Authors: Linda Camp Keith, Banks Miller, and Jennifer Holmes
Publication: Human Rights Quarterly 37: 153-187
Original source: hrq_2015.pdf
Conversion note: this Markdown was generated by text extraction from the hosted file. Review equations, tables, notes, and citations before treating it as a final web copy.
HUMAN RIGHTS QUARTERLY Human Rights Quarterly 37 (2015) 153–187 © 2015 by The Johns Hopkins University Press How Draconian Are the Changes to US Asylum Law? A Monthly Time Series Analysis (1990–2010) Linda Camp Keith,* Banks P. Miller,** & Jennifer S. Holmes***
How have recent changes to US asylum law altered who gets asylum? We investigate whether the changes wrought by the Illegal Immigration Reform and Immigrant Responsibility Act (IIRIRA) and the Real ID Act changed the decision making of immigration judges in asylum cases. We find, contrary to much of the commentary surrounding both IIRIRA and Real ID, that immigration judges became more likely to grant applicants asylum. Furthermore, we find that those applicants who are most at-risk of persecution in the countries they are fleeing are also the applicants most likely to be granted relief—a fact that became increasingly true with the implementation of IIRIRA and Real ID. I. INTRoDUCTIoN Crafting solutions to end state violence against its own citizens may present a nearly impossible task; however, international human rights law has * Linda Camp Keith is Associate Professor of Political Science at the University of Texas at Dallas. Her primary research interests are human rights and the rule of law, judicial decision making, and US asylum policy. ** Banks P. Miller is an Assistant Professor of Political Science at the University of Texas at Dallas. His primary research interests are judicial decision making, judicial politics and US asylum policy. ***Jennifer S. Holmes is Professor and Head of Public Policy, Political Economy, and Political Science at the University of Texas at Dallas. An earlier version of this article was written for the 2012 annual meeting of the American Political Science Association which was cancelled due to Hurricane Isaac. We would like to thank Will Moore for arranging an alternative cyber forum in which we were able to present the article. Our work here has benefitted from the insightful comments from Will and the other panel members.
Vol. 37 154 HUMAN RIGHTS QUARTERLY developed legal protections to end or prevent future persecution for a small set of victims or potential victims: those who are able to flee their state. Following World War II, in 1951 the United Nations General Assembly promulgated the Convention Relating to the Status of Refugees (hereafter the Convention), which sets out the rights and obligations of states in regard to refugees, including asylum and the norm of non-refoulement. 1 The core principle of non-refoulement underpins the international refugee regime. Generally speaking, this principle prohibits states from forcibly returning individuals who fear a return to their country of origin. To be eligible for protection, this concern must be based on a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group, or political opinion. 2 One hundred forty-seven states belong to the Convention or its Optional Protocol and millions of individuals have been afforded protection through international refugee law. 3 Since 1980 the United States has admitted almost 2.5 million refugees through its refugee resettlement system and has granted asylum to approximately 500,000 additional refugees. 4 While the general intent of the international refugee regime is protection of refugees, from the outset, the Convention and compliance with its obligations have reflected the political realities of the time. The refugee policies of Western receiving states have been criticized as being primarily driven by foreign policy interests, especially during the Cold War. Since the mid- 1990s, receiving countries have acted upon the increasing fear of economic opportunists abusing the system that has combined with the broader fear of potential terrorists gaining legal entry into the country through overburdened domestic asylum and refugee status determination systems. The United States and European states increasingly have taken deterrent and preventative actions to discourage individuals from choosing their country for asylum and 1. Convention Relating to the Status of Refugees, adopted 28 July 1951, U.N. Doc. A/ CONF.2/108 (1951), 189 U.N.T.S. 150 (entered into force 22 Apr. 1954). 2. Specifically, the Convention defines a refugee as a person who “owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country; or who, not having a nationality and being outside the country of his former habitual residence as a result of such events, is unable or, owing to such fear, is unwilling to return to it.” Convention Relating to the Status of Refugees, art. 1A(2), adopted 28 July 1951, U.N. Doc. A/CONF.2/108 (1951), 189 U.N.T.S. 150 (entered into force 22 Apr. 1954). 3. As of April 2011, the total number of States Parties to the 1951 Convention is 144; total number of States Parties to the 1967 Protocol is 145; the total number of States Parties to both the Convention and Protocol is 142. See UNHCR, States Parties to the1951 Convention relating to the Status of Refugees and the 1967 Protocol (2011), available at http://www.unhcr.org/3b73b0d63.html. 4. HUmaN RigHts FiRst , ReNewiNg U.s. C ommitmeNt to ReFUgee PRoteCtioN 8 (2010), available at http://www.humanrightsfirst.org/wp-content/uploads/pdf/30th-AnnRep-3-12-10.pdf.
2015 How Draconian Are the Changes to US Asylum Law? 155 to prevent potential applicants from even reaching their ports of entry. The far-reaching steps call into question these Western states’ compliance with international norms. The United States Congress has passed two major acts since 1995 to reform its asylum process in reaction to these fears. Both the 1996 Illegal Immigration Reform and Immigrant Responsibility Act (IIRIRA) and the 2005 Real ID Act were passed to prevent economic migrants and individuals who may pose security risks from entering the country on false claims of asylum. 5 Following 11 September, the US government has pursued prosecution for documents fraud among asylum applicants and aggressively enforced safe third-country requirements. Critics argue that these measures are draconian responses that are overly broad and jeopardize our commitments under international refugee law, and they fear that worthy applicants have been turned away at the border or denied asylum with increasing frequency, thus facing the real possibility of torture and other forms of persecution as they are returned home. In this study we draw upon the international relations compliance literature, in which a growing body of empirical work explores the gap between promise and practice in international human rights. In particular we join the small body of scholars who seek to understand the role of domestic actors in treaty compliance. To do so we turn to the primary actors who implement US asylum policy—immigration judges (IJs)—whom we conceive to be the street level bureaucrats who make the bulk of asylum decisions. The IIRIRA, which went into effect in April 1997 and 1998, most significantly added “a new ‘expedited removal’ procedure and ‘credible fear’ screening for asylum seekers arriving at ports of entry with no documents or with documents suspected to be false,” 6 increased the grounds for mandatory detention, and imposed “a one-year deadline on applying for asylum, delay in work authorization eligibility, prompt adjudication of asylum applications, expedited removal, and detention of asylum seekers.” 7 The Real ID Act, which went into effect in May 2005, imposed the requirement that the applicant must prove the ground on which the fear of persecution is established as the central reason for the persecution. 8 The Real ID Act is thought to make it easier for IJs to make adverse credibility determinations. We examined the effect of the two major terrorist attacks on US soil: the World Trade Center bombing in February 1993 and the 11 September 5. Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA), 1996, Pub. L. 104–208, Div. C, 30 Sept. 1996, 110 Stat. 3009–546; Emergency Supplemental Appropriations Act for Defense, the Global War on Terror, and Tsunami Relief, 2005, PL 109–13, 11 May 2005, 119 Stat 231. 6. DeboRaH e. a NkeR, Law oF a syLUm iN tHe UNiteD states 11 (2011). 7. Marissa Silenzi Cianciarulo, Terrorism and Asylum Seekers: Why the Real ID Act is a False Promise, 43 HaRv . J. oN Legis . 101, 110–11 (2006). 8. Id. at 116.
Vol. 37 156 HUMAN RIGHTS QUARTERLY attacks. We also control for political party of the executive, decreases in the US refugee program, national unemployment, and other institutional factors. We then apply the model to the monthly grant rates for two groups: first, all asylum applicants and second, only those applicants from states with the highest levels of state violence against its citizens. From a human rights perspective, we seek to understand whether these events and policy changes have increased the probability that US asylum decisions will protect those facing persecution back home or whether they have in fact increased the likelihood that asylum seekers will be returned back to a country where they are likely to be harmed. We find that both IIRIRA and Real ID raised the rate at which asylum seekers were granted relief and likely resulted in thousands of additional asylees being allowed to stay in the US. This increase in the rate at which relief is granted holds for both the general and at-risk populations, with Real ID having a particularly beneficial effect for those we have deemed at-risk. Given our data limitations, we cannot be certain that these two policy interventions did not have serious consequences for those refugees who were unable to enter the country or who were turned away at ports of entry. Nevertheless, conditional on applying for asylum and reaching the IJ stage of the process, IIRIRA and Real ID have unmistakably delivered a net benefit for those fleeing persecution. Furthermore, we find that 9/11 had a short-lived but substantively significant negative effect on the overall rate at which asylum-seekers were admitted, lasting no more than approximately one month and no effect on those applicants thought to be particularly at-risk. We also find that the World Trade Center bombing in 1993 had no discernible effect on the willingness of IJs to grant asylum. II. INTERNATIoNAL RELATIoNS PERSPECTIvE oN US ASYLUM obLIGATIoNS US refugee and asylum obligations originate in membership in the 1951 UN Convention Relating to the Status of Refugees (hereafter the Convention). The United States chose not to become a party to the original Convention (a policy decision we will discuss more fully in the section below) but subsequently did become a party to the Convention when it ratified its 1967 Optional Protocol. The Refugee Act of 1980 was promulgated to bring US statutory law into compliance with US obligations under international law. 9 Early empirical evidence generated by international relations (IR) scholars, however, has suggested that throughout the 1980s US asylum and refugee policy continued to be driven primarily by foreign policy concerns rather 9. See id. at 109–10; a NkeR, supra note 6, at 2.
