Immigration Judges and U.S. Asylum Policy

Immigration Judges and U.S. Asylum Policy investigates hundreds of thousands of U.S. asylum cases with theoretical sophistication and empirical rigor, finding that immigration judges tend to assess legally relevant facts objectively while their decisions may be subjectively influenced by extralegal facts.

Immigration Judges and U.S. Asylum Policy

Banks Miller, Linda Camp Keith, and Jennifer S. Holmes

2014 | 248 pages | Cloth $69.95
Law | Public Policy
View main book page

Table of Contents

Chapter 1. Introduction
Chapter 2. Creating a Dataset
Chapter 3. A Cognitive Approach to IJ Decision Making
Chapter 4. Local Conditions and IJ Decision Making
Chapter 5. Appealing to the Board of Immigration Appeals
Chapter 6. The Policy Gap and Asylum Outcomes
Chapter 7. IJs and Reform of the U.S. Asylum System


Excerpt [uncorrected, not for citation]

Chapter 1. Introduction

In this book we seek to enhance understandings of why immigration judges (IJs) do what they do. We perceive IJs as the linchpin of U.S. asylum policy, and we assert in these pages that understanding how IJs decide asylum cases is the best place to begin trying to grasp asylum policy in the United States. In addition, the IJs offer an interesting case study from the perspective of scholars of judicial behavior because they decide cases in highly ideological fashion even though they are analogous to trial judges, a situation that is not often depicted in the literature. We attempt to move beyond the asylum literature's focus on disparities in grant rates as the primary criticism of the U.S. asylum process. Instead we focus on theoretical constructs—largely adapted from theories of judicial decision making—that allow us to better understand the conditional nature of IJ decision making. This approach leaves us with the overriding sense that eliminating disparities in IJ adjudication is akin to tilting at windmills—the causes of variation are too deeply embedded and the contexts in which decisions are made are too varied and influential. Instead, we implicitly focus on the quality of IJ decisions—a focus we make explicit in our final chapter, where we offer several concrete suggestions for improving the quality of decision making by IJs.

U.S. asylum policy represents a unique intersection of foreign and domestic policy. The implementation and adjudication of asylum cases triggers a wide range of potentially conflicting interests including international human rights norms, national security issues, geopolitical interests, border and immigration control, and the national and state economies. Asylum adjudications are enmeshed in a complicated web of domestic immigration law, U.S. treaty obligations, and federal jurisprudence. Recently, significant changes have been made in U.S. asylum law in response the terrorist attacks on the United States and in response to fears of economic migrants flooding the country. These changes have in turn triggered concerns that the law has become too draconian to meet our treaty or broader humanitarian obligations and ultimately may have left legitimate asylum seekers vulnerable. The individual adjudications of asylum law are made in an asylum system that spans two executive branches (the Department of Justice and the Department of Homeland Security) and that are overloaded and underresourced at every stage of the process. Components of the system have been politicized, such as the hiring process of IJs under Attorneys General Ashcroft and Gonzales and the ideological culling of the Board of Immigration Appeals by Ashcroft.

The key adjudicators—IJs—have generated a significant amount of controversy in regard to the consistency and fairness of their judgments, to such an extent that some critics have concluded that the decision of whether an individual gets asylum depends mostly upon the judge the individual draws. Both IJs and the Board of Immigration Appeals (BIA) have drawn criticism from the federal courts for the quality of their work. IJs in turn point to their crushing caseloads, limited support, and complicated cases, and limited independence from the Department of Justice (DOJ)—a point repeatedly made to multiple presidents and to Congress—most recently appealing to both the Senate and the House in the current attempts at immigration reforms. These issues have generated increasing concern and concomitant scrutiny of the asylum process and its various actors by commentators, activists, and scholars. In this book we engage in a theoretically driven, systematic, and rigorous examination of the core asylum adjudicator—the IJ. Our study allows us to empirically test the key criticisms and issues raised in regard to asylum decisions and U.S. asylum policy more broadly. Our theoretical underpinning allows us to offer explanations of IJ decision making that may inform future asylum policy or reforms.

In this book, we seek to better understand U.S. asylum policy by focusing on those whom we consider to be the most important, yet relatively unstudied, actor in the convoluted asylum bureaucracy: IJs. We have decided to focus on IJs for a host of reasons, some theory driven, some substantively driven, and others data driven. Most important from our perspective, IJs decide the majority of asylum cases in the United States and decide them with a significant degree of finality. The BIA, the body responsible for review of the decisions of IJs, reviews only 47 percent of the merit asylum decisions made by IJs (a high percentage to be sure) and upholds 74 percent of the decisions that they review. Combining these percentages means that 12 percent of the decisions made by IJs are reversed. Although this reversal rate may be high in comparison to the rates of Article III courts, nonetheless IJ decisions are usually final.

Below the IJs in the asylum bureaucracy are the asylum officers (AOs). They are the first to review affirmative asylum cases (we explain the distinction between affirmative and defensive asylum claims below). AOs tend to act as a rather permeable filter in the process, culling the cases in which it is clear that the claimant has a right to asylum and then passing the more difficult cases up the chain to the IJs. About one-third of the applicants who are eventually granted asylum receive it from AOs at this early stage of the process. In short, IJs decide the majority of asylum claims and virtually all of the cases in which asylum is not clearly due to the applicant. For these reasons, IJs tend to decide substantively more difficult cases than do the AOs.

