first arrival, capacity-building policies and practices can have the additional effect, even if only indirectly, of advancing, strengthening, or rationalizing migration-control externalization.

Externalization policies are often also pursued with the stated goal of assisting third countries with migration control and management. Examples of these schemes include policies and practices that encourage both third countries and countries of origin to prevent would-be migrants from migrating through incentives for individuals to remain in place and also through physical or legal barriers (Plaut, 2015); policies and practices encouraging migrant apprehensions (interdictions, interceptions, or “turn-backs” – including on the high seas) through logistical, financial, or political support, or directly in exchange for aid; the development of readmission and incentive structures between third countries and countries of origin (Hyndman and Mountz 2008, 253); [10] financial and political support of migrant detention or interdiction practices by third countries or off-shore (ibid.); or partnerships to combat “illegal” (or irregular) migration or to build capacity of immigration or asylum systems in third countries (Haddad 2008, 196). They can also include measures implemented entirely through requirements imposed on the private sector, such as carrier sanctions, and which have the effect of preventing departure or transit of migrants to destination States. In the context of forced migration, externalization efforts may also extend to efforts aimed at diverting asylum-seekers to third countries, such as to third-country processing centers or “protected areas” near countries of origin (Andrijasevic 2010, 266).

In conjunction with and often as part of the externalization of migration controls, States have pursued a two-pronged strategy to reduce access to asylum. One way is to push, prod, and sometimes pay a country of first arrival or a transit State to seek to curb the migration flow through its own enforcement measures. The second line is to deem the transit or first arrival state a “safe third country” or “first country of asylum” that has or could have provided protection to a refugee, and to which the person can be returned after only a cursory admissibility determination.

Of course, international refugee law has incorporated the principle of "safe third country" as a limited basis for asylum denial in countries of destination (UNHCR Executive Committee 1999), so what is at issue in this context is not the principle per se, but whether the country of first arrival or transit is, in fact, safe.  This requires a factual assessment whether if provides effective protection, which is based on the following criteria:

1. no risk of persecution within the meaning of the 1951 Convention or serious harm in the previous state;
2. no risk of onward refoulement from the previous state;
3. compliance, in law and practice, of the previous state with relevant international refugee and human rights standards, including adequate standards of living, work rights, health care and education;
4. access to a right of legal stay;
5. assistance of persons with specific needs; and
6. timely access to a durable solution (UNHCR 2002).

In the rush to render asylum claims inadmissible on the basis of the safe third country concept, nominal adherence to these criteria has often been deemed sufficient even when there are evident gaps between formal acceptance of principles and their realization in practice. In effect, asserting that a person should be denied asylum because he or she could have sought an ineffective form of protection in a country of first arrival or transit operationalizes the externalization of migration control – and provides perhaps the starkest illustration of the serious threat that externalization poses to the protection of the rights of migrants, and of asylum-seekers in particular.

Part II: The Impact of Border Externalization on the Right of Migrants

Border externalization policies and practices can directly affect the human rights of migrants—and the international obligations of States to protect them—in significant ways.

First, and perhaps most importantly, by directing migrant flows to third countries, externalization influences the nature and duration of State legal obligations, as well as which States are charged under international law with the protection of the rights of migrants. For example, border externalization may prevent migrants from ever coming under the jurisdiction of destination States in an attempt to limit formal legal obligations, including the right to seek and enjoy asylum.^[11]

Such externalization policies and practices can then place significant and unequal burdens on third countries; often states with fewer resources are forced (in practice and by law) to seek to ensure the protection of migrants’ rights, including rights in the context of asylum.^[12]

When the rights of asylum seekers and migrants are then violated in such third countries as a result of a destination State’s externalization efforts, this can raise complicated issues of State responsibility for both destination states and third countries. Simply put, it is a violation of international law for States to directly support the internationally wrongful acts of another State. States may not knowingly “aid or assist” another State in the commission of an internationally wrongful act if the underlying act would be internationally wrongful if committed by the former state (International Law Commission 2001). Indeed, destination States pursuing border externalization strategies may come to be responsible (as a matter of international law) for rights violations outside of their own territory, including at a minimum, if they exert control over the acts of third countries. 

Second, externalization may actually trigger, directly or indirectly, one or more categories of rights violations. Regardless of their status or location, migrants have a range of fundamental rights that can be implicated by migration-control externalization practices and which protect migrants against abuse throughout the migration process.[13] These include rights that are implicated during transit, including while on the high seas and over land, if and when detained as well as during the expulsion or deportation process (see, e.g., UN GA 1948; UN GA 1966a; UN GA 1966b).

