Outsourcing Refoulement: The United States and the Central American Refugee Crisis

Written by Aaron Korthuis

 1. Introduction

In recent years, the U.S. immigration debate has been defined by a refugee crisis. The crisis contains two types of refugees: the refugees seeking protection abroad, like Syrians, and the asylum seekers crossing the U.S. southern border, who are primarily Central American. This piece looks at that latter crisis – the one that has in part provoked presidential candidate Donald Trump to promise a wall time and time again to his cheering supporters.

However, Donald Trump notwithstanding, the United States has already helped to build a virtual wall at the Mexican southern border. In recent years, as immigration flows from Central America dramatically increased due to gangs and drug violence, the United States rapidly expanded its support for and pressure on Mexico to enforce its southern border. Soon after the Central American refugee crisis became a major political issue, Mexico took over the work of arresting, detaining and deporting the refugees fleeing north – deporting tens of thousands of Central Americans, and using U.S. funding to do so.

This piece addresses the implications of state efforts to displace legal obligations by outsourcing their refugee problems to other states. It does so by examining U.S. practices in response to the Central American refugee crisis. I argue that the United States’ new approach to displacing its legal obligation by paying Mexico to do its enforcement work violates the Refugee Convention. Further, I suggest that victims of these new efforts to avoid the Convention’s obligation should seek to challenge such action in the legal forums available to them.

2. The Central American Refugee Crisis

In the summer of 2014, a growing problem that had brewed on the United States’ southern border for years erupted onto the scene of America’s tumultuous immigration debate. Tens of thousands of Central American migrants – many of them children – crossed the border in unprecedented numbers.[1] Department of Homeland Security (DHS) officials struggled to respond to this unique new problem, taken aback by the numbers and unable to quickly deport the new entrants. Yet just as quickly as the surge began, apprehensions at the U.S. border declined in late 2014, a trend that continued into 2015.[2] The Border Patrol apprehended 42% fewer unaccompanied minors and fewer family units in FY 2015 when compared with FY 2014.

Statistics from the Mexican border, however, belied claims that the surge in Central American migrants ever subsided. Soon after the crisis broke in U.S. news media, Mexico dramatically increased its own enforcement, leading to the apprehension of huge percentages of Central Americans long before they reach the United States. Thus, the crisis—and the refugees central to the crisis—simply moved off of the U.S. border, out of sight, and of mind.

Before turning to this problem on the U.S. and Mexican southern borders, it is important to first understand the nature of the current migrant crisis. Central American migration presents a unique problem, because many of those leaving their countries do so as refugees. The countries they leave from—El Salvador, Guatemala, and Honduras—have some of the world’s highest homicide rates.

These high violent mortality rates stem from two primary causes: drugs and gangs. Local drug groups shuttle cocaine north through the weakly governed Northern Triangle, and competition for routes among drug trafficking groups can easily spark violence as groups vie for the lucrative gains to be had by transporting cocaine. Meanwhile, gangs have torn apart most the region’s large urban centers. The extreme violence that gangs can wreak is best demonstrated by El Salvador, where a temporary gang truce in 2012 led to a dramatic fall in homicides – as well as a terrible surge in violence once the truce collapsed. Gangs enforce their local reign through threats of terror, exacting violence and punishment on those that fail to pay or respect their dominance.

This violence has prompted the flight north of thousands of Central Americans. Central Americans’ fears of the violence in their homelands is well-documented. For example, in 2014, a United Nations High Commissioner for Refugees (UNHCR) study of children fleeing north concluded that “58% . . . [of] children interviewed were forcibly displaced because they suffered or faced harms that indicated a potential or actual need for international protection.”[3]

3. The Emerging Paradigm: Outsourcing Refoulement

Nevertheless, despite the obvious protection needs, the U.S. has responded with efforts to make such migration impossible. Detention of asylum-seeking families, the application of expedited removal procedures for asylum-seekers, and raids targeting youths and families already ordered removed from the United States have become hallmarks of the U.S. response since the crisis’s inception. Yet one key aspect of the response is often overlooked. In July 2014, just as the unaccompanied and family migrant crisis peaked in numbers and in media fervor, Mexican President Peña Nieto launched the Program Frontera Sur (the “Southern Border Program”). Today, this program – which is in part financed by the United States – presents one the most severe humanitarian crises in the Western Hemisphere.