Our interest is also in analyzing the individual decision making of the IJs as opposed to summaries of aggregate trends across courts or time, or comparisons within courts. This is because we believe, theoretically, that an understanding of the asylum bureaucracy starts with an understanding of the case-level decision-making process of IJs. That is, how do IJs decide each individual case before them? Which factors matter most and how? Thankfully, data are available on the decisions made by individual IJs in cases from at least 1990 onward. Such data are not publicly available, to our knowledge, for the AOs at all and, though individualized case outcomes are available for the BIA, data are not available on the decisions of individual members in the BIA. Therefore, given our theoretically driven focus on individualized decision making, it is not possible to study the AOs or the BIA in sufficient detail. We recognize that IJs make decisions at the midlevel in a complex bureaucracy, and that the context in which they make their decisions is crucial to understanding them. Therefore, we are able to utilize the aggregate statistics that we generated for both the AOs and the BIA where appropriate to ensure that we have properly contextualized the decision making of the IJs without also overcomplicating our desired focus. Furthermore, in Chapter 5 we focus on how the BIA affects IJ decision making.

We undertake the most thorough examination of the decision making of IJs ever attempted in the literature. To that end we analyze the IJs from a number of perspectives, beginning with a basic cognitive model of their decision making in cases that incorporate the extensive literature on judicial decision making as well as the concerns of international relations scholars. We engage in a cross-sectional analysis of all of the decisions of IJs in every asylum claim decided on the merits between 1990 and 2010, a total of over half a million claims. Although Congress has limited what data are available about applicants due to privacy concerns, we do know the form of relief requested, case number and outcome, applicant's country of origin and language spoken, and whether or not the applicant had a lawyer. From this, we are able to leverage country of origin characteristics to form a more complete picture of the applicant's case for asylum. We then expand our perspective to focus on how local demographic, economic, and political factors affect IJ decision making, drawing upon dominant theories within the broader immigration literature including variations of contact and threat theory. We next account for how the prospect of BIA review might influence IJ decision making. Finally, we analyze the aggregate trends in asylum decision making across the span of our data in an attempt to better understand the dynamic nature of policy interventions in the asylum process, drawing upon key policy perspectives within the immigration literature.

Our theoretically driven focus and rigorous empirical analysis allow us to generate a substantive understanding of asylum decision making that should speak directly to various policy makers considering a variety of proposals that have been advanced for reforming the asylum system and the broader immigration system. Critics argue that past policies, laws, and DOJ regulations promulgated in response to the threat of terrorists, undocumented immigrants, and an unwieldy asylum system have generated unintended consequences for both the asylum seeker and the U.S. asylum bureaucracy. We examine these expectations empirically and believe that these analyses can inform a discussion about the potential unintended consequences of currently debated reform. Indeed, this is one of the major benefits of our approach: because we have a well-developed theory of how the IJs decide cases, we have insight into which parts of that decision-making process are amenable to change and how policy makers might go about making the desired changes. We turn next to a brief summary of asylum law and the recent controversies that have surrounded IJs.

Asylum and Refugee Status in International Law

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 rights and obligations of states to refugees. 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. For an individual 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. 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 provided legal protection to only 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" (United Nations High Commissioner for Refugees [UNHCR] n.d., 2). 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. For U.S. policy makers at that time "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" (Betts, Loescher, and Milner 2012, 21). Instead it created "two other U.S.-led organizations that were parallel to and outside the purview of the United Nations" (Loescher 2003, 7 and see also Copeland 2003, 108-9) to limit the functional scope of the UNHCR.

Over time the legal norm expanded, in large part due to the efforts of the UNHCR, which successfully pushed for universal coverage for refugees, which was achieved with the promulgation of the 1967 Protocol eliminating the original geographical and temporal restrictions (Betts, Loescher, and Milner 2012, 30). The United States strongly supported and joined the 1967 Protocol. The move clearly signaled a change from its previous unilateral Cold War policies, which we discuss in the next section.

Today the right to non-refoulement is widely considered to be a part of customary international law and thus would apply even to states that are not formally a party to the 1951 Convention or Protocol (Goodwin-Gill and McAdam 2007, 345 and citations within). In 2001, the state parties to the Convention issued a declaration "reaffirming their commitment to the 1951 Convention and the 1967 Protocol" and "recognizing in particular that the core principle of non-refoulement is embedded in customary international law" (UNHCR n.d., 4).

U.S. Asylum and Refugee Law and Policy

Asylum law in the United States is domestic law that is "expressly based on international law (Anker 2011, 2). As Cianciarulo notes, the passage of the 1980 Refugee Act, "which sought to give statutory meaning to our national commitment to human rights and humanitarian concerns . . . ushered in a new era of refugee protection" (Cianciarulo 2006, 109-10). The 1980 Refugee Act was the first congressional act to specifically address refugees and asylum seekers. Previously, Congress had controlled immigration through rigid quota systems such as those implemented under the Immigration and Nationality Act of 1952 (INA), although presidents spanning FDR to LBJ had creatively skirted congressional limits by directing their attorneys general to use their power of parole to allow aliens, and thus unilaterally had admitted large numbers of refugees outside of the quota systems.