In addition to this core set of rights, some migrants may be covered by specific bodies of law that provide a higher standard of protection than is available to other migrants. For example, child migrants are recognized to have a special status under international law, as are those fleeing rights abuses in their countries of citizenship of habitual residence, such as refugees and asylum-seekers or stateless people (see UN GA 1951; UN GA 1990a). At times, migrants also encounter situations during transit that increase their vulnerability or trigger the attachment of additional rights, such as the rights of victims of trafficking or other crimes.^[14]

The perilous journey undertaken by many migrants, including on the high seas, as well as clandestine efforts by some to cross increasingly militarized (and sometimes closed) borders, can expose them to violations of the right to life and the right to seek and enjoy asylum and can also implicate their rights as victims of crime and abuse (such as by traffickers) (see UNOHCHR 2014, 1-2). Externalization can also increase demand for both third country resources for and interest in apprehension of migrants, and this, in turn, can increase the likelihood of apprehension and both the likelihood and the duration of detention. Yet, throughout this entire process, migrants maintain fundamental rights, including to liberty and security of person (including a presumption against detention on the basis of migration status) and, if deprived of liberty and otherwise, rights against torture and ill-treatment (See, e.g. UN GA 1966a, articles 7, 9; UN GA 1984).

In addition to being associated with increased enforcement practices (and detention of migrants), externalization can implicate asylum rights and prohibitions against refoulement. All migrants have the right to leave any country, including their own; refugees and asylum seekers also have rights to seek and enjoy asylum and not to be punished for illegally entering a country to do so (See UN GA 1948, article 13; UN GA 1951, article 31(1)). Perhaps the most important obligation, that of non-refoulement, prohibits States from returning a refugee to territories where her or his life or freedom would be threatened on account of race, religion, nationality, membership of a particular social group, or political opinion or anyone to a state in which they would be exposed to the danger of torture or cruel, inhuman or degrading treatment of punishment (UN GA 1951, article 33; 1966a, article 7; UN GA 1984, article 3; UN HRC 1992).[15]

In some regions and circumstances, those fearing mistreatment are also protected under human rights law from return to a broader set of harms, including generalized violence or serious deprivation of a range of human rights (see, e.g., UNHCR 1984, part III(3))[16] These obligations require that States provide migrants access to screening and examination of any refugee or asylum claims, including in situations of mixed migratory flows (where not all migrants may merit or require international protection) (see UNHCR 2016b).^[17]

Enforcement pressures on third countries can increase the difficulty of crossing borders for asylum seekers and refugees as well as the ability to seek or access procedures for determining refugee status. Given that states are forbidden to practice “indirect” or “chain” refoulement—in which refugees are returned to a third territory where they face persecution, harm, or the serious threat of being returned to their home country or another territory to face persecution or torture—migration-control externalization can implicate the very cornerstone of the human rights regime.^[18]

Part III: Regional Case Studies

The United States, Australia, and the European Union are the primary destination States or regions that have engaged in specific externalization strategies aimed at preventing migrants and asylum seekers from irregularly reaching their shores.

The United States

U.S. efforts to externalize migration controls go back at least to the Reagan Administration. [19] The 1981 Interdiction Agreement between the United States and Haiti authorized the U.S. Coast Guard to interdict Haitian vessels on the high seas, detain the passengers, and return them to Haiti.[20] Despite the inclusion of assurances in Reagan’s Executive Order that the United States would respect the principle of nonrefoulement, U.S. officials brought fewer than one dozen of the 22,651 Haitians interdicted at sea to the United States to pursue asylum claims during the highly repressive Duvalier dictatorships between 1981 and 1990 (Taft-Morales and Sullivan 1993, 9).

A September 30, 1990 coup that overthrew Haiti’s first democratically elected president, Jean-Bertrand Aristide, caused the number of Haitians taking to boats to spike. U.S. President George H.W. Bush attempted to continue the Reagan policy of shipboard screening, but ran into legal challenges that the screening was inadequate.[21] On May 23, 1992, he issued the Kennebunkport Order, which threw out Reagan’s nod to the principle of non-refoulement, and authorized summary returns of interdicted Haitians with no refugee screening whatsoever.[22]In Sale v. Haitian Centers Council, the U.S. Supreme Court countenanced the Kennebunkport Order policy of summary return of refugees interdicted on the high seas.[23] The UN High Commissioner for Refugees called the Sale ruling “a setback to modern international refugee law” (UN High Commissioner for Refugees Responds 1993).

As widespread human rights violations continued in Haiti and Haitians continued to seek protection in the United States, the Clinton Administration briefly tried shipboard screening, but  decided in July 1994 to detain interdicted Haitians temporarily at the U.S. naval base in Guantánamo, Cuba rather than summarily return them to Haiti (Wasem 2011, 5). They were not screened for refugee status but held pending their return to Haiti (ibid., 4). It was clearly conveyed to the Haitians at Guantanamo that they would never be resettled in the United States (Gordon 1994).