In recent years—the years during which Central American migration rapidly increased—the U.S. has provided increasing financial and political support to strengthen Mexico’s southern border enforcement.[4] For example, in fiscal year 2015, Congress approved a budget of $79 million more than the administration requested for the Merida Initiative (a U.S. funding stream for Mexican security efforts); this money was “to be used for helping Mexico secure its southern border and implement justice sector reforms.”[5] This $79 million came in addition to the $86.6 million already appropriated for enhanced border enforcement.[6] According to the Congressional Research Service, Congress appropriated these large sums to support Mexico’s efforts beginning in 2013 to “implement[] a southern border security plan.”[7]

U.S. officials have made clear that increased Mexican border enforcement serves U.S. interests. For example, in January 2015, President Obama praised Mexico for increasing enforcement at the southern border to control the U.S.’s migration problem, saying, “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.”[8] President Obama recently reiterated his thanks to Mexico in a speech at the UN.[9] Other officials have also made this connection.

The United States’ considerable financial support and its high-level praise for Mexico’s southern border enforcement closely tie the United States to what occurs at the Guatemalan border. The timing of Nieto’s Program Frontera Sur (July 2014) suggests that United States has played a catalytic and sustaining role in Mexico’s southern border.[10]

The consequences of this support have been grave. Mexico’s Program Frontera Sur is designed to turn around Central Americans in their tracks. The practical result is that the United States, through its financial and political support, has exported, or “outsourced” its humanitarian crisis to Mexico. This policy presents grave concerns for the protection of human life and suggests a serious failure to comply with international law.

The degree to which Mexico intensified migrant seizure and detention after launching Program Frontera Sur is evident in Mexico’s own border enforcement statistics. From October 2013 to April 2014—right before the “crisis” news exploded in the United States—Mexico apprehended 49,893 Central Americans.[11] During the same period a year later, from October 2014 to April 2015, Mexico detained 92,889 Central Americans, nearly double the prior year.[12] At the same time, U.S. apprehensions between these two periods fell by 92,303 persons, from 162,751 to 70,448.[13] The conclusion is unmistakable: Mexican southern border enforcement following the 2014 crisis dramatically reduced U.S. apprehensions, because fewer Central Americans reached the U.S. border.

The mass apprehension and deportation of Central Americans suggest that Mexico is engaged in a large scale violation of the Refugee Convention. Two reasons compel this conclusion. First, Mexican asylum statistics demonstrate the country’s lack of capacity to identify and process asylum claims. Indeed, in 2014, a year in which 107,814 people were deported,[14] only 451 people were granted refugee status (and 79 were given “complimentary protection”) out of 2,137 original applicants.[15] Second, Mexico’s procedures for identifying protection needs and processing applications are demonstrably inadequate. To give one example, credible reports exist that Mexico distributed pre-filled screening forms that state an individual does not have a fear of being returned to their home country.[16] It is also undisputed that the procedural protections afforded to Central American migrants in Mexico pale when compared to protections afforded in the United States, where serious concerns also exist about compliance with international refugee law. These procedural shortcomings, when contrasted with the high rate of successful screening done for refugees at the U.S. southern border, demonstrate unequivocally that Mexico is in engaged is returning refugees to their country of origin. The woeful procedural inadequacies also underscore the conclusion that the United States is supporting and encouraging the return of refugees at the Mexican southern border.

4. U.S. Liability for Its Support of Mexican Refoulement

The 1951 Convention Relating to the Status of Refugees[17] and the 1967 Protocol Relating to the Status of Refugees[18] form the bedrock of today’s international legal obligations concerning refugees. The Convention places the most important restriction in refugee law, non-refoulement, on signatory states. In Article 33, the Convention adopts the principal of non-refoulement: “No Contracting State shall expel or return (“refouler”) a refugee in any manner whatsoever to the frontiers of territories where his life or freedom would be threatened on account of [being a member of a protected class].”[19]

But can this obligation make the United States legally responsible for Mexico’s refugee law violations? The question depends first on the Refugee Convention’s extraterritorial application to state actions outside of areas where the state exercises de jure jurisdiction.

At first glance, there seems to be no plausible argument for such an expansive notion of refoulement. Indeed, in Sale v. Haitian Centers Council, Inc., the U.S. Supreme Court adopted an especially restrictive interpretation of Article 33, holding that the Refugee Convention does not “place[] any limit on the President’s authority to repatriate aliens interdicted beyond the territorial seas of the United States”.[20] Yet internationally, this interpretation has been largely discredited. The United Nations High Commissioner on Refugees, the Inter-American Commission on Human Rights, and the European Union, have all determined that Article 33 places the same legal obligations on the way that states handle refugees even where the state is acting outside it formal territory.