The 1980 Refugee Act repealed the INA's geographical and political limitations and lifted numerical caps on the number of annual asylum grants (Cianciarulo 2006). Cianciarulo argues that with the passage of the 1980 Act (along with the Supreme Court's recognition of the implications of the Act in INS v. Cardoza-Fonseca), "asylum was no longer an ad hoc, marginal immigration procedure entirely subject to the whims of policy" (110). Anker (2011, 17) argues that Congress's intent in enacting the legislation was "to conform provisions of U.S. law to the Refugee Convention." 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" (as cited in Gibney 1988, 111). However, this does not mean that it was entirely free of foreign policy or national security interests. Gibney (2004b, 157) argues that even after the 1980 law, "through State Department opinions, the ideological predilections of the Administration found their way into Immigration and Naturalization Service (INS) asylum decisions" (see also Ferris 1987, 126). We discuss this more in Chapter 3 when we examine the role of U.S. foreign policy interests in asylum outcomes.

Today, 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. Refugees apply for status outside of the United States, while individuals requesting asylum do so from within the United States or upon arrival at a port of entry. Refugee quotas and regional allocations are set by the president, who consults with Congress. In addition to INA eligibility, refugees must come from a country that is of "special humanitarian concern to the United States" and must not be resettled in another country or ineligible due to security, criminal, or other factors as determined by U.S. Citizenship and Immigration Services (USCIS) (Martin and Yankay 2012, 1-2). While asylum is discretionary under U.S. law, there are no numerical limits, as there are for refugees. The United States has been a major refugee receiving state. As Gibney (2004b, 132) notes, no Western state has admitted more refugees than has the United States since the end of World War II. However, refugee flows into the country have dramatically decreased, particularly since the September 11 attacks. Recently, the number of refugee admissions has subsequently begun a slight recovery, as seen in Figure 1.1, which plots the number of asylum grants by IJs compared to annual refugee inflow from 1990 to 2010. There is a striking negative relationship between the two series, which have (at the annual level) a correlation of -.91. Each series is plotted on its own scale, with the number of asylum cases on the right axis and refugee inflow on the left axis. Since these are the two paths through which individuals can escape persecution and seek to gain legal protection within the United States, it is reasonable to find that trends across the two potential routes to safety in the United States interact. Exploring this connection is beyond the scope of this work, but we plan to address the relationship in subsequent work that focuses on the broader U.S. refugee system.

The Asylum Process

Today U.S. law provides three treaty-based forms of relief or protection for individuals fleeing persecution: (1) asylum and (2) withholding of removal—which are based on the 1951 U.N. Refugee Convention and the 1980 Refugee Act—and (3) protection against return under Article 3 of the Convention Against Torture (CAT). The CAT narrowly prohibits the return of a person to another country if there are substantial grounds to believe he or she would be subjected to torture. Since 1999, the United States has also been bound under this obligation through the Foreign Affairs Reform and Restructuring Act of 1998. Each of the three forms of protection offers different levels of relief or benefits and each has somewhat varying legal thresholds that must be met. We discuss these fully in Chapter 2. Next we turn to the asylum process and the institutions and actors within the various stages of the process.

The asylum process today involves two executive departments—the DOJ and the Department of Homeland Security (DHS). Generally speaking, jurisdiction over the asylum process between the DOJ and DHS can be demarcated as follows: "DHS has jurisdiction over 'border' or credible fear interviews and first instance affirmative asylum applications (for persons who voluntarily apply before the institution of removal proceedings)" and "DOJ has jurisdiction over asylum applications determined in the course of removal proceedings, as well as over withholding of removal and applications for protection the Convention Against Torture" (Anker 2011, 12). In January 1983, the Executive Office for Immigration Review (EOIR) was created as a separate agency within the DOJ through an internal DOJ reorganization that combined the BIA with the IJ function previously performed by the former Immigration and Naturalization Service (INS), making the immigration courts independent of INS, the agency that at the time was charged with enforcement of U.S. immigration laws.

The EOIR is charged with administering immigration courts nationwide. The EOIR, located in Falls Church, Virginia, is headed by a director who reports directly to the deputy attorney general. Within the EOIR, the Office of the Chief Immigration Judge provides overall program direction of fifty-nine immigration courts throughout the United States and has administrative supervision for approximately 260 IJs. This number has been slowly increasing over time, in an attempt to reduce the backlog of cases. The BIA serves as the highest administrative tribunal adjudicating immigration matters, and has responsibility for interpreting and applying immigration laws nationally. The BIA is constituted by a directive of the attorney general and is authorized to have as many as fifteen members who serve at the pleasure of the attorney general. The BIA has broad authority to review decisions of IJs and does so through a paper review process.

IJs are administrative adjudicators who are formally appointed by the deputy attorney general; however, the EOIR and the Chief Immigration Judge handle their hiring. Current qualifications set by the attorney general require only that the candidates have seven years of prior legal experience. IJs arguably have less structural independence than federal judges and potentially less independence than administrative law judges. Nonetheless, they maintain a high degree of independence. The INA (Section 240) states that "in deciding the individual cases before them . . . IJs shall exercise their independent judgment and discretion." IJs (originally known as Special Inquiry Officers) act as trial-level judges at this stage with asylum hearings being somewhat adversarial in process if the applicant has an attorney.