President Clinton extended this externalization of migration control to Cuban boat and raft migrants as well. After a decades-long U.S. policy of welcoming all rescued or irregularly-arriving Cubans to the United States and automatically granting them residency status, President Clinton, in August 1994, announced that boat people and rafters fleeing Cuba would, like the Haitians, be interdicted and taken to Guantánamo. This new U.S. policy toward Cuban rafters was confirmed in a “Joint Communique on Migration” issued by the Cuban and U.S. governments on September 9, 1994. The communique stated that Cubans “rescued at sea attempting to enter the United States will not be permitted to enter the United States, but instead will be taken to safe haven facilities outside the United States.” In addition, the communique stated that the Cuban government would “take effective measures in every way it possibly can to prevent unsafe departures using mainly persuasive methods” (U.S. Department of State 1994).

President George W. Bush reverted to the policy of interdicting and summarily returning Haitian boat people, including possible refugees, to Haiti. When a wave of political violence in late February 2004 resulted in a second ouster of President Aristide, and Haitians once again took to boats, President George W. Bush announced, “I have made it abundantly clear to the Coast Guard that we will turn back any refugee that attempts to reach our shore” (Seper 2004).[24]

The Obama Administration has continued high seas interdictions and cursory shipboard screening. Those found to have “credible fears” are brought to Guantánamo where they undergo a refugee status determination without the benefit of legal representation. The few who are recognized as refugees are held at Guantánamo pending third country resettlement; they are not considered for resettlement to the United States.[25]

U.S. externalization of migration controls has a long history on land as well. In 1989, an internal U.S. Immigration and Naturalization Service (INS) memo called on the INS liaison in Mexico “to secure the assistance of Mexico and Central American countries to slow down the flow of illegal aliens into the United States” (Frelick 1991, 2). In June 1989, the INS newsletter, Commissioner’s Communique, reported on “cooperation with the Government of Mexico to stem the flow of Central Americans through that country, including the establishment of checkpoints along the transit corridors and the deportation of intercepted Central Americans” (Jones and Roney 1989). While cooperation between U.S. immigration officials and their Mexican counterparts has ebbed and flowed over the years, the idea of pressuring and supporting Mexico to control Central American migration across its remote 500-mile southern border with Guatemala has had a lasting appeal for U.S. policymakers faced with the challenges of policing the nearly 2,000-mile U.S. border with Mexico.

Through the Mérida Initiative, a cooperative bilateral security agreement between the United States and Mexico that dates from 2008,[26] the U.S. Congress has appropriated about $2.5 billion in assistance to Mexico, of which more than $1.3 billion in equipment and training has been delivered to Mexico, as of April 2015 (Seelke and Finklea 2015). One of the “four pillars” of U.S. security assistance to Mexico through the Mérida Initiative is support to “create a 21st century border” (pillar 3). This assistance is to “[fa]cilitate legitimate commerce and movement of people while curtailing the illicit flow of drugs, people, arms, and cash” (U.S. Embassy-Mexico 2014). While Mérida Initiative funding involves far more than migration control, “curtailing the illicit flow of…people” has remained an important priority for the United States (ibid.).

This became an even higher priority in the summer of 2014 when extreme gang violence and poverty in the Central American countries of Honduras, Guatemala, and El Salvador caused tens of thousands of people, many of them unaccompanied minors, to flee to the U.S. and Mexico.

President Obama committed his administration to meeting the surge “with an aggressive, unified, and coordinated Federal response on both sides of the border.”[27] He told Congressional leaders that he intended “surging law enforcement task forces in cooperation with our international partners, with a focus on stepped-up interdiction and prosecution.”[28] A week later, he asked Congress for an emergency supplemental appropriation of $3.7 billion “to comprehensively address this urgent humanitarian situation.”[29] The highest levels of U.S. diplomacy were engaged to persuade Mexico and the three Central American source countries, El Salvador, Honduras, and Guatemala, to create “concrete ways that we can work together to stem the flow of migrants taking the dangerous trip to the United States.”[30]

In direct response to the summer 2014 surge in unaccompanied Central Americans arriving at the U.S. border, the U.S. Department of Homeland Security (DHS) launched Operation Coyote, which it said was “designed to stem the flow of illegal Central American migration” (Kubiak 2015, 7-8). The operation involved the deployment of DHS investigators to Mexico and Central America “to share criminal intelligence with foreign partners and build capacity in human smuggling and human trafficking enforcement” (Human Rights Watch 2016b, 52). By the end of May 2015, this effort had resulted in 1,037 criminal arrests in Mexico and the region (Kubiak 2015, 7-8).