2015 How Draconian Are the Changes to US Asylum Law? 157 than its human rights commitment. 10 Subsequent studies have also generally concluded that US economic and material interests influence grants more than human rights conditions. 11 Thus, there appears to be a potential gap between promise and practice, a gap that is similar to the one found throughout the larger literature on human rights treaty compliance. 12 The empirical work which has sought to understand the [dis]connection between commitment and compliance in regard to international human rights treaties has largely turned to two competing sets of theoretical explanations within the field of international relations: realism and constructivism. Realists perceive states as rational actors whose behavior is based primarily upon narrow self-interest. 13 State commitments to international humanitarian or rights norms are perceived as “cheap talk” that give way to more substantive state interests. 14 Conversely, constructivists emphasize the emergence and diffusion of international human rights norms through networks of domestic and transnational actors, who not only shape the discourse of international human rights but who also rally publics to convince states to formally accept and to adhere to these norms. 15 In addition, states are expected to comply with these norms because states have a propensity to comply with their legal obligations 16 or because they generally aspire to comply with the norm of pacta sunt servanda (agreements must be kept). 17 However, the world society approach suggests the possibility of a decoupling effect between treaty commitment and practice. 18 This approach perceives states to be embedded in an integrated cultural system that “promulgates cognitive frames and 10. For a discussion of refugee admissions as a function of foreign policy versus universal human rights, see o PeN boRDeR ? C LoseD soCieties ?: t He etHiC aL aND P oLitiC aL issUes (Mark Gibney ed., 1988); Mark Gibney & Michael Stohl, Human Rights and U.S. Refugee Policy, in id. at 151; Mark Gibney, Vanessa Dalton, & Marc Vockell, USA Refugee Policy: A Human Rights Analysis Update, 5 J. ReFUgee stUD . 33 (1992); g iL LoesCHeR & J oHN a. sCaNLaN , CaLCULateD kiNDNess ReFUgees aND a meRiCa ’s HaLF -o PeN DooR 1945– PReseNt (1998). 11. See, e.g., Lisa Hassan, Deterrence Measures and the Preservation of Asylum in the United Kingdom and United States, 13 J. ReFUgee stUD . 184 (2000); mattHew J. g ibNey , t He etHiCs aND PoLitiCs oF a syLUm : LibeRaL DemoCRa Cy aND tHe ResPoNse to ReFUgees (2004). 12. See, e.g., Linda Camp Keith, The United Nations International Covenant on Civil and Political Rights: Does it Make a Difference in Human Rights Behavior? 36 J. PeaCe ReseaRCH 95 (1999); Oona Hathaway, Do Human Rights Treaties Make a Difference? 111 yaLe L. J. 1935 (2002); Emilie M Hafner-Burton & Kiyoteru Tsutsui, Human Rights in a Globalizing World: The Paradox of Empty Promises, 110 a m. J. soC . 1373 (2005). 13. keNNetH w aLtz , t HeoR y oF iNteRN atioN aL PoLitiCs (1979); RobeR t o . keoHaNe , a FteR HegemoNy : CooPeRatioN aND DisCoRD iN tHe w oRLD PoLitiC aL eCoNomy (1984). 14. John J. Mearscheimer, The False Promise of International Institutions, 19 iNt ’L seCURity 5–26 (1994). 15. See, e.g., maRgaRet keCk & katHR yN sikkiNk , a Ctivists beyoND boRDeRs : a DvoC aCy NetwoRks iN iNteRN atioN aL PoLitiCs (1998); t He PoweR oF HUmaN RigHts : iNteRN atioN aL NoRms aND DomestiC CHaNge (Thomas Risse, Stephen C. Ropp & Kathryn Sikkink eds., 1999). 16. Lo Uis HeNkiN, How Natio Ns beHave : Law a ND Fo Reig N Po LiCy (1979). 17. Abram Chayes & Antonia Handler Chayes, On Compliance, 47 iNt ’L o Rg . 175, 185 (1993). 18. See Hafner-Burton & Tsutsui, supra note 12.
Vol. 37 158 HUMAN RIGHTS QUARTERLY normative prescriptions that constitute the legitimate identities, structures, and purposes of modern nation-states.” 19 Thus, with the proliferation of human rights treaties codifying human rights norms, states’ legitimacy or “good nation” identity is increasingly linked to the formal acknowledgment of these norms. 20 However, as Cole notes, many states join the traditionally weak human rights regime, “not out of deep commitment, but because it signals their probity to the international community” and thus “a decoupling is endemic to the human rights regime.” 21 This norms-versus-interest debate is reflected in IR-based asylum literature, although the link to the treaty compliance and the related empirical literature is typically overlooked by the empirical asylum literature, which has sought to identify whether norms or interests drive asylum outcomes. 22 More recently, the compliance literature has broadened beyond IR theory and has begun to consider the role of domestic institutions. The domestic institutions approach dismisses the assumption of a unitary state actor and instead recognizes the role of numerous domestic actors and institutions within the state (particularly in democratic regimes) which may affect the regime’s calculation of costs related to commitment and [non]compliance. Democratic electoral processes and legal institutions are seen as providing the public and other political actors with the tools and venues through which they can hold the regime accountable should it fail to keep its international commitments. 23 From a broader perspective, democratic affinity for the rule 19. Wade M. Cole, Sovereignty Relinquished? Explaining Commitment to the International Human Rights Covenants, 1966–1999, 70 a m. soC . Rev . 477 (2005); see, e.g., John W. Meyer, John Boli, Greg M. Thomas, & Francisco O. Ramirez, World Society and the Nation-State, 103 a m. J. soC .144 (1997). 20. Id.; Christine Min Wotipka & Francisco O. Ramirez, World Society and Human Rights: An Event History Analysis of the Convention on the Elimination of All Forms of Discrimination against Women, in t He g LobaL DiFFUsioN oF maRkets aND DemoCRa Cy (2007). 21. Cole, supra note 19, at 477; this argument has a parallel in the immigration gap theory posited by Cornelius and Tsuda who argue that international pressures can create gaps in policy between what the host state might intend and what is actually implemented; see Wayne A. Cornelius & Takeyuki Tsuda, Controlling Immigration: The Limits of Government Intervention, in CoNtRoLLiNg immigRatioN : a g LobaL PeRsPeCtive 13 (Wayne A. Cornelius et. al 2004). We turn to this theory in the next section. 22. Marc R. Rosenblum & Idean Salehyan, Norms and Interests in US Asylum Enforcement, 41 J. PeaCe Res. 677 (2004); Idean Salehyan & Marc R. Rosenblum, International Relations, Domestic Politics, and Asylum Admissions in the United States, 61 PoL . R es. Q. 104 (2008); Andy J. Rottman, Christopher J. Fariss & Steven C. Poe, The Path to Asylum in the US and the Determinants for Who Gets In and Why, 43 iNt ’L migRatioN Rev . 3 (2009); Linda Camp Keith & Jennifer S. Holmes, A Rare Examination of Typically Unobservable Factors in U.S. Asylum, 22 J. ReFUgee stUD . 224 (2009); Jennifer S. Holmes & Linda Camp Keith, Does the Fear of Terrorists Trump the Fear of Persecution in Asylum Outcomes in the Post-September 11 Era?, 43 Ps: P oL . sCi. 431 (2010); Linda Camp Keith, Jennifer S. Holmes & Banks Miller, Explaining the Divergence in Asylum Grant Rates among Immigration Judges: An Attitudinal and Cognitive Approach, 35 Law & PoL ’y 261 (2014). 23. Linda Camp Keith, International Principles for Formal Judicial Independence: Trends in National Constitutions and Their Impact (1976 to 1996), 85 JUDiCatURe 194 (2002); Eric
2015 How Draconian Are the Changes to US Asylum Law? 159 of law and respect for constitutional constraints and judicial processes in the domestic context arguably carry over to the international context and thus increase the likelihood that democratic regimes will honor their international legal commitments. 24 Assumptions of the domestic-institutions approach have received much more specific empirical attention than those of realist and rationalist theory; presumably because scholars can more readily observe and measure domestic institutional contexts than cost/benefit calculations or the diffusion of norms. However, as far as we know, no study has examined the actual behavior of actors on the ground implementing treaty commitments. Our contribution to this literature is to pull the focus to the street-level bureaucrat: the immigration judge (IJ). The IJ primarily enforces the granting of asylum and the protection of the norm of non-refoulement, which is one of the core components of the United States commitment under the Refugee Convention. III. ASYLUM AND REfUGEE STATUS UNDER INTERNATIoNAL AND US LAw Following World War II, in 1951 the United Nations General Assembly promulgated the Convention Relating to the Status of Refugees, which sets out the rights and obligations of states in regard to refugees. The core principle of non-refoulement underpins the international refugee regime. Generally speaking, this principle prohibits states from forcibly returning individuals who do not want to be returned to their country of origin owing to a wellfounded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group, or political opinion. While the general intent of the regime is protection of refugees, from the outset the Convention reflected the Eurocentric political realities of the time, and only provided legal protection to a limited set of refugees. To qualify, individuals had to be refugees as a “result of events occurring before 1 January 1951.” In addition, countries were able to exclude refugees from outside of Europe. To do so, they were allowed to make a “declaration when becoming party, according to which the words ‘events occurring before 1 January 1951’ are understood to mean ‘events occurring in Europe’ prior to that date.” 25 Over Neumayer, Asylum Recognition Rates in Western Europe: Their Determinants, Variation and Lack of Convergence, 49 J. CoNFLiCt Res. 43 (2005); Oona A. Hathaway, Why Do Countries Commit to Human Rights Treaties?, 8 J. Co NFLiCt Res. 588 (2007); Emilia Justyna Powell & Jeffrey K. Staton, Domestic Judicial Institutions and Human Rights Treaty Violation, 53 iNt ’L stUD . Q. 149 (2009); LiNDa CamP keitH , PoLitiC aL RePRessioN : CoUR ts aND Law (2012). 24. Beth A. Simmons, International Law and State Behavior: Commitment and Compliance in International Monetary Affairs, 94 a m. PoL . sCi. Rev . 819 (2000). 25. Office of the United Nations High Commissioner for Refugees, Convention and Protocol Relating to the Status of Refugees: Introductory Note, at 2 (2011), available at http:// www.unhcr.org/3b66c2aa10.html [hereinafter UNHCR Introductory Note].
Vol. 37 160 HUMAN RIGHTS QUARTERLY time the legal norm has expanded, in large part due to the efforts of the UN High Commissioner on Refugees, who successfully pushed for universal coverage for refugees. This universal refugee coverage was achieved with the promulgation of the 1967 Protocol, which eliminated the original geographical and temporal restrictions. 26 Today the right to non-refoulement is considered a part of customary international law and thus applies even to states that are not formally a party to the 1951 Convention or its Protocol. 27 In 2001, the state parties issued a declaration “reaffirming their commitment to the 1951 Convention and the 1967 Protocol” and “recogniz[ing] in particular that the core principle of non-refoulement is embedded in customary international law.” 28 As Loescher and Betts, Loescher, and Milner note, during the 1950s the United States considered refugee policy too important to its national security interests to permit the United Nations to control it. 29 For US policy makers, “the most important aspects of American refugee policy were maintaining international attention devoted to refugees from Communist countries, encouraging emigration from the Eastern bloc, and minimizing appeals for assistance funds to refugees.” 30 The United States, which had played a strong role in restricting the scope of international protection of refugees under the Convention, did not join the 1951 Convention. Instead, it initially sought to limit the functional scope of its primary UNHCR organ and created “two other U.S.-led organizations that were parallel to and outside the purview of the United Nations.” 31 Contrarily, the United States strongly supported and joined the 1967 Protocol.32 Copeland identifies multiple factors that led to “an easy and non-controversial” ratification process in the US Senate, including strong domestic support (evidenced by the support of eighty-six voluntary agencies) within the US Department of State, and compatibility with existing US law and practice. 33 The move clearly signaled a change from its previous unilateral Cold War policies; however, as Marissa Cian- 26. a LexaNDeR betts , g iL LoesCHeR & James miLNeR, t He UNiteD NatioNs HigH CommissioNeR FoR ReFU- gees (UNHCR): t He PoLitiCs aND PRaCtiCe oF ReFUgee PRoteCtioN iNto tHe t weNty -FiRst CeNtUR y 29 (Thomas G. Weiss ed., 2008). The UNHCR reports that the geographic restrictions have been maintained by only a “limited number of states.” UNHCR Introductory Note, supra note 25, at 2. 27. g Uy g ooD wiN -g iLL & JaNe mCa Dam , t He ReFUgee iN iNteRN atioN aL Law (3rd ed. 2007). 28. UNHCR Introductory Note, supra note 25, at 4. 29. Gil Loescher, UNHCR at Fifty: Refugee Protection and World Politics, in PRobLems oF PRoteCtioN : t He UNHCR, ReFUgees , aND HUmaN RigHts 3, 7 (Niklaus Steiner, Mark Gibney & Gil Loescher eds., 2003); betts , LoesCHeR & miLNeR, supra note 26, at 7, 21, respectively. 30. a LexaNDeR betts , g iL L oesCHeR & J ames miLNeR, t He U NiteD N atioNs H igH C ommissioNeR F oR ReFUgees (UNHCR): t He PoLitiCs aND PRaCtiCe oF ReFUgee PRoteCtioN 21 (2012). 31. Loescher, supra note 29, at 7; see also Emily Copeland, A Rare Opening in the Wall, in PRobLems oF PRoteCtioN , supra note 29, at 108–09. 32. Copeland, supra note 31, at 109–10. 33. Id.