The Asylum Officer Corps (AOC) was created as a part of the INS belatedly in October 1990—a decade after the 1980 Refugee Act and in response to a push for further reform of asylum in 1990 during the Bush administration; the AOC was "intended to ensure that political-asylum rulings are 'fair and sensitive'" (Koehn 1991, 232). AOs were placed under the authority of the DHS by the Homeland Security Act of 2002. They are now housed within the new USCIS, which has responsibility for enforcing federal immigration laws and administering immigration and naturalization benefits. There are eight asylum offices across the country, with over three hundred AOs serving.

A noncitizen who is physically present in the United States may seek asylum through either an affirmative or defensive process (see Figure 1.2). In the affirmative process, the applicant voluntarily identifies himself or herself through an application with the USCIS. The individual may or may not have valid status in the United States at the time of the application, but the application is not initiated during removal proceedings. In the affirmative process, once an application is filed, the applicant will receive notice to be fingerprinted and then will receive a notice to appear for an interview with an AO, who will review the application in a nonadversarial process in which the applicant must bring his or her own interpreter if desiring one. AO decisions are reviewed by a supervisor with the AOC. AOs are empowered to grant asylum, but the rates are rather low, as we can see in Table 1.1. The grant rate ranges from a low of 22 percent in 1997 to a high of 47 percent in 2000. As of 1995, if the application is denied and the applicant does not have valid immigration status, the applicant is referred to the immigration court for a de novo hearing and is now in removal proceedings. At this point the asylum seeker enters the defensive asylum process. The referral rates have ranged from 49 percent in 2000 to 73 percent in 1996.

In the defensive process, typically the noncitizen has been apprehended within the United States and is in removal proceedings in immigration court when the applicant makes application for asylum. A second stream of defensive applicants consists of aliens who arrive at a U.S. port of entry without proper documentation and who are placed in the expedited removal procedures that went into effect in 1998 under the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA), which we will discuss further in subsequent chapters. If these individuals express a fear of persecution, they are detained and receive a "credible fear" interview with an AO; otherwise the immigration officer at the port of entry can deny admission and summarily remove the aliens. If the aliens are found credible by the AO, the individuals are referred to an IJ for a hearing.

In the defensive process, applicants can apply for all three forms of relief if appropriate. The EOIR provides the applicants with an interpreter for the IJ hearings, but representation is not provided. If the applicants are not represented, the IJ guides them through the proceedings and advises them of their rights and options for relief or removal (Alexander 2006; Transactional Record Access Center [TRAC] 2009). Judges usually provide an oral decision at the end of each hearing due to caseload constraints. On the right side of Table 1.1 we report the IJ grant rates. In the last column we report the grant rates for any form of relief (including asylum, withholding of removal, or withholding of removal under the CAT), which range from a low of 17 percent in 1995 to a high of 58 percent in 2010. Of the three forms of relief, applications for asylum have consistently been the most successful, ranging from grant rates of 17 percent in 1995 to 49 percent in 2010. Withholding of removal has been granted considerably less often, with a high of 7 percent or 8 percent from 2005 forward, and relief under the CAT has rarely been granted thus far. Understanding these time trends is important in attempting to describe the current state of U.S. asylum policy, and we take up that analysis in subsequent chapters.

Both applicants and the DHS may appeal IJ decisions to the BIA, but the government appeals few cases in practice. IJ decisions are subject to the reasonableness standard of appellate review, which requires that the BIA find that the IJ reached unreasonable conclusions in order to reverse a decision. The facts of a case are reviewed for clear error by the BIA, whereas application of the law is reviewed de novo. Table 1.2 presents the annual number and percentages of asylum cases appealed in between 1990 and 2010, along with the percentage of appeals that favor the asylum seeker. As noted at the outset, about half of all applicants eventually seek review with the BIA, but only about one-fourth of those petitions that are reviewed are overturned. The peak year for asylum appeals was 2003, with just over twenty thousand cases appealed, although, as an overall percentage of decided cases, the rate is much higher before 1996 than after. Similarly, in general it appears that the BIA was friendlier to applicants in the early 1990s than at any other period in our data. We explore these trends more in Chapter 5. Board decisions may be referred to the attorney general for review at either the attorney general's request or the request of DHS. The attorney general may vacate any BIA decision and instead issue his or her own decision. Applicants for relief may seek judicial review of final agency decisions with the appropriate federal circuit court. In the past decade, it has been the voice of federal circuit judges (joined by the media and the law profession) that has raised serious questions about the asylum process, specifically in relation to IJs.