The Congressional Research Service estimated that U.S. State Department funding for equipment and training to support immigration enforcement on Mexico’s southern border would exceed $86.6 million prior to the enactment of the FY 2015 appropriation (Seelke and Finklea 2015, 16). The U.S. Congress increased the President’s $115 million request for FY2015 Mérida Initiative by another $79 million, and specified that it was to be used “for helping Mexico secure its southern border” and implement justice sector reforms (ibid., 7). Significant funding for Mexican Mérida Initiative programs continued in the Obama Administration’s FY 2016 request.[31] Additional funds also flowed to Mexico’s military from the U.S. Department of Defense’s counter-narcotics budget to bolster its capacity to control Mexico’s southern border.[32]

The combination of funding and technical support as well as diplomatic and other pressures from the United States, spurred the Mexican government to action. The same day as Obama’s emergency supplemental budget request, July 8, 2014, Mexican President Enrique Peña Nieto announced the start of the Programa Frontera Sur (Southern Border Program). Although a detailed program was never presented publicly, the Mexican government established a new administrative body within the Ministry of Interior, the Office of Coordination for Comprehensive Migration Assistance at the Southern Border, to coordinate border control and migration management, and federal officials were designated to manage migration policy in the southern Mexican states of Campeche, Chiapas, Quintana Roo, and Tabasco (Poder Ejecutivo 2014, 1, 2; Monroy 2015). Moreover, in September 2014, the New York Times reported that “under pressure from the United States . . .  Mexico in recent weeks has taken a rare step toward stemming the flow of migrants, sweeping [Central American migrants] off trains, setting up more roadway checkpoints and raiding hotels and flophouses where they congregate on their journey north” (Villegas and Archibold 2014).

Apprehensions of non-Mexican migrants along the southwestern border fell by 57 percent between October 2014 and April 2015 compared to the same months the previous year, from 162,700 to 70,400 (Central America's 2015). As early as September 2014, when the number of Central Americans appearing at the U.S. border decreased, U.S. Secretary of Homeland Security Jeh Johnson issued a press release showing the statistical drop and saying that the U.S. government is “pleased that the Mexican government has itself taken a number of important steps to interdict the flow of illegal migrants from Central America bound for the United States” (Johnson 2014).

In parallel, apprehensions of Central American migrants in Mexico rose significantly – by 75 percent, from 53,078 to 92,889 (Secretaria de Gobernacion 2013, 117; Secretaria de Gobernacion 2014, 117; Central America’s Unresolved Migrant Crisis 2015). Mexico’s National Immigration Institute (INM) reported a 79 percent increase in the number of Central Americans deported from Mexico in the first four months of 2015 (Stillman 2015).

The U.S. government has characterized the efforts to stem the flow of Central American migrants to the U.S. border as a success, and has given much of the credit for this success to Mexico for its enhanced enforcement efforts. As President Obama said at a joint White House press conference with Mexican President Peña Nieto on January 5, 2015:

I very much appreciate Mexico’s efforts in addressing the unaccompanied children who we saw spiking during the summer.  In part because of strong efforts by Mexico, including at its southern border, we’ve seen those numbers reduced back to much more manageable levels.

U.S. externalization efforts extended to the Central American source countries as well. In June of 2014, Honduran law enforcement units which had received funding and training from the U.S. State Department’s Bureau of International Narcotics and Law Enforcement (INL) “launched an operation to intercept children and families attempting to cross the border from Honduras into Guatemala” (Jesuit Conference 2014, 2). These Special Forces apparently units collaborated on two tactical operations, Operation Rescue Angel and Operation Coyote (ibid.). According to reports, all three units received equipment and special training from U.S. Border Patrol, U.S. Immigration and Customs Enforcement or other U.S. migration control and law enforcement entities (see ibid.; Honduras amuralla 2014; Carcamo 2014a; Carcamo 2014b).

Australia

Australia’s migration-control externalization policy, called the “Pacific Solution,” dates from August 2001 when the Australian navy began interdicting migrants on the high seas (Hathaway 2002; Magner 2004). That month, a Norwegian freighter, the Tampa, rescued migrants from an overcrowded vessel and sought to bring them to Australia’s Christmas Island, as the next “place of safety,”—the proper place, according to international maritime law, to disembark people rescued from vessels in distress.[33]After the Norwegian ship captain defied Australia’s order and entered its territorial waters, the Australian military blocked the Tampa off the coast of Christmas Island, prevented its passengers from disembarking, and transferred them to an Australian military vessel.

During the weeks-long stand-off at sea, the Australian Parliament passed legislation with the explicit objective of stemming the unauthorized maritime arrival of asylum seekers (HRW 2002, 12-13). Among other provisions, the legislation, which came into effect in late September 2001, excised Christmas Island and other outlying territories from Australian immigration law, thus prohibiting asylum seekers in these erstwhile Australian territories from lodging asylum claims in Australia (HRW 2002, 12-13).