2015 How Draconian Are the Changes to US Asylum Law? 161 ciarulo argues, US assent to the Protocol “did not have a significant impact on asylum processing” and until 1980 the US continued to “enforce the narrow parameters, low ceiling on approvals, and strict burden of proof” mandated by the 1952 Immigration and Nationality Act and its subsequent amendments 34 which form the basic source of domestic US asylum law.35 She argues that the 1980 Refugee Act which “sought to give statutory meaning to our national commitment to human rights and humanitarian concerns” and “ushered in a new era of refugee protection.” 36 The 1980 Refugee Act, which amends the Immigration and Nationality Act, was the first congressional act that specifically addressed refugees and asylum seekers. The statute repealed the 1952 Act’s geographical and political limitations and lifted numerical caps on the number of annual asylum grants. 37 Deborah Anker argues that Congress enacted the legislation “in order to conform provisions of U.S. law to the Refugee Convention.” 38 Specifically, it codified into US law the international definition of a refugee. The US Supreme Court in I.N.S. v. Cardoza-Fonseca noted that “if one thing is clear from the legislative history of the new definition of ‘refugee,’ and indeed the entire 1980 Act, it is that one of Congress’ primary purposes was to bring United States refugee law into conformance with the [Protocol].” 39 According to the House committee report, the Act represented an intention to emphasize and make paramount “the plight of refugees themselves, as opposed to national origins or political considerations.” 40 Cianciarulo argues that with the passage of the 1980 Act (along with the Supreme Court’s recognition of the implications of the Act in Cardoza-Fonseca), “asylum was no longer an ad hoc, marginal immigration procedure entirely subject to the whims of policy.” 41 However, she also observes that “over the next twentyfive years, asylum would emerge as a unique, complex body of law and a lightning rod for the national immigration debate, forcing the country to balance traditional humanitarian interests against weighty national security concerns.” 42 The 1980 statute also instructed the US Attorney General to establish uniform procedures for the treatment of asylum claims, presumably with the goal of producing consistent and fair outcomes across adjudicators. In general, there are two paths through which individuals may claim refugee status in the United States: either as a refugee or as an asylee. Both paths require that individuals fulfill the definition of refugee in the INA. 34. Cianciarulo, supra note 7, at 108. 35. See a NkeR, supra note 6. 36. Cianciarulo, supra note 7, at 109, 110. 37. Id. at 109. 38. a NkeR, supra note 6, at 17. 39. I.N.S. v. Cardoza-Fonseca, 480 U.S. 421, 436 (1987). 40. Mark Gibney, A “Well-Founded Fear” of Persecution, 10 HUm. Rts . Q. 109,111 (1988). 41. Cianciarulo, supra note 7, at 110. 42. Id. at 110.
Vol. 37 162 HUMAN RIGHTS QUARTERLY Generally speaking, refugees apply for status outside of the US, while individuals requesting asylum do so from within the US or upon arrival at a port of entry. Our primary focus is the asylum process. First, we address the administrative structure for asylum processing; next, we turn to the refugee process, which we believe may be interconnected with the asylum process. We believe a discussion of both in tandem is important because they are both potential routes for safety into the United States for those who fear persecution at home. A. overview of the US Asylum Structure and Immigration Judges Adjudication of asylum cases in the United States occurs entirely within the executive branch, with some eligibility for judicial review. The Executive Office for Immigration Review (EOIR), which was created in January of 1983 as a separate agency within the Department of Justice, is the primary tribunal that handles asylum cases for applicants who are in removal proceedings. EOIR’s immigration judges (IJs) make determinations through adversarial hearings. Justice Department regulations allow applicants (and the government) administrative appeal to the Board of Immigration Appeals (BIA). The BIA was established by the Attorney General and is housed within the EOIR. Federal law also allows applicants to appeal to the appropriate federal circuits, with some exceptions. 43 Potential asylees who are not in removal proceedings may make application for asylum through the US Asylum Office within the US Citizenship and Immigration Services (USCIS) within the Department of Homeland Security (DHS). If the asylum officer (AO) does not grant asylum, he or she refers the applicant for removal proceedings and the applicant will receive a Notice to Appear before an IJ and have the opportunity to present their case de novo before an IJ or they may choose to voluntarily depart. In this article we focus specifically on the immigration judge who functions as the equivalent of a “trial level” adjudicator in asylum cases, which are heavily fact-driven. We consider IJs to be the primary decision makers in the asylum process because they make the final determination for virtually all asylum applications. The BIA reviews only a fraction of cases. IJs, in effect, make the decisions in the difficult cases; AOs, on the other hand, grant easy cases and refer the rest to IJs. Immigration Judges are formally appointed by the US Deputy Attorney General; however, the Executive Office of Immigration Review (EOIR) and 43. See Stephen Legomsky, Restructuring Immigration Adjudication, 59 DUke L. J. 1634 (2010); Jaya RamJi -NogaLes , a NDRew i. sCHoeNHoL tz , PHiLiP g. sCHRag & eDwaRD keNNeDy, ReFUgee RoU - Lette : DisPaRities iN a syLUm a DJUDiCatioN aND PRoPosaLs FoR ReFoRm (2009).
2015 How Draconian Are the Changes to US Asylum Law? 163 the Chief Immigration Judge handle their hiring. 44 In most years there are more than 230 IJs serving in fifty-five immigration courts in twenty-seven states and Puerto Rico. IJs arguably have less formal independence than federal judges and potentially less independence than administrative law judges. Nonetheless, they maintain a high degree of independence. The Immigration and Nationality Act (INA) states that “in deciding the individual cases before them. . . . IJs shall exercise their independent judgment and discretion.” 45 Despite their lack of life tenure, IJs experience a good deal of autonomy in their decision making because of: 1) the large volume of cases they decide (approximately three times the number decided by a typical federal district court judge) 46; 2) the low probability of appellate review by the Board of Immigration Appeals (BIA) 47 and the federal circuit courts 48; and 3) the standard of appellate review—which is reasonableness—to which the IJs are subjected. This exceedingly low standard requires that the BIA find that the IJ reached unreasonable conclusions in order to reverse a decision where the facts of a case are reviewed for clear error by the BIA. In contrast, the application of the law is reviewed de novo. In essence, we perceive IJs as judges-as-bureaucrats, with ample discretion and broad civil service protections. 49 b. US Refugee Admissions Program Refugee quotas and regional allocations are set by the President, who consults with Congress. In addition to INA eligibility, refugees must come from 44. Current qualifications (set by the attorney general) only require that the candidates have seven years of prior legal experience. See Executive Office for Immigration Review, Fact Sheet: Executive Office for Immigration Review Immigration Judge Hiring Initiative, AILA InfoNet Doc. No. 10031261 (Mar. 11, 2010), available at http://www.aila.org/content/ default.aspx?docid=31511. 45. 8 CFR § 1003.10 (b). 46. In 2009, 250 IJs were tasked with deciding approximately 350,000 cases or 1400 cases per judge per year, compared to 400 cases per year per district court judges in the same year. 47. In FY 2007, the BIA received 36,606 appeals of 351,051 cases from the immigration courts in Fiscal Year 2006. In short, only about 10 percent of cases decided by IJs are reviewed by the BIA; see U.s. DePaR tmeNt oF JUstiCe , Fy 2013 statistiCs yeaRbook , V1 (2013). 48. In 2005 there were 12,873 appeals from the BIA to the federal circuit courts; see The Third Branch, BIA Appeals Still Significant Part of Federal Appellate Caseload, May 1997, available at http://www.uscourts.gov/news/TheThirdBranch/07-05-01/BIA_Appeals_Still_Significant_Part_of_Federal_Appellate_Caseload.aspx. If all of these appeals involved asylum (and many did not, although we are unable to determine how many), then the number of cases reviewed by federal appellate courts accounts for less than 5 percent of the 265,000 cases decided by IJs in 2005. 49. See, e.g., JoHN bReHm & sCott g ates . w oRkiNg , sHiRkiNg , aND sabotage : bUReaUCRatiC ResPoNse to a DemoCRatiC PUbLiC 192 (1997).