Immigration Judges: Controversies and Issues

Law professors and the media have increasingly scrutinized immigration courts and the behavior of IJs. This attention has been fueled by "federal appeals courts around the country complain[ing] of a pattern of biased and incoherent decisions on asylum and rebuk[ing] some immigration judges by name for 'bullying' and 'brow-beating' people seeking refuge from persecution" (Bernstein 2006; see also Legomsky 2007). In March 2007 the New York Times reported confirmation from the EOIR that "11 of the nation's roughly 215 immigration judges had been temporarily suspended from courtroom duties since June, 'based on concerns about how they were conducting immigration proceedings'" (Bernstein 2007). Bernstein's story of New York IJ Jeffrey Chase illustrates some of the frustrations experienced by IJs as well as the attorney general and the federal circuit courts. Judge Chase, who was a former human rights advocate and chairman of the American Immigration Lawyers Association's National Asylum Reform Task Force, had "rallied on behalf of people from China seeking asylum," but as Bernstein notes "before long, incredulous tirades became his trademark in many Chinese asylum cases, according to court records and interviews with a dozen lawyers" as his frustration grew over "a pattern of boilerplate claims that he suspected had been concocted by smugglers." The Second Circuit Court of Appeals rebuked Judge Chase in multiple decision for "'pervasive bias and hostility,' 'combative and insulting language,' and remarks 'implying that any asylum claim based on China's coercive family planning policies would be presumed incredible' and criticized the judge's decisions for 'a plethora of errors and omissions'" (Bernstein 2007, n.p.). Judge Posner of the Seventh Circuit Court of Appeals has been a frequent critic of IJs. TRAC notes in particular Posner's decision to vacate a particular IJ decision "based on six 'disturbing features'" in which he expressed the view that "these disturbing features bulk large in the immigration cases" (TRAC 2006, n.p.). While Judge Posner cautioned that the cases before him may not be completely representative, TRAC concluded differently:

TRAC's analysis of the decisions of most of the nation's immigration judges about tens of thousands of different asylum cases, however, provides powerful evidence that the problems of the immigration court go far beyond the failings of a few rotten apples—the individual judges criticized by Attorney General Gonzales. Rather, the examination of the case-by-case records appear[s] to document a far broader problem: a long-standing, widespread and systematic weaknesses in both the operation and management of this court. (n.p.)
The criticism of IJs has been compounded by reports of significant variation in grant rates across judges, even among those serving on the same courts (Legomsky 2007; Ramji-Nogales, Schoenholtz, and Schrag 2007, 2009). For example, the U.S. Government Accountability Office (GAO) noted an example in 2008 in which the likelihood of receiving a grant of asylum from the IJ most likely to grant asylum was 420 times greater than the likelihood of receiving asylum from the IJ least likely to grant asylum in the same court (GAO 2008, 34). The U.S. Commission on International Religious Freedom (2008, 115) concluded that outcomes of individual asylum claims have come "to depend largely on chance; namely, the IJ who happens to be assigned to hear the case." These disparities across courts and across judges have raised significant questions about the quality and consistency of justice in immigration courts, and for many observers and legal practitioners they present a frustrating dilemma.

On the other hand, IJs and their national association have sought to draw attention to the serious institutional constraints under which they work. IJs face an unrelentingly harsh workload. The chief judge of the Second Circuit reported that IJs must finish at least five cases per business day to stay current. TRAC reports that IJs typically handle sixty-nine cases a week and must dispose of twenty-seven cases per week (TRAC 2011). And despite recent reforms, IJs face a growing case backlog with a new all-time high backlog of 267,752 at the end of December 2010 (TRAC 2011). IJs complain about the "the constant drumbeat of case completion goals," the requirement that judges must "rule promptly at the end of the hearing in the form of lengthy, detailed and extemporaneous oral opinion with little or no time to reflect or to deliberate," that making credibility determinations "is extremely, extremely difficult," and, most important, that "there is not enough time to do research and adequately read about country conditions" (Lustig et al. 2008, 65-66). In addition, IJs typically have little staff assistance; most courts are not assisted by a clerk or bailiff, and the judges often have to operate their own tape machines. Clearly, these are overburdened and underresourced courts with high stakes for applicants. The president of the judges association recently complained that "for some people, these are the equivalent of death penalty cases, and we are conducting these cases in a traffic court setting" (as cited in Becker and Cabrera 2009, n.p.). In subsequent chapters we offer an explanation for how these circumstances influence the way IJs make decisions and in part shape the disparities in granting asylum for which the IJs are often criticized. This work builds upon a nascent body of empirical literature that we address next.

Empirical Research on U.S. Asylum Outcomes

Until recently, the empirical literature on U.S. asylum has largely been oriented toward international relations and thus has consistently focused on the relationship between the United States and asylum-sending states rather than the individual decision makers. The theoretical perspectives with which asylum outcomes have been approached in most of these studies have assumed that the state is a unitary actor, and thus the empirical analysis has focused on aggregate levels of asylum granted to persons fleeing dominate asylum-sending states. The early asylum literature in political science demonstrated the dominant role of geopolitical and material interests over human rights in U.S. asylum policy (Gibney 1988; Gibney and Stohl 1988; Gibney, Dalton, and Vockell 1992; Loescher and Scanlan 1998). Subsequent studies that have examined the effects of social and economic conditions of applicants' countries of origins have generally concluded that U.S. interests influence grants more than human rights conditions (Hassan 2000; Gibney 2004a) and that applicants perceived to be economic migrants are more likely to be rejected. However, recent empirical studies paint a more nuanced picture, suggesting that while national interests may trump human rights concerns, both national interests and normative considerations influence these outcomes (Rosenblum and Salehyan 2004; Salehyan and Rosenblum 2008; Keith and Holmes 2009; Rottman, Fariss, and Poe 2009; Holmes and Keith 2010) in addition to domestic politics (Salehyan and Rosenblum 2008). Most researchers working on asylum outcomes in Western European states have also resorted to aggregate grant rates (for example, Holzer and Schneider 2001; Neumayer 2005; Hamlin 2012), but Holzer, Schneider, and Widmer's (2000) study of asylum applications in Swiss cantons is a notable exception.