While there is evidence to suggest that Australian officials examined the U.S. playbook in devising the Pacific Solution and in many respects “followed in the footsteps of the United States in its restrictive policies on asylum seekers who arrive by sea,” they also added some new variations (Magner 2004, 58). Australia’s new twist on externalization excised Australian territories, for immigration law purposes, by a stroke of the pen rather than through any extraterritorial actions, creating the sort of “rights free zone” that had proved so convenient for the United States at Guantánamo (HRW 2002, 63).

During September and early October 2001, Australia detained irregular maritime arrivals on Christmas Island or on board Australian naval ships and then transferred them to the Pacific-island country, Nauru, or to Papua New Guinea, where they were confined in camps. (HRW 2002, 12-13). At the time, Nauru was not a party to the 1951 Refugee Convention and Papua New Guinea, though a party to the Convention, had attached numerous reservations to their accession and did not have in place adequate laws and structures to provide effective protection to refugees and asylum-seekers (UNHCR 2010). Australian immigration officials conducted refugee status determinations in Nauru and Papua New Guinea but not for the purpose of granting asylum under Australian law, but rather—as the United States was doing at Guantánamo—to consider recognized refugees for discretionary resettlement; the asylum seekers in Nauru and Papua New Guinea had no recourse under any national law to challenge the refugee status determinations (HRW 2002, 62-65).

In late October 2001, Australia also began interdicting migrant boats and forcibly returned them to Indonesian waters. Indonesia was notified, but no specific reception arrangements were made and no agreement was in place guaranteeing the protection of the intercepted refugees from refoulement (HRW 2002, 12-13).

Refugee processing was closed at Manus Island in 2004 and at Nauru in 2008. Nauru became a party to the Refugee Convention in 2011, but lacked the capacity to provide effective asylum procedures and refugee protection (UNHCR 2012b, 38).[34]

In July 2011 Australia announced an “arrangement” to transfer irregular maritime asylum seekers to Malaysia, but Australia’s High Court halted that plan, finding it to be procedurally invalid because of Malaysia’s poor refugee protections and the risk of refoulement (Phillips 2014).^[35] The High Court ruling ended the Malaysia deal, but the Parliament scrapped the provision of the Migration Act upon which the High Court had based its ruling, and switched its offshore processing scheme away from Malaysia, not a party to the Refugee Convention, and back to Nauru and Papua New Guinea, both nominally parties to the Convention. 

In August 2012, the Offshore Processing and Other Measures Bill authorized the government to transfer irregular migrants arriving by sea to Nauru or to Manus Island, Papua New Guinea, where they would be held indefinitely while their refugee claims were processed (HRW 2012). The legislation only targeted asylum seekers who arrived irregularly by boat. (ibid) The claims of asylum seekers who arrived by air, even with improper documents, would continue to be processed while they remained in Australia under the law. (ibid) On July 19, 2013 Prime Minister Kevin Rudd announced that all irregular maritime asylum seekers would be sent to Manus and Nauru for processing and that none found to be refugees would be resettled to Australia (Parliament of Australia 2014).

In December 2014, the Australian Parliament passed legislation that removed most references to the Refugee Convention from Australia’s Migration Act of 1958, instead creating a “new, independent and self-contained statutory framework” to allow Australia to pursue its own interpretation of its obligations under the Convention (KCIRL 2014). Among other things, the legislation also gave the Minister of Immigration and Border Protection extraordinary powers to detain people on the high seas and to transfer them to any country chosen by the minister and with very limited possibilities for judicial review (ibid). The new law added that “the designation of a country to be an offshore processing country need not be determined by reference to the international obligations or domestic law of that country” (ibid). Prime Minister Tony Abbott promoted his “stop the boats” policies as a model for other countries confronting the challenge of irregular migration, including European countries (see Norman 2015).

Australia has continued and expanded the Pacific Solution under the Abbott Government (ASRC 2015). This expansion includes a regional plan by Australia, Indonesia, Sri Lanka, and Malaysia to deter migration by sharing intelligence and information about the identities of migrants, cooperating on naval patrols and border security, launching media campaigns to dissuade migration, and increasing the speed at which migrants are deported (The Coalition 2014). The Abbott and Turnbull Governments’ part in this plan has been directed by the military-led Operation Sovereign Borders, which includes a joint agency task force aimed at preventing “people smuggling” (KCIRL 2015). Under Operation Sovereign Borders, migrants who are not returned to their “sending” country continue to be detained for offshore processing in substandard conditions (Australia asylum 2016; Ostrand 2014).