Similarly, numerous scholars have rejected the narrow interpretation provided by the Supreme Court in Sale. Guy S. Goodwin-Gill, a leading refugee law scholar, summarizes the consensus position: “The principle of non-refoulement [is] a rule of customary international law, the core element of which is the prohibition of return in any manner whatsoever of refugees to countries where they may face persecution. The scope and application of the rule . . . regulat[e] State action wherever it takes place, whether internally, at the border, or through its agents outside territorial jurisdiction.[21] Such a firm state responsibility is key to the effective operation of the Refugee Convention. As Professor James Hathaway notes, “[I]f states were able with impunity to reach out beyond their borders to force refugees back to the risk of being persecuted, . . . the entire Refugee Convention – which is predicated on the ability of refugees to invoke rights of protection in state parties – could, as a practical matter, be rendered nugatory.”[22]

But can this broad prohibition apply to U.S. support for Mexico’s refoulement violations? The Convention’s obligations certainly apply in the context of third-party processing agreements, where countries intercept refugees and bring them elsewhere to screen them for refugee claims. The basic idea in these situations is that where a state exercises control over an individual, it becomes responsible for what happens to the individual once they are transferred elsewhere.

These basic principles – the extraterritorial application of the Convention, and Convention’s obligations wherever a state exercises control over individuals – provide a colorable legal argument that the United States is in violation of the Refugee Convention for its support of Mexico. First, one might argue – for example, before the Inter-American Human Rights Commission (IAHCR) – that the United States has in fact exercised considerable control over the situation at the Mexican southern border. The United States has poured tens, if not hundreds, of millions of dollars into Mexican border security in recent years. Similarly, the United States exercised considerable leverage to move Mexico to launch Programa Frontera Sur. Mexican President Pena Nieto announced the program on July 8, 2014, at the height of the southern border crisis, and as noted above, its rollout immediately resulted in apprehension and deportation of tens of thousands of Central Americans.

Basic principles of legal causation might also lead an international body like the IAHCR to conclude that U.S. actions give rise to legal liability under the Refugee Convention. But for U.S. pressure and U.S. funding, many tens of thousands more Central American refugees would not face deportation in Mexico.[23] This would seemingly violate Article 33’s plain language that a state cannot “expel or return (“refouler”) a refugee in any manner whatsoever” back to the country where their life is endangered because of the refugee’s beliefs or membership in a group.[24]

Additionally, the U.S. might be liable for Mexico’s refoulment violations under a theory of aiding and abetting, a claim that a refugee could might bring in U.S. courts. Indeed, the scholarly consensus again supports a prohibition on aiding and abetting refoulement: “[A]ctions which amount to aiding, abetting, or otherwise assisting another state to breach Art. 33 are themselves in breach of the duty of non-refoulement.”[25] The concept of aiding and abetting has also been enshrined in international criminal law, which some U.S. courts have repeatedly recognized. For example, some U.S. courts have allowed claims to proceed against corporations based on a theory of aiding and abetting criminal actions undertaken by state leaders.[26]

Assuming a plaintiff could surmount problems of sovereign immunity when suing a state, aiding and abetting liability might allow a plaintiff to challenge the U.S. support for Mexico in one of two ways. First, a plaintiff might allege aiding and abetting non-refoulement in a U.S. court. This approach benefits from directly invoking Article 33 and the protections it affords. Nevertheless, aiding and abetting liability is largely confined to international criminal law, and may not extend to a cause of action rooted in Article 33. Further, a court might not agree with view that non-refoulement is customary international law that can give rise to a cause of action.

Second, a plaintiff might also invoke criminal law doctrine itself, and allege aiding and abetting of torture. Different issues might emerge here, such as mens rea, and how to assess that mens rea if the state is a defendant. However, this approach would benefit from the jus cogens status of a ban on torture, as well as the acceptance of aiding and abetting liability in criminal law.

5. Conclusion

The United States’ recent actions in support of Mexican border enforcement have put up yet another wall to refugees seeking to flee their countries of origin. This piece has suggested that U.S. policymakers should a serious look at these policies in light of international law to abandon its illegal and ill-informed support of Mexican southern border enforcement. I have also proposed that short of such a reversal, legal alternatives exist that may allow an international or domestic court to declare the U.S. support unlawful, or to award damages to an individual who suffered as a result of these policies. It remains to be seen whether the next administration is willing to dismantle this illegal, extraterritorial wall, rather than continue to build it up.