Vol. 37 164 HUMAN RIGHTS QUARTERLY a country that is of “special humanitarian concern to the United States” and not be resettled in another country or ineligible due to security, criminal, or other factors as determined by USCIS. 50 Since asylum is one of the two paths through which individuals can escape persecution and gain legal protection within the US, it is reasonable to expect that trends across the two paths will interact. In this article we are mostly concerned with introducing refugee inflow as a control variable in our models.
to annual refugee inflow from 1990–2010. 51 There is a striking relationship between the two series, which have (at the annual level) a correlation of -0.75. Each series is plotted on its own scale, with asylum cases on the left axis and refugee inflow on the right axis. Interestingly, between 2002 and 2007 asylum claims actually outnumber refugee claims, mostly due to the precipitous drop in total refugee admissions that occurred in the wake of 9/11. We can see from Table A1 in the Appendix that during the years following 2001, the actual ceiling set by the President varies little; for the first six years, the ceiling was set at 70,000 and the for the next three at 80,000. What is significant is the number of refugees actually arriving into the United States. In the two years following 9/11 the US only admitted a combined 55,000 refugees (the fewest in the program’s entire history) due the post-9/11 program review, 52 which strongly suggests a gap between the US commitment to and compliance with the Refugee Convention. While the number of refugees admitted have recovered somewhat, the numbers remain well below the ceilings, in part due to immigration-related security measures passed in response to 9/11 that require long security screening processes. 53 Andrew Schoenholtz argues that US officials perceived the refugee resettlement program as “being particularly vulnerable to security problems” and thus was the “first refugee protection casualty of the terrorist attacks”; on the other hand, the asylum system, which had been revamped in 1995 (under IIRIRA), did not suffer the same fate, and only began to receive attention from Congress three years after the attacks. 54 The overriding point we wish to emphasize with Figure 1, however, is the close relation- 50. DaNieL C. maR tiN & James e. yaNkay , DePaR tmeNt oF HomeLaND seCURity . o FFiCe oF immigRatioN statistiCs , ReFUgees aND a syLees : 2011 a NNUaL FLow RePoR t 1–2 (2012), available at http:// www.dhs.gov/refugees-and-asylees-2011. 51. Table A1, Appendix details the yearly flow of refugees into the US as compared to the rest of the world. 52. Donald M. Kerwin, The Use and Misuse of “National Security’’ Rationale in Crafting U.S. Refugee and Immigration Policies, 17 iNt ’L J. ReFUgee L. 749 (2005); Donald M. Kerwin, The Faltering US Refugee Protection System: Legal and Policy Responses to Refugees, Asylum-Seekers, and Others in Need of Protection, 31 ReFUgee sURvey Q. 1, 14 (2012). 53. Kerwin, The Faltering US Refugee Protection System, supra note 52, at 14. 54. Andrew I. Schoenholtz, Refugee Protection in the United States Post-September 11, 36 CoLUm . HUm. Rts . L. Rev ., 323, 324 (2005).
2015 How Draconian Are the Changes to US Asylum Law? 165 ship between refugee inflow and the caseload faced by IJs: as the number of refugee admitted decreases, the number of asylum cases increases, and this may especially be the case because the resettlement program took a hit after the 9/11 attacks. We examine this relationship more in our models below but turn next to global trends in asylum policy. Iv. GLob AL TRENDS Tow ARD DETERRENCE AND PREvENTIoN Since the mid-1990s the United States and its European counterparts have increasingly taken more deterrent and preventative actions to discourage individuals from choosing their country as a target for asylum and to prevent potential applicants from even reaching their ports of entry. Schuster finds there to be a convergence among Western European countries’ asylum policies towards the lowest common denominator. She recognizes three restricting practices: “restricting access to the state’s territory, restricting access to welfare as a means of discouraging applications, and the substitution of temporary protection for permanent asylum.” 55 These policies have included
Vol. 37 166 HUMAN RIGHTS QUARTERLY various means of preventing asylum seekers from gaining access to the states’ territories such as redefining asylum seekers as civil war refugees (Germany, Italy, the Netherlands, and Sweden), amending the constitution to create a buffer zone (Germany), and taking preventative steps such as sending Airline Liaison Officers to certain countries to check passenger documents before take-off. 56 Similarly, Anita Böcker and Tetty Havinga point to the example of Belgium’s short-lived “2 x 5 percent rule” in which “asylum seekers from a country which accounted for more than 5 percent of the applications of the previous year but for which the refugee recognition rate was lower than 5 percent, would be refused entry unless they were able to prove that deportation to their country of origin would constitute a threat to their lives.” 57 Additionally, during this time period from 1985–1994, there were numerous restrictions across Western Europe, such as visa requirements for refugees from Bosnia (Sweden and Denmark). 58 Böcker and Havinga also note two types of restrictions aimed at tightening asylum that overlap somewhat with those reported by Shuster: those aimed at preventing entry and those making the country’s asylum procedures less attractive. 59 These measures included “visa requirements, carrier sanctions and pre-flight checks in particular countries of origins,” as well as policies that include “shortening procedures and facilitating expulsion of rejected applicants (e.g. limited appeal possibilities, accelerated procedures for ‘manifestly unfounded’ applications, special procedures at airports, enforced detentions of rejected applicants) and measures restricting freedom of movement and limiting the rights of asylum seekers (e.g. compulsory housing in reception centers, [no right to work, and so forth]).” 60 As Astri Suhrke notes, the global extent of these trends and the implications are ominous for asylum seekers: The most dramatic feature of the refugee scene in the 1990s was the globalized restriction on asylum. In both the North and the South the very institution of asylum seemed in danger. As a UNHCR official complained in 1997, “non-compliance with international treaty obligations for refugees is becoming something of a global norm.” 61 Clearly, there is a concern that the protections that are supposed to be afforded by international refugee norms are under attack, leaving fewer options to the vulnerable. Decline in the availability of asylum in the US would be particularly troubling given that it appears to act somewhat as a buffer for those who cannot enter under normal refugee procedures. 56. Id. at 121–22. 57. Anita Böcker & Tetty Havinga. Asylum Applications in the European Union: Patterns and Trends and the Effects of Policy Measures, 11 J. ReFUgee stUD . 245 (1998). 58. Id. at 258. 59. Böcker & Havinga, supra note 57, at 258. 60. Id. at 258–259. 61. Astri Suhrke, Burden-Sharing during Refugee Emergencies: The Logic of Collective versus National Action. 11 J. ReFUgee stUD . 396, 396 (1998).
2015 How Draconian Are the Changes to US Asylum Law? 167 v. CHANGES IN US ASYLUM PoLICY AND LAw Scholars and practitioners similarly fear that the recent changes to US policy may have diminished the ability of individuals to seek refuge in the United States and weakened US compliance with the legal norms to which it has committed. Many critics of current US asylum policy link the changes to the War on Terror following 11 September; however, the most significant legal component of this change was promulgated in reaction to the 1993 World Trade Center bombing, which was codified in the Illegal Immigration Reform and Immigrant Responsibility Act (IIRIRA) and signed into law in September 1996. 62 The most substantive changes that IIRIRA makes to the asylum process include “a new ‘expedited removal’ procedure and ‘credible fear’ screening for asylum seekers arriving at ports of entry with no documents or with documents suspected to be false,” 63 as well as an increase in the grounds for mandatory detention, a one-year deadline on applying for asylum, and delays in work authorization eligibility. 64 First, the law initiates requirements for expedited removal procedures at ports of entry for people without credible fear of return, without documentation, or with documents that were viewed as fraudulent by the secondary immigration inspector. 65 With few exceptions, these immediate deportations are not subject to administrative or judicial appeal. This initial selection process has been credited for the decrease of interviews with asylum officers, resulting in a subsequent decrease in asylum. 66 The expedited removal process has been implemented in piecemeal fashion in three phases: 1) expedited removal of those entering land and air ports began in April 1997; 2) expedited removal was extended to those entering sea ports in November 2002; and 3) expedited removal was expanded to areas within 100 miles of land borders between ports of entry in August 2004. 67 Because the majority of people enter via land or airports, we focus on April 1997 as the implementation of the expedited removal process. Concern has been raised regarding the implementation of these policies by the secondary inspectors because of the real possibility that some individuals with a legitimate fear of return have been deported under these streamlined procedures. 68 Deportation under this process carries an additional penalty: those people who have been deported 62. See Schoenholtz, supra note 54; RUtH e. w asem , CoNgRessioN aL ReseaRCH seRviCe , U.s. immigRatioN PoLiCy oN a syLUm seekeRs 4–5 (2005) [hereinafter w asem ]. 63. a NkeR, supra note 6, at 11. 64. Cianciarulo, supra note 7, at 110–11. 65. Schoenholtz, supra note 54, at 325–26; w asem , supra note 62, at 5. 66. Id. 67. Id. 68. Allen Keller, Andrew Rasmussen, Kim Reeves, & Barry Rosenfeld, Evaluation of Credible Fear Referral in Expedited Removal at Ports of Entry in the United States, in stUD y oN a syLUm seekeRs iN exPeDiteD RemovaL (2005).
Vol. 37 168 HUMAN RIGHTS QUARTERLY by expedited removal proceedings are banned from reentry into the United States for five years.69 As Schoenholtz reports, no nongovernmental agency has been allowed to study the implementation and impact of the expedited removal process. 70 However, three governmental agencies (the General Accounting Office, the United Nations High Commissioner for Refugees, and the US Commission on International Religious Freedom) have been allowed access and each agency’s study has found mixed results, suggesting that problems continue to exist with the process which may result in some individuals being turned away who are bona fide asylees. 71 Second, the law’s requirement that asylum seekers apply within one year of their arrival except under a very narrow set of circumstances places an additional hurdle in the path of a significant portion of asylum seekers. Specifically, as of April 1998, applicants must, in effect, document their arrival. The law also put into place the mandatory detention of people claiming asylum without appropriate documentation, which may also serve as a significant deterrent to potential asylum seekers. 72 In addition, the law empowers the Attorney General to permanently bar individuals from receiving asylum benefits if they knowingly filed a frivolous claim. 73 IIRIRA also establishes the ineligibility of convicted criminals or dangerous or previously settled people, and imposes strict safe third country obligations. 74 We focus on the two major components in our time series models—the expedited removal process and the one-year deadline, each of which has different timing requirements in regards to implementation. Since the implementation of these major components varied, we conceptualize the impact of IIRIRA as two separate interventions. As Flynn and Patterson note, “[d]epending on one’s perspective, these amendments to INA may be considered overly harsh or long overdue mea- 69. INA § 212(a)(9)(A)(i). 70. Schoenholtz, supra note 54, at 332. 71. g oveRNmeNt a CCoUNtabiLity o FFiCe, sigNiFiC aNt v aRiatioN existeD iN a syLUm o UtComes aCRoss immig Ratio N Co URts a ND JUDges , GAO-08-940 (2008), available at http://www.gao.gov/products/ GAO-08-940; Us C ommissioN oN iNteRN atioN aL ReLigioUs FReeDom , RePoR t oN a syLUm seekeRs iN exPeDiteD RemovaL : RePoR t CaRD t wo yeaRs LateR (2008), available at http://www.uscirf. gov/reports-briefs/special-reports/report-asylum-seekers-in-expedited-removal; Rachel L. Swarns, U.N. Report Cites Harassment at American Airports of Asylum Seekers, N.y. t imes , 13 Aug. 2004, available at http://www.nytimes.com/2004/08/13/national/13immig. html. 72. Dete Ntio N w at CH Netwo Rk, FaCts a bo Ut ma NDato Ry Dete Ntio N (2011), available at http://www. detentionwatchnetwork.org/sites/detentionwatchnetwork.org/files/MD%20factsheet%20 9-25-11-FINAL.pdf. 73. INA § 208(d) (6). 74. INA § 208(d)(2); in addition, it denies employment authorization for asylum seekers; in return, their case is supposed to be decided within 180 days, but delays that stem from the applicant stop the clock Immigration lawyers have reported to us that the clock is stopped while documents are authenticated by the State Department—a process which can drag out for months—and, often, bureaucratic inertia leads to the clock not being restarted after the documents are received.