The guiding principle of Australia’s Pacific Solution has been the “no advantage” test, to ensure that “no benefit is gained through circumventing regular migration arrangements.”[36] As a consequence of this approach, Australia has refused the option of resettlement in Australia to any refugees from Nauru or Papua New Guinea. Its search for third countries willing to resettle refugees from its outsourcing centers led it to conclude a deal with Cambodia in 2014 to accept refugees from Nauru. The resettlement agreement, coupled with $40 million in development aid, ignored concerns about safety and the lack of capacity of the Cambodian government (Human Rights Watch 2015). The first group of four refugees arrived in Phnom Penh in June 2015 at a cost to Australia of another $15 million, but by March 2016 three of the original four had chosen to go back to their home countries rather than remain in Cambodia (Hasham 2016). With two additional refugees resettled from Nauru, there were only three refugees still in Cambodia, as of March 2016 (ibid). 

In April 2016, the Supreme Court of Papua New Guinea ruled that detention of asylum seekers and refugees on Manus Island violated PNG’s constitution and the PNG prime minister then called on the Australian government to find alternative arrangements of its detainees on Manus Island and to relocate refugees not wanting to stay there (HRW 2016d).

The European Union

In 2003, British Prime Minister Tony Blair’s cabinet and Home Office circulated a policy paper called “A New Vision for Refugees,” which proposed that the European Union establish Regional Protection Areas (RPAs) near refugee-producing countries, which would have the purpose both to contain refugees in countries of first arrival and to serve as places to which asylum seekers that had arrived in Europe could be deported (Noll 2015). As proposed, returned asylum seekers as well as refugees arriving from their home countries would be processed in Transit Processing Centers (TPCs) for possible resettlement in the European Union (ibid). Finding insufficient support for the proposal, the UK withdrew it prior to the June 2003 European Council meeting in Thessaloniki, and it was never formally considered (HRW 2006, 5). But the New Vision has persisted through the years and echoes loudly in the 2016 EU-Turkey Action Plan for stemming the irregular flow of migrants and asylum seekers into the European Union.

In September 2005, the European Commission proposed Regional Protection Programmes (RPPs), a variation on the New Vision’s RPA, which purportedly would strengthen protection capacity in the regions close to refugee flows (European Commission 2005). Human Rights Watch noted at the time that while “the RPPs’ goals of strengthening the protection capacity and improving access to durable solutions in the target countries are laudable, [t]he RPPs concept raises concerns . . . that the EU will use the existence of such programs as a pretext to declare the target countries ‘safe third countries.’ The EU could then return asylum seekers and migrants who transited through these countries even though effective protection could not be guaranteed” (Human Rights Watch 2006, 4).

During the next ten years, the EU continued its efforts to put externalization into practice through cooperation with key migration transit countries. In Ukraine, at the time an important transit country for migrants and asylum seekers, the EU and its member states bordering Ukraine provided resources, signed readmission agreements to facilitate summary returns, and conditioned closer ties on Kiev’s cooperation with migration-control efforts that led to asylum seekers being warehoused in the country (HRW 2010). In Libya, EU member state Italy led the way in encouraging the Gaddafi government to prevent migrants from departing its shores by boat for Italy and Malta and to accept the summary return of those who did depart, despite the abuses against migrants and asylum seekers and lack of effective protection in that country (HRW 2009).

The EU has also pursued a series of EU-wide readmission agreements with third countries to facilitate not only the return of rejected asylum seekers and other migrants who are nationals of those countries but in the case of transit counties, the nationals of other countries, a move designed in part to overcome the practical difficulties of deporting rejected asylum seekers to some countries (Rais 2016).

The “safe third country” concept was codified in Europe in the Asylum Procedures Directive (APD), one of the key elements of the EU’s Common European Asylum System, in 2005.[37] After the Asylum Procedures Directive passed, UNHCR said that it “may lead to breaches of international refugee law if no additional safeguards are introduced, and make it harder for refugees to have their asylum claims properly heard in Europe.” After the European Commission proposed a recast of the APD in 2009,[38] the UN refugee agency in January 2012 said:

“UNHCR continues to question the utility and consistency with international refugee law of the European safe third country concept. No minimum principles and guarantees appear to govern the procedure under Article 39 of the Amended Proposal. The implication is that access to territory and to an asylum procedure may be denied altogether to asylum seekers who may have protection needs. Such a denial could be at variance with international refugee law” (UNHCR 2012a).

The political impact refugee crisis in Europe in 2015, with the arrival of more than 1 million people by sea, 84 percent of whom came from the world’s top ten refugee-producing countries, gave fresh impetus to the EU’s interest in externalization (UNHCR 2016e). The EU first sought to shift responsibility towards the non-EU states on the Western Balkan migration route and then ultimately to Turkey, the main transit country for the 850,000 arrivals in Greece in 2015. At the same time, it renewed its efforts to create migration partnerships with African countries, establishing an emergency trust fund for Africa in November 2015 with the explicit aim of reducing migration to Europe (EC 2015).