[1] Frances Robles, Wave of Minors on Their Own Rush to Cross Southwest Border, N.Y. Times (June 4, 2014), http://www.nytimes.com/2014/06/04/world/americas/wave-of-minors-on-their-own-rush-to-cross-southwest-border.html.

[2] U.S. Border Patrol Southwest Border Family Unit and UAC Apprehensions (FY 2014 – FY 2016), https://www.cbp.gov/newsroom/stats/southwest-border-unaccompanied-children/fy-2016 (last visited October 20, 2016).

[3] United Nation High Commissioner for Refugees, Children on the Run 6 (2014).

[4] See Clare Ribando Seelke & Kristin Finklea, Cong. Research Serv., RL41349, U.S.-Mexican Security Cooperation: The Mérida Initiative and Beyond, p. 16-17 (2015).

[5] Id. at 7.

[6] Id. at 16

[7] Id. See also Adam Isaacson et al., Washington Office on Latin America, Mexico’s Other Border: Security, Migration, and the Crisis at the Line with Central America 24 (2014), http://www.wola.org/sites/default/files/ Mexico%27s%20Other%20Border%20PDF.pdf.

[8] Remarks by President Obama, January 6, 2015, https://www.whitehouse.gov/the-press-office/2015/01/06/remarks-president-obama-and-president-pe-nieto-after-bilateral-meeting.

[9] David Nakamura, Obama Thanks Mexico for ‘Absorbing’ Central American Refugees. His Own Administration Wants to Turn Them Away, Wash. Post (Sept. 20, 2016), https://www.washingtonpost.com/news/post-politics/wp/2016/09/20/obama-thanks-mexico-for-absorbing-central-american-refugees-his-own-administration-wants-to-turn-them-away/.

[10] Id. (“Obama and Vice President Biden pressured Mexican President Enrique Peña Nieto to tighten his country’s borders with its neighbors and intercept those who made the journey north under the guidance of human smugglers.”).

[11] Press Release, Mexico Now Detain More Central American Migrants Than the United States, Washington Office on Latin America (June 11, 2015).

[12] Id.

[13] Id.

[14] Clay Boggs et al., New Data Show That Mexico Has Intensified Its Immigration Operations Without Building An Adequate Refugee Protection System, Washington Office on Latin America (June 24, 2015), http://www.wola.org/commentary/ new-data-show-that-mexico-has-intensified-its-immigration-operations-without-building-an-adequate-refugee-protection-system/.

[15] Comisión Mexicana de Ayuda a Refugiados, Estadisticas COMAR, http://www.comar.gob.mx/ es/COMAR/Estadisticas_COMAR (last updated Nov. 13, 2015).

[16] Ongoing Migration from Central America: An Examination of FY2015 Apprehensions Before the S. Comm. on Homeland Security and Governmental Affairs, 114 Cong. (2015) (statement of Duncan Wood, Director, Mexico Institute at the Woodrow Wilson International Center for Scholars).

[17] Convention Relating to the Status of Refugees, July 28, 1951, 19 U.S.T. 6259, 189 U.N.T.S. 150 [“1951 Convention”].

[18] Protocol Relating to the Status of Refugees, Jan. 31, 1967, art. I, § 2, 19 U.S.T. 6261, 606 U.N.T.S. 267.

[19] 1951 Convention, art. 33, sec. 1.

[20] 509 U.S. 155, 187 (1993).

[21] Guy S. Goodwin-Gill & Jane McAdam, The Refugee in International Law 248 (2007).

[22] James C. Hathaway, The Rights of Refugees Under International Law 163-164 (2005).

[23] See Restatement (Third) of Foreign Relations Law of the United States § 207 cmt. e (1987) (“There is no well-developed jurisprudence of causation in international law, but international law generally applies concepts of causation, including proximate cause, not unlike those in the law of the United States and other developed states.”)

[24] 1951 Convention, art. 33, sec. 1.

[25] James C. Hathaway et. al, The Michigan Guidelines on Protection Elsewhere, 28 Mich. J. Int’l Law 207, 213 (2007).

[26] Presbyterian Church of Sudan v. Talisman Energy, Inc. 582 F.3d 244 (2d Cir. 2009).

You may also like...

3 Responses

  1. yahdi siradj says:

    hi Aaron Korthuis, thank you for this great information The United States and the Central American Refugee Crisis ,thanks

  2. Emily Scarlett says:

    How recent is this piece?

Leave a Reply

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