2015 How Draconian Are the Changes to US Asylum Law? 169 sures to simplify and streamline removal procedures in an effort to deport criminal aliens in greater numbers and gain control over illegal immigration.” 75 Critics have been particularly concerned about the effect of the expedited removal process, especially in light of the decreasing number of asylum seekers identified in the process. 76 We believe from the perspective of IJ decision-making that the implementation of IIRIRA expedited removal process in April 1997 increased the probability that IJs will grant asylum in two ways: 1) winnowing down the pool of applicants, removing many of the bogus applicants or applicants with weaker claims, and ultimately leaving a stronger pool of applicants; and 2) increasing the credibility of asylum applications in IJs’ perceptions because of their expectations of the law’s effect and intent. The changes in the IJs’ perceptions would likely lead to a stronger willingness to give applicants in general the benefit of the doubt. However, it is also possible that they would give the benefit more to those applicants who at least have an underlying foundation of documented human rights abuse in the country of origin to give some substance to their claim. Unfortunately, government-imposed data restrictions prevent us from determining the extent to which individuals with bona fide claims and a strong probability of a grant of asylum were instead removed at a port of entry and denied the chance to make a claim before an AO or IJ. In regard to the one-year filing deadline, which went into effect in April 1998, we also expect a positive effect on grant rates. While the expedited removal process is presumably winnowing out weaker cases, we believe the one-year deadline is in effect injecting into the IJ case pool a set of stronger cases. In other words, AOs are now feeding a new stream of affirmative cases into the IJ pool—cases which AOs might have previously granted asylum but no longer can because of the one-year deadline issue that the IJ must resolve. As we noted above, due to the nature of the asylum process, it is the difficult cases that AOs pass on to IJs. The body of referred cases may include easy cases in terms of the overall merit of the applicant’s claim; however, this new set of cases brings with them the legal hurdle of having missed the one-year deadline, an oversight for which there is only a limited set of exceptions. 77 Thus, we temper our expectations somewhat in regard to the effect of the one-year deadline. The second immigration law linked to the War on Terror, the Real ID Act, was passed in May 2005. The motivation fueling the passage of the law was not limited to concerns about terrorists getting in the country but also “economic opportunists.” The perception at the time was that the 75. Patricia Flynn & Judith Patterson, Five Years Later: Fifth Circuit Case Law Developments Under the Illegal Immigration Reform and Immigrant Responsibility Act, 53 bayLoR L. Rev ., 557, 560 (2001). 76. Schoenholtz, supra note 54, at 332. 77. Id. at 344.
Vol. 37 170 HUMAN RIGHTS QUARTERLY asylum processing delays due to a backlog of several years had “allowed economic migrants, unscrupulous individuals, or even potential terrorists to avoid deportation and then to abscond as their applications went unexamined.” 78 Nonetheless, the bill’s author, House Judiciary Chairman James Sensenbrenner (R-Wis), clearly promoted the bill’s intention as the prevention of terrorists gaining legal status in the US through its asylum system. 79 Cianciarulo argues the law is based on the xenophobic misconception that “all non-citizens are potential terrorists, that an application for asylum is a free and easy pass into the United States, and that U.S. law does not give immigration officials sufficient authority to remove unwanted non-citizens from the country.” 80 However, she observes that the law ignores the reality that “asylum—despite never having been a particularly popular avenue for terrorists to pursue—had become even more difficult to abuse.” 81 The Real ID Act requires that asylum applicants meet a somewhat higher standard of proof in asylum hearings. For instance, if the IJ requests proof corroborating what she believes is otherwise credible testimony, then in most circumstances the applicant must provide such proof or fail in her claim. Furthermore, it is possible that minor inconsistencies in the application can be used by IJs to make adverse credibility determinations. The Real ID Act also “requires asylum seekers to demonstrate that their race, religion, nationality, membership in a social group, or political opinion represents ‘a central reason’ for the persecution they suffered or fear.” 82 While some scholars perceive the Real ID Act as codifying existing case law regarding corroboration requirements, credibility determinations, reasons for persecution, and nexus requirements, 83 others fear that this congressional attempt makes significant departures from case law that may leave room for adjudicators to abuse their discretion. For example, Cianciarulo fears that since the statute does not define the concept of centrality, it may create an “opportunity for adjudicators to require more proof of causation” 84 than current case law actually requires and thus has the potential to make asylum 78. Cianciarulo, supra note 7, at 110. 79. Cianciarulo, supra note 7, at 102; Sensenbrenner stated that: There is no one who is lying through their teeth that should be able to get relief from the courts, and I would just point out that this bill would give immigration judges the tool to get at the Blind Sheikh who wanted to blow up landmarks in New York, the man who plotted and executed the bombing of the World Trade Center in New York, the man who shot up the entrance to the CIA headquarters in Northern Virginia, and the man who shot up the El Al counter at Los Angeles International Airport. Every one of these non-9/11 terrorists who tried to kill or did kill honest, law-abiding Americans was an asylum applicant. We ought to give our judges the opportunity to tell these people no and to pass the bill. 80. Id. at 104. 81. Id. at 114. 82. Kerwin, The Use and Misuse of “National Security” Rationale, supra note 52, at 757. 83. a NkeR, supra note 6. 84. Cianciarulo, supra note 7, at 118.
2015 How Draconian Are the Changes to US Asylum Law? 171 difficult for those who need it most, resulting in “devastating consequences for bona fide asylum applicants while providing no additional protection against fraudulent claims.” 85 The authors’ discussions with immigration attorneys practicing in Dallas suggest that the Real ID Act allows judges to deny claims based on minor inconsistencies that do not relate to the heart of the asylum claim and may be easily explainable. However, if the overall effect of the Act is merely to codify existing case law, then we would expect to find no effect on the asylum grants. If critics are correct, we would expect to see declines in the probability of a grant after the law comes into effect even among those who are most likely to have a valid claim. From our discussion of the expectations of scholars, we derive the following hypotheses for the law and policy shocks: • IIRIRA (the April 1997 implementation of the expedited removal process and the April 1998 implementation of the one-year deadline) will have a positive effect on the overall grant rate, with the one-year deadline producing a smaller impact than the expedited removal. • Real ID Act, which passed and went into effect on May 11, 2005 (modeled as May 2005) will have a negative effect or no effect on the overall grant rate. In returning to the question of whether these laws diminish US compliance with its Convention obligations, our expectations are that the statutes have had an overall beneficial effect on asylum seekers who make it into the system. What is difficult to assess due to data restrictions is the extent to which these laws have had a deterrent effect on would-be asylum seekers who may have had a bona fide claim to asylum. We return to this issue again in the subsequent sections. While the underlying interest of this article is clearly policy-driven and our methodology is more data-driven than theoretical, we believe the analysis provides an important historical framework for future empirical asylum research within the extant literature. Much of the empirical literature has focused on the tension between foreign policy interests versus compliance with the international norm of non-refoulement. This debate is viewed primarily through the lens of the larger norms-versus-interest debate in international relations between realists/rational functionalists—who focus on national self-interest 86—and constructivists—who expect the power of international norms to build through networks of domestic and transnational actors—finally pressuring states to formally accept and to comply with these norms. 87 85. Id. at 105. 86. See w aLtz , supra note 13, keoHaNe , supra note 13. 87. See Keck & Sikkink, supra note 15; t He PoweR oF HUmaN RigHts , supra note 15.
Vol. 37 172 HUMAN RIGHTS QUARTERLY The early asylum literature largely demonstrated the influence of geopolitical and material interests over human rights in US asylum policy, 88 while more recent work demonstrates that both normative concerns and national interests influence asylum grants.89 However, the effects of national interests have varied over time. For example, Marc Rosenblum and Idean Salehyan found a strong preference for asylum seekers from communist countries, but this effect vanishes after the Cold War; then, in the new post-Cold War era, human rights conditions, US domestic interests such as trade policy, and stemming the flow of illegal immigration have influenced asylum outcomes. 90 The 11 September attacks shifted the national interest focus to factors linked to terrorism. While Andy Rottman, Christopher Fariss, and Steven Poe found that these attacks had no observable effect on IJ grant rates, 91 Jennifer Holmes and Linda Keith demonstrated that the criteria that influenced asylum grants shifted after the terrorist attacks. Prior to 11 September speaking Arabic had a beneficial effect for an applicant, as did being from a country which had the presence of al Qaeda and which was a state sponsor of terror, but after the attacks the effects shifted to the other direction, substantially decreasing the success of some applicants. 92 Since 11 September the US has more actively controlled ports of entry, which, as Schoenholtz notes, is where asylum seekers often make claims. Additionally, the US has implemented pre-screening procedures at airports abroad. 93 We expand our exploration beyond just the September 11 attacks and include the first World Trade Center attack as well and thus test the effects of the following shocks: • The first World Trade Center attack in 1993 (26 February 1993 modeled as March 1993) will have a negative effect on the overall grant rate. • The September 11 attack (modeled as September 2001) will have a negative effect on the overall grant rate. The asylum literature has increasingly focused on domestic conditions as well. Lisa Hassan and Mark Gibney find that general social and economic 88. See, e.g., Gibney & Stohl, supra note 10; Gibney et al., supra note 10; Loes CHeR & sCa NLa N, supra note 10. 89. See, e.g., Rosenblum & Salehyan, supra note 22; Rottman, Fariss & Poe, supra note 22; Jennifer S. Holmes & Linda Camp Keith, Determinants of Asylum Grants: A Rare Examination of Factors Typically Unobservable in U.S. Asylum Decisions, 22 J. ReFUgee stUD . 224 (2009); Jennifer S. Holmes & Linda Camp Keith, Does the Fear of Terrorists Trump the Fear of Persecution in Asylum Outcomes in the Post September 11 Era? 43 PS: POL. SC. & POL. 431 (2010). 90. Rosenblum & Salehyan, supra note 22. 91. Rottman, Fariss & Poe, supra note 22, at 20. 92. Holmes & Keith, supra note 89, at 434 (Table 3), 435. 93. Schoenholtz, supra note 54, at 329–30.