In March 2016, the EU and its member states used inducements of future visa-free travel for Turkish nationals, a renewed path toward EU membership, and billions of euros in aid to seek Turkey’s assistance in preventing Syrian refugees from crossing into the EU (EC 2016a). On March 18, 2016 the EU and Turkey came to a formal agreement to return “all new irregular migrants crossing from Turkey into Greek islands” back to Turkey (ibid). The agreement states that such returns will take place “in full accordance with EU and international law.” The agreement cites the APD, and suggests that asylum requests from Syrians, Iraqis, Afghans, and others arriving irregularly by boat from Turkey will be ruled inadmissible because Turkey will be identified as a “safe third country” or as a “first country of asylum” to which asylum seekers can be returned with less than full examination of their asylum claim (ibid.).

At this writing the EU-Turkey migration agreement is still in its infancy, but Turkey in any case falls short of the APD’s criteria for either “safe third country” or “country of first asylum.”[39] Importantly, any Syrian, Iraqi, or Afghan returned to Turkey would not be allowed to request refugee status there because Turkey excludes non-Europeans from qualifying for refugee status (Kirişci 2005). Syrians have access to temporary protection in Turkey offering some access to assistance and more recently to the labor market, though, in practice significant barriers remain (HRW 2016e). But Iraqis, Afghans, and other nationalities of asylum seekers are not even eligible for temporary protection. The directive also requires the safe third country to respect the principle of non-refoulement, the prohibition on the forced return of refugees.[40] That principle not only forbids governments from deporting refugees to places where their lives or freedom would be threatened but also from rejecting asylum seekers at their borders who would face such threats. At the very time the EU was announcing its migration control deal, Turkey had closed its border to tens of thousands of Syrians fleeing a military offensive in the northern Syrian city of Aleppo (HRW 2016a).

As of this writing, the deal’s deterrent effect coupled with the closure of the Western Balkan migration route appears to be working; the number crossing the Aegean has dropped dramatically (UNHCR 2016a). But if no Syrians reach Greece by boat and the war in Syria drags on, Turkey will be left with an ever-growing refugee population. And while the deal envisions voluntary resettlement of Syrians from Turkey to the EU in exchange for the return of Syrians from Greece to Turkey, the fate of non-Syrians is far more uncertain.

Whatever the outcome of the EU-Turkey deal, it is clear that externalization is now the main plank of EU migration policy. This was emphasized in June 2016, when the European Commission announced a new Migration Partnership Framework (EC 2016b). While there are positive elements in the framework, including the possibility of greater resettlement and other legal routes for migration, the overall approach is consistent with the EU’s efforts during the refugee crisis to deflect responsibility and legal obligations away from EU member states and onto transit and origin countries, including through the EU-Turkey deal. The short-term objectives of the framework are to “save lives in the Mediterranean Sea; increase the rate of returns to countries of origin and transit; and “enable migrants and refugees to stay close to home and to avoid taking dangerous journeys.” (ibid.)

While much of the framework restates long-standing policy approaches, the emphasis on economic development as the main tool to tackle root causes of migration, the choice of potential partners including governments, like Sudan and Eritrea, with poor records when it comes to protecting human rights, and the effort to condition general EU aid explicitly on migration cooperation are worrying elements that will require close scrutiny in the months and years to come.

Part IV: Recommendations for Promoting Government Migration Policies Protective of Human Rights

Laws, policies and practices that externalize migration controls are rights-threatening when they have the intention or effect of blocking access to territory in ways that frustrate the right of any person to leave their country of nationality or any other country, or the right of anyone to seek asylum outside their country. They are also rights-threatening when they have the effect (or the intention) of causing or making more likely violations of other human rights of migrants, including asylum-seekers. There is ample scope for destination or transit countries to engage in capacity building and to assist with migration management, including as part of supporting security screening and law enforcement, in ways that are consistent with international human rights law and standards. Such laws, policies and practices should be reformed to systematically and directly incorporate rights protections into migration management (including through increasing the protective capacity of authorities involved in migration control in third countries) as well as to condition destination- and transit-State funding, training and other assistance to third countries on the implementation of a minimum set of human rights protections in law and in practice.

Rather than promoting the externalization of migration controls, States and other donors could also pursue increased support to regional and international organizations that provide or promote the protection of migrants’ rights, including the rights of asylum seekers, in third countries and countries of origin. Such parallel resources could be directed to enhancing the capacity of all States along migration routes, including through robust training of immigration and border security officials in human rights and refugee protections. These could in particular include advocacy and training efforts to ensure that officials without a human rights or refugee protection mandate, but whose work triggers or follows apprehension and processing of migrants, such as border guards and law enforcement officials, understand and respect the rights of all of the migrants they encounter.