In our previous work, we focused on a more appropriate level of analysis, the individual IJs' decisions (Keith and Holmes 2009; Holmes and Keith 2010; Keith, Holmes, and Miller 2013; Miller, Keith, and Holmes 2013). This move to the level of the individual decision maker is an important theoretical break with unitary actor assumptions in the international relations (IR) literature. Similarly, Ramji-Nogales, Schoenholtz, and Schrag (2007 and 2009) examined IJ decisions, though they studied individual AO and IJ aggregate grant rates rather than the actors' particular case decisions. Congressionally mandated data restrictions have made it impossible for nongovernmental bodies to have access to individual case factors such as applicant characteristics and types and level of evidence presented. In a rare field study of an NGO's asylum cases, we (Keith and Holmes 2009) found that none of a variety of evidentiary factors mattered in the judges' individual decisions once conditions that signaled U.S. interests were controlled. We are aware of only one other study that was able to gain access to such data. Koehn (1991) accessed fifty-nine individual files of Ethiopian asylum seekers in the New York and Washington, D.C., courts. And in our 2010 (Holmes and Keith) study of individual IJ decisions, we found that September 11 shifted the way in which judges assessed certain factors such as whether the applicant spoke Arabic or was from a state either that sponsored terrorists or in which al-Qaeda was present.

While these studies have engaged the more appropriate individual level of analysis, they, like the rest of the asylum literature, do not offer a theoretical understanding of how the individual judges make these decisions, a crucial thread that we began to explore (Keith, Holmes, and Miller 2013). In that article we presented a variation of the attitudinal model of judicial behavior that we modified by incorporating a cognitive model of decision making. There we argued that some pieces of information before IJs in asylum cases are treated more objectively while others are treated more subjectively. This theoretical model allowed us to account for informational cues that influence IJ decisions while at the same time assessing the impact of national interests and human rights conditions. We continue to expand upon this theoretical approach and empirical exploration in subsequent chapters. Finally, most of the literature has conceptualized the dependent variable as a dichotomous decision, which our subsequent work (Miller, Keith, and Holmes 2013) began to question, given the more nuanced nature of U.S. asylum law, which presents the IJs with different standards of evidence depending upon the form of relief requested and different levels of benefits to the applicant depending upon the form of the relief granted. We explore this issue further in Chapter 2. Our book fills several significant gaps in this empirical literature and makes a significant contribution to the broader literature dealing with U.S. asylum law and policy. In the next section we discuss the theoretical perspectives of each component of our analyses. We also give a brief overview of our findings and their implications.

Our Contribution to the Literature

First, we created a comprehensive new dataset on IJ decision making that spans two decades and contains the most complete set of covariates of which we are aware. In creating the dataset we addressed core limitations in data examined in most of the empirical literature on U.S. asylum decisions. For example, for a variety of reasons most of the asylum studies in the IR literature have focused on aggregate grant rates. And even though empirical studies within the law literature have focused more appropriately on the individual asylum case and IJ decisions, this literature has tended to severely truncate the population of IJ asylum decisions, in terms of temporal dimension, the selection of high volume immigration courts, or a focus on select countries of origins or only affirmative applications. We created a more complete dataset, which we present in Chapter 2 and that examines all asylum cases decided on the merits between 1990 and 2010. We also created or gathered a significant set of case characteristics that allow us to leverage more information in multivariate analyses. Thus, we are able to specify our models to test assumptions about the countries of origins, the immigration courts, and various factors relating to the applicant. Another significant innovation in our dataset is that we move beyond the traditional treatment of the asylum outcome as a dichotomous choice in which the applicant is granted relief or not (see, e.g., Ramji-Nogales, Schoenholtz, and Schrag 2007; Rosenblum and Salehyan 2004; Salehyan and Rosenblum 2008; Rottman, Fariss, and Poe 2009; Holmes and Keith 2010; Keith, Holmes, and Miller 2013). While we acknowledge that this approach to the asylum process is reasonable, here we explore the potential loss of information that is theoretically and statistically important. We demonstrate in Chapter 2 that asylum outcomes are sometimes more appropriately operationalized a polychotomous choice that reflects an ordering of benefits that adhere to the applicants, as opposed to a simple grant or deny decision or a legal ordering of the standards applied in the various forms of relief.

One of our most significant innovations vis-à-vis the asylum decision making literature is to create a measure of the ideology for IJs. To create our measure of policy predispositions, in Chapter 2 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 (although we occasionally refer to it as a measure of ideology for convenience). This variable becomes the core component of the cognitive model we present in this book. In addition to gathering a wealth of information on the factors that affect the decision making of IJs, we also collected data on which cases were appealed to the BIA as well as data on many dimensions of the local and national economic and political environments, which are theoretically important in understanding individual asylum decisions and U.S. asylum policy. In Chapter 2 we give an in-depth description of the data and the processes we employ to explore and create the new measures such as the ordered dependent variable and IJ ideology score, and we present an initial descriptive picture of the data and salient trends in asylum outcomes.