2015 How Draconian Are the Changes to US Asylum Law? 173 conditions of the applicants’ countries of origins influence grants more than human rights conditions.94 In other words, applicants who seem to be economic migrants have less success in gaining asylum. High unemployment can affect asylum rates in two ways. First, asylum seekers may avoid seeking asylum in countries that have high unemployment. 95 Second, there is evidence of the concern about the cost of asylum recipients. For example, Boswell cites the effort of the British and German governments who distribute asylum recipients across regions in response to concern of the burden of asylees on the receiving communities. 96 Building upon this literature, we control for monthly unemployment, expecting a negative relationship. In addition we add four more control variables. First, we include a smoothed series of the monthly refugee flow to the United States to the relationship we observe in Figure 1. We include three measures that capture political context: 1) a dummy variable for when a Democratic administration controls the White House (and therefore the Attorney General is a Democrat); 2) a dummy variable for the Department of Justice under Attorney General Alberto Gonzales to control the politicalization of the hiring process during his tenure; and 3) a measure of the mean liberalism of the IJs on the bench for each month (see Appendix). Finally, we include a dummy variable that controls for the proportion of cases that are referred from asylum officers. Our analysis will allow us to build upon the extant empirical literature, modeling the impact of the two terrorist attacks in a monthly time series of IJ grant rates that simultaneously examines the major legal changes that follow the attacks and controls for a variety of factors that are theoretically important. vI. ANALYSIS We include two dependent variables in our analysis. Our first dependent variable is a monthly measure of the aggregate grant rate for immigration judges, measured from January 1990 through December 2010. Table 1 below provides descriptive statistics on this variable, as well as other variables included in our models. Our second dependent variable is also a monthly measure of the grant rate for IJs, but for a narrower pool of applicants, those we term “at-risk.” These are applicants who come from states with patterns of gross human rights abuse, which we operationalize as a score of 4 or 5 on Gibney’s Political Terror Scale, a scale that measures personal integrity abuse by the state and is based on human rights country reports published 94. Hassan, supra note 11, at 190; g ibNey , t He etHiCs aND PoLitiCs oF a syLUm , supra note 11, at 132. 95. Neumayer, supra note 23, at 64. 96. Christina Boswell, Burden Sharing in the European Union: Lessons from the German and UK Experience, 16 J. ReFUgee stUD . 316, 317–322 (2003).
Vol. 37 174 HUMAN RIGHTS QUARTERLY by the US Department of State. 97 We separate these cases out in order to control for bad human rights conditions in the country of origin. In other words, these are people who are more likely to be persecuted if returned to their country of origin. 97. Political Terror Scale (PTS), PoLitiC aL t eRRoR sCaLe .oRg , available at http://www.politicalterrorscale.org.
General Pool At-Risk Pool Mean Std. Min. Max Mean Std. Min. Max Dev. Dev. Grant Rate 0.372 0.128 0.142 0.626 0.449 0.155 0.170 0.730 Grant Rate t-1 0.371 0.128 0.142 0.626 0.447 0.156 0.152 0.730 Grant Rate t-2 0.369 0.128 0.142 0.626 0.444 0.156 0.152 0.730 Grant Rate t-11 — — — — 0.426 0.153 0.152 0.729 Number of Cases 2174 697 529 3573 1112 469 259 2320 Number of Cases t-1 2170 702 529 3573 1109 471 259 2320 Number of Cases t-2 2166 708 529 3573 1107 473 259 2320 IIRIRA Expedited Removal 0.746 0.435 0 1 0.746 0.435 0 1 IIRIRA One Year 0.692 0.463 0 1 0.692 0.463 0 1 Real ID 0.308 0.463 0 1 0.308 0.463 0 1 Nine Eleven 0.005 0.067 0 1 0.005 0.067 0 1 First WTC Bomb 0.005 0.067 0 1 0.005 0.067 0 1 Gonzalez AG 0.145 0.353 0 1 0.145 0.353 0 1 Democratic Admin 0.543 0.499 0 1 0.543 0.499 0 1 Average IJ Ideology 1.255 0.059 1.094 1.392 1.255 0.059 1.094 1.392 Unemployment Rate 0.057 0.015 0.038 0.100 0.057 0.015 0.038 0.100 Unemployment Ratet-1 0.057 0.015 0.038 0.100 0.057 0.015 0.038 0.100 Refugee Flow 5936 2277 1931 11276 5936 2277 1931 11276 Refugee Flow t-1 5955 2295 1931 11276 5955 2295 1931 11276 Refugee Flow t-2 5977 2320 1931 11276 5977 2320 1931 11276 Asylum Officer Referrals 2306 1360 0 5926 2306 1360 0 5926 Asylum Officer Referrals t-1 2300 1368 0 5926 2300 1368 0 5926 Asylum Officer Referrals t-2 2293 1375 0 5926 2293 1375 0 5926 Asylum Officer Referrals t-10 2235 1436 0 5926 2235 1436 0 5926
includes vertical lines for the interventions in which we are primarily interested in including: the first World Trade Center bombing (March 1993), the implementation of IIRIRA (April 1997 and April 1998), 11 September 2001, and the implementation of the Real ID Act (May 2005). First, it appears that the grant rate has increased generally over the time period of our study. More
2015 How Draconian Are the Changes to US Asylum Law? 175 importantly, we can see in the graph that applicants from at-risk countries largely trend with the general set of applications. However, following the implementation of IIRIRA and Real ID, the at-risk applications have a much higher grant-rate except for the months immediately before and after Real ID is passed. In September 2001 the overall rise in grant rates from IRRIRA tapers off for both groups prior to the Real ID going into force with the grant rates of the two groups converging in May 2005. To understand how the interventions of interest may have affected not just relief rates but also the numbers of applicants reaching this stage of the process we present Figure 3, which graphs the number of claims for relief for each of the two groups. We can see in May 2005 that the claims that are made by at-risk applicants peaks, and thus the precipitous increase in the rates at which these at-risk applicants are granted relief is likely tied to the reduction in the number of those cases which reach IJs. If we further examine this figure we can see a clear increase in claims overall as the Cold War ends, which levels-off around 1996 and seems unaffected by IIRIRA’s two provisions. We see a similar increase in claims among the at-risk group in terms of the Cold War ending; however, there is a subsequent decline in the number of claims and then a gradual overall increase as IIRIRA is imple-
Vol. 37 176 HUMAN RIGHTS QUARTERLY mented, peaking in May 2005, when the Real ID Act goes into force. From that point forward, the number of claims from at-risk applicants decreases significantly until it levels off somewhat in 2008. We return to these trends in the analysis below. The total number of claims for relief also declines precipitously following Real ID. It is clear from these trends that IIRIRA has had an impact on the composition of the IJs’ caseload, increasing the proportion of claims from at-risk applicants, while Real ID does not seem to have an impact on the composition of the claims, only the overall number of claims. It is difficult to explain from a legal perspective the significant drop in the number of claims following Real ID; instead, the decrease suggests potential asylum seekers are selecting not to identify themselves and make a claim, which would predict an effect that is counter to our hypotheses. In the time series models we control for the monthly number of claims. In addition, we control for the number of cases referred to the IJs by asylum officers (AOs), who serve as the first official encountered by those asylees who apply affirmatively. A sharp drop in the number of cases referred by the AOs to the IJs might indicate that the policy interventions and shocks had their effects by reducing the number of applicants making it into the system. Thus, the addition of the AO referral series helps to account for this possibility.
2015 How Draconian Are the Changes to US Asylum Law? 177 We approach our modeling task as primarily an intervention model, 98 where what we are most interested in is how policies (IIRIRA and Real ID) and events (the 1993 World Trade Center bombing and the 9/11 attacks) affect the likelihood that an applicant will be granted asylum by an IJ. We present two models in Table 2. Table 3 presents the immediate and longrange effects for the variables that achieve statistical significance. Models 1 and 2 are our basic models of the data generating process and include a minimum number of controls. In Model 1 the dependent variable is the grant rate for the overall pool of asylum applicants in a given month and for Model 2 the dependent variable is the grant rate for the at-risk pool of applicants. Included in both models are lags of the dependent variable on the right-hand side of the equation, thus each model can be described as an auto-distributed lag (ADL) model. The ADL approach is general and does not impose any a priori assumptions on the structure of the data while still allowing for dynamic interpretation of variables and accounting for autoregression. 99 We feel this modeling approach is necessary given the general lack of theory that would inform precise dynamic specifications for the interventions of interest (more on this below) and the highly dynamic nature of the time series. The ADL model is appropriate where the dependent time series is stationary, meaning that past shocks to the series are not permanent. Evidence for the stationarity of all of the models is present in the cumulative coefficients for the lags of the dependent variables. Stationarity is indicated when these lags are cumulatively less than one, as they are, for instance, in both model 1 (0.75) and model 2 (0.76). We determined the number of autoregressive lags to include for each series by including the minimum number of lags which rendered the errors from the series white noise. For the series of overall grant rates it was necessary to include two lags of the dependent variable to create white noise residuals. For the at-risk time series an additional lag was necessary for the dependent variable. To summarize, by including the required lags we are confident that we have created time series that are unlikely to generate spurious results. In general, all models fit the data well, with highly statistically significant F–statistics, and large adjusted R 2 statistics (although the adjusted R2 is artificially inflated given the inclusion of lags of the dependent variables). More importantly, as suggested by Luke Keele and Nathan Kelly, 100 we report the Breusch-Godfrey LaGrange Multiplier test for serial autocorrelation to 98. w aLteR eNDeRs, a PPLieD eCoNometRiC t ime seRies 269–70 (1995) (explaining the time-series intervention model). 99. Suzanna De Boef & Luke Keele, Taking Time Seriously, 52 a m. J. PoL . sCi. 184 (2008). 100. Luke Keele & Nathan J. Kelly, Dynamic Models for Dynamic Theories: The Ins and Outs of Lagged Dependent Variables, 14 PoL . a NaL . 186 (2006) (explaining the utility of the Breusch-Godfrey LM test).
Vol. 37 178 HUMAN RIGHTS QUARTERLY
Model 1: General Model 2: At-Risk Coeff. S.E. Coeff. S.E. Grant Rate t-1 0.531 0.068 0.524 0.070 Grant Rate t-2 0.224 0.069 0.227 0.071 Grant Rate t-11 — — -0.023 0.051 Number of Cases -5.63E-06 6.36E-06 -1.34E-05 1.16E-05 Number of Cases t-1 1.86E-06 6.63E-06 5.47E-06 1.25E-05 Number of Cases t-2 7.23E-06 6.15E-06 -7.69E-05 1.14E-05 IIRIRA Expedited Removal 0.028 0.010 0.039 0.014 IIRIRA One Year 0.019 0.011 0.027 0.015 Real ID 0.029 0.009 0.051 0.012 Nine Eleven -0.043 0.026 0.000 0.029 First WTC Bomb -0.008 0.025 -0.008 0.029 Gonzalez AG -0.005 0.009 -0.012 0.10 Democratic Admin -0.012 0.006 0.004 0.006 Average IJ Ideology 0.029 0.045 -0.007 0.052 Unemployment Rate -0.278 1.129 -0.735 1.315 Unemployment Ratet-1 0.603 1.107 0.836 1.285 Refugee Flow 8.27E-06 5.83E-06 8.94E-06 6.67E-06 Refugee Flow t-1 -7.25E-06 8.57E-06 -9.19E-06 9.86E-06 Refugee Flow t-2 7.06E-06 5.79E-06 -1.03E-06 6.49E-06 Asylum Officer Referrals -2.35E-06 4.42E-06 4.33E-06 4.98E-06 Asylum Officer Referrals t-1 3.5E-06 5.86E-06 -2.38E-07 6.6E-06 Asylum Officer Referrals t-2 6.64E-07 4.54E-06 -3.9E-06 5.08E-06 Asylum Officer Referrals t-10 -1.38E-06 2.19E-06 2.22E-06 2.29E-06
Constant -0.017 0.058 0.083 0.067 N 221 221 Adj. R2 0.97 0.97 LM Test χ2 0.397 0.373 p-value 0.529 0.541 Bolded coefficients significant at p<0.05 (one-tailed).