Finally, States and other donors could provide increased support to civil society actors whose work directly or indirectly supports protection of the basic needs and rights of migrants, refugees, and asylum seekers, including to humanitarian protection. These efforts could be targeted to civil society actors well-positioned to advocate for migrants’ and asylum seekers’ rights. It is particularly important that international human rights organizations and civil society actors monitor the externalization of border control. There are several ways in which this can be done. To begin, civil society groups should educate each other on government practices in their regions. Civil society can seek and disseminate information about government practices and regional dynamics, including through civil society platforms and networks, and organize and engage with international and regional human rights bodies that monitor States’ compliance with human rights laws. Civil society groups can also directly engage with law enforcement efforts and seek to ensure migrants’ and asylum seekers’ rights are not violated in the context of the prosecution of alleged criminals. Civil society groups can also engage in documentation and monitoring of compliance with international law and standards, promoting the incorporation of international human rights law into domestic law and efforts to promote accountability for those migration-control externalization policies and practices that violate human rights – and can bring legal challenges to border externalization in domestic courts and international and regional human rights bodies whenever it contravenes international or national law.

The following are specific policy recommendations targeted to States and corresponding to the Regional Case Studies in Section III above:

To the United States

1. Stop the high seas interdiction, shipboard screening, and summary returns of irregular maritime migrants and asylum-seekers in the Caribbean or elsewhere. Do not outsource screening and refugee processing to the US Naval Base at Guantánamo Bay, Cuba, but rather bring all individuals expressing a need for international protection to the territorial United States to have their claims assessed.
2. Redirect funding and support for Mexico’s Southern Border Program away from interception of Central American migrants and asylum seekers and to improving Mexico’s capacity to protect, receive and process asylum seekers, including by improving its capacity to provide them social support while their claims for refugee status or other forms of protection are pending, and to integrate recognized refugees.
3. Condition any US funding of third countries engaged in immigration and border control on their demonstrated compliance with national and international human rights standards as well as anti-corruption measures.
4. Do not use US aid money to increase border militarization aimed at stopping migrants from leaving their home countries in search of international protection.

To Australia

1. End the policy of offshore detention and processing of asylum seekers in Nauru, Manus Island, or in any other sites that unable or unwilling to provide effective rights protections to asylum seekers.
2. Stop the policy of turning back migrants and asylum seekers arriving by boat to countries of origin or countries of transit, such as Indonesia, Sri Lanka, and Vietnam, after cursory screening.
3. Allow irregularly arriving maritime asylum seekers to have their claims for protection heard in Australia and provide asylum in Australia for those found to be deserving of international protection.
4. Offer the possibility of settlement in Australia to all refugees currently in Nauru and Manus Island.

To The European Union

1. Do not designate or otherwise treat Turkey (or any other country) as a safe third country or as a first country of asylum for the return of asylum seekers unless it meets the following principles:
   1. The country in law and practice provides for refugee status or a status that is equivalent to that required by the 1951 Refugee Convention or its 1967 Protocol to people of all nationalities deserving of international protection;
   2. The country provides a legal regime, such as a temporary protection regime, to ensure that migrants are not returned to a country experiencing armed conflict and indiscriminate violence where they face a risk of serious harm and that carries with it the possibility of refugee status, including for situations that become protracted;
   3. The country protects labor rights of lawfully present refugees and in particular the right to work, such as by providing work authorization and access to the labor market for recognized refugees as well as beneficiaries of temporary protection;
   4. The country protects the rights to health and to education for recognized refugees as well as beneficiaries of temporary protection, or provide access to social services;
   5. The country protects the rights of liberty and security of recognized refugees as well as beneficiaries of temporary protection and refrains from detaining asylum seekers, particularly children, on the grounds of illegal entry or stay.  
   6. The country demonstrates respect for the principle of non-refoulement, including prohibiting any practice that could have the effect of returning an asylum seeker or refugee to places where they would face serious risk of violation of fundamental human rights, where their lives or freedom would be threatened, or where they would be at risk of persecution, torture, or cruel, inhuman or degrading treatment or punishment, including through interception, rejection at the frontier, or indirect refoulement. 
2. Do not use the Turkish or Libyan or other third country coastguards or border guards to prevent would-be asylum seekers from exercising their rights to leave any country or to seek asylum.
3. Do not make assistance through the EU-Africa Trust Fund, the Khartoum Process or any other migration compact under the EU Partnership Framework conditional on African or other countries preventing individuals from leaving their countries of nationality or habitual residence, but rather fund their need to build capacity to respect and protect human rights and to address development needs.

Abrogate the the unfair and unworkable Dublin Regulation system for determining EU Member State responsibility for examining asylum claims based on the country of first arrival and in its place institute a mechanism for assigning responsibility for examining asylum claims to Member States according to each state’s assessed capacity, while taking into consideration the dignity, autonomy, and wishes of asylum seekers and rights to family unity and reunification.