Second, we draw upon two disparate disciplines and their literatures to present a model of cognitive decision making for IJs. The law literature has been primarily driven by the assumption that international and U.S. asylum law will be applied consistently and fairly. This literature has largely responded normatively to the evidence of significant disparity and inconsistency in grant rates within the asylum system. The IR literature, with its focus on the relationship between states, has largely been concerned with the foreign policy dimension of U.S. asylum decisions, along with the normative and behavioral questions that are raised by connection. These two literatures have largely operated in isolation from each other, and at the same time both literatures have failed to draw upon a third body of work, the judicial behavior literature, that greatly enhances understanding of asylum outcomes.

In Chapter 3 we integrate these approaches to leverage their insights and supplement them with a variation of the attitudinal model from the judicial behavior literature that draws upon insights from psychological models of decision making. We believe that a comprehensive model of IJ decision making approximates a situation in which an IJ, under tremendous time pressure and unsure of the credibility of an asylum seeker, will use policy predispositions to help process both legally relevant and legally irrelevant facts. By incorporating a cognitive model of decision making, we show how IJs consider some pieces of information objectively while other information is treated subjectively. This approach allows us to account for informational cues that influence decisions while assessing the impact of national interests and human rights interests as reflected in the IR debate. In brief, we draw upon Braman and Nelson (2007), among others (e.g., Simon 2004; Bartels 2010), and argue that the inherent lack of certainty in legal standards, the frequent lack of corroborating evidence, and the complexity of credibility determinations invite the use of motivated reasoning in which policy-motivated goals overwhelm accuracy goals. More specifically, directional or policy goals will induce what Bartels (2010) has termed top-down decision making in which the judge brings a theory or predisposition to the case. In contrast, in a bottom-up approach facts are evaluated deliberately and decision making is constrained by the manner in which the law dictates the treatment of those facts. In essence, we posit that the policy preferences of IJs influence their decisions in asylum cases, but that U.S. asylum law also imposes some constraints on the use of the policy proclivities of an IJ. We explore the theoretical links fully in Chapter 3.

We find that the influence of these competing factors is conditional upon a judge's policy preferences toward asylum. More specifically, we find that the policy predispositions of the IJs play a dominant role in explaining the discrepancies in asylum grant rates that have disturbed law professors. Moreover we find that liberal IJs respond to certain nonlegal factors differently than their more conservative colleagues—findings that strongly implicate norms of fairness and consistency. At the same time, we also find some limited evidence that the law constrains the decision making of the IJs with respect to applicant characteristics. The analyses in Chapter 3 have theoretical implications and implicate several salient policy issues. In terms of theory, the analyses expand and improve upon the compliance literature by examining actual on-the-ground state actors, who engage in behavior controlled by U.S. international commitments. While previous studies in the IR literature have shown the role of both national interests and humanitarian concerns, our cognitive model offers explanations of how these factors play out in the IJs' decision-making process. Thus, the underlying implication for political science is the importance of cross-discipline research, in the sense that IR theories benefit from the introduction of judicial behavior theories by specifying an underlying causal mechanism for many of the central IR findings in what is perceived as "state" action. In regard to the work of law professors, the research in Chapter 3 strongly emphasizes the need for theory to understand data. To our knowledge, this is the first time the variation in grant rates among IJs has been examined using theoretical approaches from the study of judicial behavior. We also make two contributions to the judicial behavior literature. First, we offer an extensive exploration of the application of the attitudinal approach in a trial-level-like context where policy preferences are not thought to explain a good deal of variation in judicial decision making. Here we find that attitudes can play a prominent role in decision making given the right context (see Zorn and Bowie 2010). Second, we offer an application of psychological approaches to understanding judicial choices. Our cognitive model is innovative because it tells us not only that policy preferences matter, but also how and when they matter. In this way we also provide some evidence for the constraint of law in judicial decision making, which some scholars characterize as a central question in the literature (e.g., Braman 2010; Segal and Spaeth 2002).

Within the asylum literature there is no work that we are aware of that examines whether local economic or demographic conditions have an impact on asylum decisions, unlike the broader immigration literature's competing threat and contact theories that are very much focused on the local context. In this literature the primary debate centers on whether increased interaction with immigrants can either incite hostility or promote acceptance. We expand our cognitive model in Chapter 4 to integrate expectations derived from long-standing debates within the immigration literature to further understand the asylum decisions of IJs. We argue that understanding local influences on IJ decision making is important if we are interested in understanding and reducing the variability between adjudicators and across immigration courts. In the context of IJ decision making a very particular set of local conditions is likely to influence the IJ's decisions, and these extralegal factors, similar to those related to material and security concerns we examine in Chapter 3, are likely to be evaluated in a more top-down fashion or more subjectively than legal factors and thus will be contingent upon the IJ's policy predispositions. In other words, we believe that IJs might weigh these locally situated extralegal factors differently, depending on their own ideological proclivities.