Immediate Impact Long Run Impact At-Risk General At-Risk General IIRIRA Expedited Removal 3.9 2.8 15.7 11.4 IIRIRA One Year 2.7 1.9 10.8 7.8 Real ID 5.1 2.9 20.2 11.8 Nine Eleven — -4.3 — — Democratic Administration — -1.2 — -4.8
2015 How Draconian Are the Changes to US Asylum Law? 179 ensure that we have accurately modeled the data generating process for each time series. Non-significant values indicate that auto-correlation is not present, as is the case in all of our models. Additional tests indicate that heteroscedasticity is not a concern. In short, we are confident that we have accurately modeled the data generating process and proceed with a substantive interpretation of the coefficients. The structure of the various interventions of interest is not clear from the asylum literature. For instance, with respect to the implementation of IIRIRA, the law is generally considered to have gone into effect in April of 1997, while others note it was not really fully in effect until April of 1998 due to the one-year deadline implementation. 101 In addition, it is clear that the scope of IIRIRA was expanded once its basic provisions were implemented; there was an expansion of the application of the law to sea ports in November 2002 and an additional expansion again to areas within 100 miles of land borders in August 2004. 102 In the intervention model we code the IIRIRA intervention for expedited removal as 1 after April 1997 and 0 before; we coded the intervention for the IIRIRA in a similar way, with the one-year deadline as 1 after April 1998 and 0 before. Real ID is also specified as a step-function, taking a value of 1 after May 2005 and 0 before. We specify the policy interventions as step-functions, and not pulse functions, because theoretically these laws, once in effect, are permanent changes to the asylum process. Alternatively, we test the two terrorist attacks of interest as pulse functions, with the first World Trade Center bombing coded a 1 in March 1993 and 0 otherwise. 103 We coded the 9/11 bombing as equal to one in September 2001 and 0 otherwise. We tested a multitude of specifications for IIRIRA, Real ID, 9/11 and the first World Trade Center bombing and included those that the data dictated as best. We determined this by estimating multiple models with the different intervention structures and choosing the model which minimized the Akaike Information Criterion (AIC), indicating the best fit to the data. 104 This approach is consistent with our desire to give preference to the data generating process, and not theory, in specifying our models. Using this approach leads us to treat the two terrorist attacks as pulse functions (temporary interventions) rather than permanent interventions. Of course, to the extent that the attacks were significant in motivating either 101. Philip G. Schrag et al., Rejecting Refugees: Homeland Security’s Administration of the One-Year Bar to Asylum, 52 w m. & maRy L. Rev . 651 (2010). 102. Schoenholtz, supra note 54, at 326. 103. We tested alternative specifications of the WTC bombing, including coding the even as occurring in February, when the actual bombing occurred. Tests indicate that the March specification is better fit to the data and this makes some sense given that the bombing occurs on February 26, late in the month for the effects to be present. 104. Shuhua Hu, Akaike Information Criterion, www4.ncsu.edu (2007), available at http:// www4.ncsu.edu/~shu3/Presentation/AIC.pdf.
Vol. 37 180 HUMAN RIGHTS QUARTERLY IIRIRA or Real ID, their effects are somewhat underestimated. Nevertheless, we believe that the pulse specification accurately captures the immediate effect either attack may have had. Turning to Table 2, we find that both IIRIRA interventions perform similarly in that both interventions increase the grant rate, with the expedited removal intervention having a stronger effect than the one-year time limit intervention in that the beneficial effect of each intervention is almost double for the at-risk group than what it is for the general pool of applicants. The April 1997 IIRIRA intervention, which represents the start of the expedited removal process, produces an immediate increase in the grant rate of 3.9 percentage points in the at-risk model and 2.8 in the general pool, with a long range effect of an increase of 15.7 percentage points in the at-risk group compared to 11.4 in the general pool. 105 In short, there is clear evidence that that the implementation of this component of IIRIRA raised the probability of an applicant receiving asylum and that the primary beneficiaries of that increase were those likely to be most vulnerable upon return to their home countries. To put this substantive effect into perspective, we estimate that between January 2000 and December 2010 an additional 25,569 at-risk applicants were granted asylum by IJs because of the implementation of the IIRIRA expedited removal provision—an estimate which uses the long-range effect of IIRIRA expedited removal reported in Table 3. The second IIRIRA intervention, which delineates the April 1998 implementation of the one-year filing deadline, affects both pools but the effect is considerably smaller than that of the first April 1997 intervention: the immediate substantive impact is an increase in the grant rate of 2.7 percentage points for the at-risk pool and 1.9 percentage points in the general pool of applicants with a long range effect of an increase of 10.8 and 7.8 percentage points, respectively. The effect of Real ID follows a similar pattern, but with a greater difference between the two pools of applicants, with an immediate impact on the at-risk group that is an increase of 5.1 percentage points and a long term impact of an increase of 20.2 percentage points, compared to the general pool in which the immediate impact is an increase of 2.9 percentage points and the long term impact an increase of 11.8 percentage points. Indeed, this is what we would expect from a simple visual examination of Figure 2, which shows an increase for both pools but a rather substantial separation in grant rates for the general pool and the at-risk pool after the implementation of Real ID. For the at-risk applicant pool we estimate that from January 2006 through December 2010 the implementation of Real ID led to the granting of an additional 11,956 applications. As we noted above, the trend graphs had 105. Long-run effects are calculated as follows, using the values from models 1 for the IIRIRA expedited removal intervention: .028/(1-.531-.224). See De Boef & Keele, supra note 99, for further explanation.
2015 How Draconian Are the Changes to US Asylum Law? 181 suggested that with Real ID potential asylum applicants were self-selecting out of the pool faced with the potentially tougher standards imposed by the law. And it is likely that the weakest potential applicants are opting out due to Real ID and thereby continuing to strengthen the application pool beyond that of IIRIRA. Overall, the analysis here makes it quite difficult to argue that IIRIRA and Real ID have had a draconian effect on those most vulnerable to persecution. Turning to the two terrorist attacks, we see that the effect of the bombing of the World Trade Center in March of 1993 is negative, as we would expect, but it is not statistically distinguishable from zero in either of the models. The terrorist attacks in September 2001 had an immediate negative effect on the likelihood that an applicant would be granted asylum by an IJ but only in the general pool models. Our estimate of the immediate effect of 9/11 on the grant rate for the general pool of applicants is a decrease of 4.3 percentage points. This is in line with our expectations given the broader literature on immigration and the more specific work on asylum. We tested additional dynamic specifications for the effect of 9/11, including lags of the policy shock which would allow the attack to have a permanent effect on the grant rate; we rejected these lags as unnecessary to explain the time series based on tests using the AIC. 106 This is important information that suggests that the effect of 9/11, while immediate and substantively substantial, was transitory, having little effect on grant rates beyond the month of the attack. Indeed, we find that though there is a clear effect for the general pool of applicants there is no statistically distinguishable effect in the at-risk series. None of the control measures in the model are statistically significant. Only Democratic administration in the general model achieves what would be an appropriate level of statistical significance but it is not in the predicted direction—the coefficient suggests a negative effect on grant rates. In addition, the models included the first difference and two lags of the number of cases decided by IJs in a given month (with lag structures specified to create white noise errors for the series) but none of the coefficients achieve an acceptable level of statistical significance. While caseloads may be a significant component of the asylum system, the fluctuations in aggregate caseload do not appear to have an effect on aggregate grant rates. 106. This holds given our specification of the 9/11 intervention as a pulse function instead of a step function, a decision we also tested against the notion of minimizing the AIC. Because we treat 9/11 as a pulse-function, absent lags of this variable, it does not have a long-range effect on the time series. This same method is used to determine the structure of the intervention for the World Trade Center bombing in March 1993.
Vol. 37 182 HUMAN RIGHTS QUARTERLY vII. CoNCLUSIoNS AND IMPLICATIoNS In this article we set out to address changes in US asylum from the perspective of ending human rights violations. We sought to discover whether recent terrorist attacks and core changes in US policy have had the draconian effect feared by some critics of greatly diminishing the strength of the international norm of non-refoulement and the protection it is intended to afford those fleeing persecution. While government-imposed data restrictions hamper our ability to make a full assessment, we can provide an initial discussion about the impact of the terrorist attacks and subsequent legal changes. The terrorist attacks have little observable impact on the overall rate. Our analyses demonstrate that the implementation of IIRIRA has increased the probability of a grant of asylum for those who make it past the ports of entry into the United States, and this is especially so for those who are fleeing states with high levels of human rights abuses. Our analyses also demonstrate that Real ID has had an even stronger effect on both the at-risk pool and the general pool of applicants. The net effect seen here is that both statutes have strongly increased the odds that applicants who are most at-risk will be granted relief. This result is especially significant given the US government’s trend in reducing refugee flows. While we believe these results represent a positive assessment of US compliance with its international and domestic obligations, we must emphasize that we are only able to examine their effect on a subset of the population that might potentially possess a legitimate claim to asylum in the US. We are unable to directly test the effect of these laws on would-be asylum seekers who are removed at ports of entry without a credible fear hearing and the extent to which would-be asylees are deterred from seeking relief because of the filing deadline or the possibility of being placed in removal proceedings and detention. Nevertheless, we believe that our controls for caseload and AO referrals help to control for this possibility in our models. Finally, we emphasize that the impact of 9/11 and of the policy interventions, IIRIRA and Real ID, are different depending on the population being examined. It should hearten observers of the US asylum system that those applicants coming from countries with the most dangerous human rights situations are more likely to receive asylum than those fleeing less dangerous places. The different treatment of these two groups of applicants also suggests that IJs may be relying on something other than their own policy predispositions in adjudicating asylum cases (or at least not solely on those predispositions)—it suggests at least some minimal role for the law and human rights concerns in the determination of asylum applications. 107 Our analysis suggests that Congressional policies with an attempt to limit access to asylum have actually benefited those most at-risk of persecution. Surprisingly, for those who reach the adjudicatory stage of the asylum process, more and not less protection is potentially available from an IJ. 107. Keith, Holmes & Miller, supra note 22.
2015 How Draconian Are the Changes to US Asylum Law? 183
Refugee Admissions The peak refugee ceiling is in 1992, matching the peak of global refugees. The ceiling falls to a low of 70,000 from 2002–2007. However, the rough proportion of the US ceiling to global refugees is fairly stable in this time period (about .007 percent). There is more variation in the number of refugee arrivals. Actual refugee arrivals fall sharply after 2001, reflecting increased security related scrutiny of refugees before admission.
Year US Refugee Arrivals US Refugee Admission Ceilings Global Refugees 1990 122,066 125,000 17,395,979 1991 113,389 131,000 16,854,795 1992 115,548 142,000 17,838,074 1993 114,181 132,000 16,325,525 1994 111,680 121,000 15,753,691 1995 98,973 112,000 14,896,087 1996 75,421 90,000 13,357,087 1997 69,653 78,000 12,015,350 1998 76,712 83,000 11,480,860 1999 85,285 91,000 11,689,358 2000 72,143 90,000 12,064,599 2001 68,925 80,000 12,031,996 2002 26,788 70,000 10,594,055 2003 28,286 70,000 9,592,795 2004 52,840 70,000 9,560,280 2005 53,738 70,000 8,661,988 2006 41,094 70,000 9,877,703 2007 48,218 70,000 11,388,967 2008 60,107 80,000 10,489,812 2009 74,602 80,000 10,396,540 2010 73,293 80,000 10,549,686 Sources: o FFiCe oF immigRatioN statistiCs , 2011 yeaRbook oF immigRatioN statistiCs , 39 (2012) (Table 13: Refugee Arrivals for Fiscal Years 1980 to 2011), available at https://www.dhs.gov/sites/ default/files/publications/immigration-statistics/yearbook/2011/ois_yb_2011.pdf (Note: Excludes Amerasian immigrants except in fiscal years 1989 to 1991.); Refugee ceilings: o FFiCe oF ReFUgee Resett LemeNt , a NNUa L RePo Rt to Co Ng Ress (for each year between1990 and 2000), available at http:// www.acf.hhs.gov/programs/orr/resource/annual-orr-reports-to-congress; o FFiCe o F immig Ratio N statistiCs , a NNUaL FLow RePoR t : ReFUgees aND a syLees (2006, 2008 & 2011); Global refugees: “Refugee population by Origin, UNHCR Total Refugee population by country of asylum, 1960-2010 & Total Refugee population by origin, 1960–2010”, available at http://www.unhcr.org/statistics/ Ref_1960_2010.zip.
Vol. 37 184 HUMAN RIGHTS QUARTERLY Measuring the Policy Proclivities of Immigration Judges One typical approach to measuring judicial ideology is to use the ideology or party of the appointing president as a proxy for the ideology of the judge, in various guises, 108 in various guises, including using these proxies specifically in asylum cases. 109 To create our measure of policy predispositions, we create a factor score that summarizes the contribution of a number of background characteristics to the policy predispositions of a judge toward asylum cases. This is a more policy-specific method of measuring a judge’s policy predisposition—our measure is specific to asylum decisions and we do not intend it as a general measure of judicial ideology. 110 A subject specific approach such as ours is often necessary when scholars study specialized judicial decision makers, because more general approaches may not uncover the relevant ideological dimensions of decision making. 111 Our approach builds upon earlier studies that have looked to judges’ background characteristics, believing that they represent a socialization process which ultimately results in votes. 112 The prior experiences of judges serve as indicators of judicial perspectives on cases. 113 In addition, there is strong evidence that the background characteristics that we consider have been used by key actors in the asylum decision making bureaucracy specifically as cues for the likely policy preferences of asylum adjudicators. For instance, when Janet Reno sought to increase the size of the BIA in the 1990s, she was careful to select members who had backgrounds in academia, private practice, and advocacy on behalf of immigrants to balance what was thought to be an overweening dominance by those with a background in government enforcement of immigration laws. 114 Further, when Ashcroft sought to streamline the BIA he removed those judges that 108. See, e.g., Tracey E. George & Lee Epstein, On the Nature of Supreme Court Decision Making, 86 a m. PoL . sCi. Rev . 323 (1992); Mark A. Zupan, Measuring the Ideological Preferences of U.S. Presidents: A Proposed (Extremely Simple) Method, 73 PUb. CHoiCe 351 (1992). 109. Melissa Ocepek & Joel Fetzer, The Causes of Pro-Immigration Voting in the United States Supreme Court, 44 iNt ’L migRatioN Rev . 659 (2010). 110. For a further discussion of this distinction, see Joshua B. Fischman & David S. Law, What Is Judicial Ideology, and How Should We Measure It?, 29 w asH . U.J.L. & P oL ’y 133 (2009). 111. Lawrence Baum, What Judges Want: Judges’ Goals and Judicial Behavior, 47 PoL . Res. Q. 749 (1994); Nancy Staudt, Lee Epstein & Peter Wiedenbeck, The Ideological Component of Judging in the Taxation Context, 84 w asH . U.L. R ev . 1797 (2006); Banks Miller & Brett C. Curry, Expertise, Experience and Ideology on Specialized Courts: The Case of the Court of Appeals for the Federal Circuit, 43 L. & soC ’y Rev . 839 (2009). 112. Gerard S. Gryski & Eleanor C. Main, Social Backgrounds as Predictors of Votes on State Courts of Last Resort: The Case of Sex Discrimination, 39 w este RN Po L. Q. 528 (1986). 113. C. Neal Tate & Roger Handberg, Time Binding and Theory Building in Personal Attribute Models of Supreme Court Voting Behavior, 1916–88, at 35 a m. J. PoL . sCi. 460 (1991). 114. Schoenholtz, supra note 54, at 353.
2015 How Draconian Are the Changes to US Asylum Law? 185 had previous experience working in NGOs and on behalf of immigrants. To this end, we examine the IJ’s career path to create a tightly focused proxy for a policy predisposition toward immigration rights and asylum. We believe that our measure is a strong proxy for asylum liberalism in the IJs’ career paths that likely reflects a socialization process that we discuss further below. Additionally we believe that it accounts for the early career selections of some IJs that may indicate an underlying policy proclivity that is subsequently strengthened through additional career socialization. For example, a conservative individual may be more likely to seek out a job as an INS agent or a prosecutor. These career experiences then are likely to reinforce those underlying proclivities. First, using un-rotated factor analysis and several background variables related to the immigration judges, we create a factor, which we believe captures an underlying immigration liberalism that will drive decision making. In essence this is a measure of policy predisposition, although we believe that it is most accurately described as capturing the degree to which an IJ has been socialized to give the benefit of the doubt to asylum seekers. We included the following variables in the factor analysis: experience at INS, experience at the Department of Homeland Security (other than INS), experience at the EIOR, experience as a prosecutor, experience at an NGO, experience at an immigration related NGO, military experience, academic experience, private practice experience, prior judicial experience (other than as an IJ) and corporate experience. These variables are included because previous scholarly work and personal observation of these judges lead us to believe that these characteristics are strong proxies for the liberalness of a judge. For most of the characteristics we have a priori expectations about how the characteristic will relate to asylum or immigration liberalism. For instance, we believe that experience in the Department of Homeland Security, at INS or as a prosecutor should indicate skepticism toward asylum claimants. On the other hand, it is straightforward to expect that those who have worked for NGOs or for immigration NGOs more specifically are highly likely to view asylum claimants sympathetically, given that many of these NGOs represent or support the representation of immigrants. As is common practice, we retain only the factors which have eigenvalues above 1. 115 We then used these elements to create a factor score using regression techniques. The results of the factor analysis and regression scoring method are displayed in Table A2 below. The retained factor has an eigenvalue of 1.42 and explains 78 percent of the variance in the data. No other factor in the data approaches an eigenvalue of 1. The factor loadings for the various background character - 115. RiCHaRD g oRsUCH , FaCtoR a NaLysis (2d ed. 1983).
Vol. 37 186 HUMAN RIGHTS QUARTERLY
Factor Loadings Factor 1 INS Experience -0.414 DHS Experience -0.226 EOIR Experience -0.119 Former Prosecutor -0.098 NGO Experience 0.685 Immigration NGO Experience 0.679 Military Experience -0.178 Academic Experience 0.331 Private Practice Experience 0.298 Prior Judicial Experience 0.019 Corporate Experience 0.092 Regression Scoring Coefficients Factor 1 INS Experience -0.184 DHS Experience -0.097 EOIR Experience -0.034 Former Prosecutor -0.025 NGO Experience 0.359 Immigration NGO Experience 0.363 Military Experience -0.058 Academic Experience 0.120 Private Practice Experience 0.129 Prior Judicial Experience 0.007 Corporate Experience 0.034 istics are exactly as we would predict, with the previous INS, DHS, EOIR, prosecutorial and military experience suggestive of a conservative view toward asylum seekers. Alternatively, NGO experience, immigration-specific NGO experience, and academic experience are all predictably suggestive of increased liberalism toward asylum seekers. Prior judicial experience is slightly indicative of greater asylum liberalism as are prior private practice experience and corporate experience. We did not have strong a priori predictions for these three variables, but their inclusion seems to explain some variance in asylum liberalism. As a check on the robustness of this approach to measuring the proclivities of IJ toward asylum cases, we also create a measure that we call presidential liberalism, which is a factor score combining a dummy variable for the appointing president and the DW-NOMINATE score from Poole and Rosenthal. 116 Our method for creating this factor score is identical to the method described above and we do not report the factor loading here, although we again retained one factor with an Eigenvalue above 1. Our two 116. keitH PooLe & HowaRD RoseNtHaL , CoNgRess : a P oLitiC aL -eCoNomiC HistoR y oF RoLL CaLL v otiNg (1997).
2015 How Draconian Are the Changes to US Asylum Law? 187 measures of judicial policy proclivity are positively correlated, with a modest correlation of r = 0.15. Our measure of asylum liberalism ranges from 0 to 4.08, with higher scores indicating a greater proclivity to grant relief to aliens in asylum hearings. The median score is 0.954, meaning that most of the IJs in our data lean toward conservatism in the granting of asylum. Put differently, the majority of IJs have backgrounds that suggest that they will look skeptically at asylum claims. The core of immigration judges becomes slightly more liberal over time in our data, with the median asylum liberalism score increasing from 0.954 in 1990 to 1.061 in 2010. These averages are weighted based on the number of cases each IJ decides in our data.