We find evidence that IJs are responsive to local conditions in highly conditional ways in that liberal and conservative IJs react differently to the same set of local conditions. This finding has important implications as not just locality influences the chances of an asylum seeker being granted asylum, but also within a community the random draw of a judge influences how those local conditions affect the applicant's chances of relief. If IJs not only react to different local conditions but also react differently to the same local conditions, then significantly reducing the variability in their decision making may be more difficult than first contemplated.

We also broaden our focus outside the immigration court to the BIA, which serves as the highest administrative tribunal adjudicating immigration matters. For many asylum seekers, the BIA has historically been their last and best chance to challenge a final deportation order. For IJs, the BIA is the most immediate venue of review in which their decisions have historically faced a real possibility of being overturned or remanded; however, due to significant reforms in the DOJ, IJs are now considerably less likely to have their decisions disturbed than they were previously. Our contribution here is primarily in examining how the behavior of the BIA might alter the decision making of the IJs and asylum applicants. In Chapter 5 we first examine the institutional context in which the BIA operates, particularly examining changes made by recent attorneys general in regard to the BIA's structure and its methods of decision making. These reforms have been criticized within the legal literature and by practitioners and adjudicators at all levels including the federal circuit courts. In Chapter 5 we present a theoretical model for the asylum seekers' decision to appeal to the BIA drawing upon the rational actor approach, and we find strong evidence of rationality on the part of asylum applicants.

In Chapter 5 we also examine BIA outcomes themselves in what we believe is one of the more rigorous statistical examinations in the literature. We approach BIA decisions from a functional perspective, as our interest in this book focuses primarily upon the applicant and the IJ. Thus we are more interested in how the BIA informs our understanding of the individual asylum seeker's behavior and more important how it may shape IJ behavior. Drawing upon the judicial behavior literature, we examine the BIA decisions from an error correction perspective as well as from a legal policy making perspective. While we find support for both functions, but we find stronger support for the policy making function of the BIA over its error correction function. Our findings in regard to this executive branch appellate institution are consistent with the judicial behavior literature.

Another contribution we make in Chapter 5 is to enhance our understanding of the role of the attorney general in the asylum process through the creation of regulations concerning the structure of the BIA and its procedures. We demonstrate that Ashcroft's streamlining reforms clearly changed the outcomes of the BIA and had implications for both the policy-making function as well as the error correction function of the BIA. We also find evidence that overall the Ashcroft streamlining reforms appear to have polarized the BIA ideologically. We then are able to apply this understanding to our cognitive model of IJ decision making. Here we find some evidence that the IJs behave strategically in response to institutional changes put into place by Ashcroft and Gonzales, but once the streamlining reforms are instituted and affirmance becomes the norm with the BIA, IJs resume voting their policy preferences.

One of our biggest contributions to the asylum literature is to broaden the perspective beyond that of the asylum applicant, in which we focus on the probability of individual relief, to that of the policy maker and system outputs. In Chapter 6 we examine key statutes passed by Congress in the past two decades in regard to asylum. Following upon the 1993 World Trade Center bombing, Congress passed IIRIRA, which substantially overhauled the asylum process. Then following the September 11 attacks Congress passed the Real ID Act, which sought to put tighter restrictions on potential terrorist migrants and those coming to the United States for economic, as opposed to humanitarian, reasons. We turn to the broader immigration literature and examine the laws' intent and consequences from the policy gap perspective. This approach highlights "significant and persistent gaps [that] exist between official immigration policies and actual policy outcomes" (Cornelius and Tsuda 2004, 4) that stem from the embedded economic realities of U.S. labor market needs or from principal-agent problems in regard to implementation. In Chapter 6, to test the policy gap perspective, we identify the intent of the two statutes by examining congressional debate and public statements. We find that the rhetoric largely links the statutes to the threat of terrorism and to reducing potential abuse of the asylum system by economic migrants. We then examine our expectations in a monthly time series models, focusing on whether the intended consequences materialize, controlling for the potentially intervening shocks of the World Trade Center bombing and the September 11 attacks, along with key factors that we know from the previous chapters are likely to influence asylum outcomes. We also examine the human rights critics' concerns that the effects of the laws are draconian and have made it difficult for bona fide asylum applicants to avail themselves of the protection promised under U.S. statutory and treaty commitments. In Chapter 6 we find overwhelming evidence to support the policy gap expectation of unintended consequences. Specifically we find that IIRIRA and Real ID had net positive effects on the number of applicants granted relief, not the negative effect that critics had feared. Our findings suggest that the two laws likely increased the quality of applications reaching IJs.

In Chapter 7 we summarize our results and talk about what they suggest is likely to work, and how, when it comes to reforming the asylum decision-making process in the United States. Our focus in this final chapter is on applying our empirical insights to determine where change is likely both to be feasible and to have the intended consequences. Our conclusion there is that, generally speaking, eradicating or significantly reducing variation in outcomes for asylum applicants is likely to be difficult. We instead seek to focus attention not on the outcome but on the process of decision making. We do so by emphasizing that the quality of the determination made by an IJ should be of more concern than the actual outcome since it is likely impossible to know whether any particular applicant should be judged credible or incredible. Of course, assessing the quality of any given decision is also no easy feat (Mitchell 2010; Wistrich 2010) and considerable work, both theoretical and empirical, remains to determine how we might assess quality from outside the actual judging process. We offer some tentative suggestions for reforms that are likely to improve the quality of the decisions made by IJs, with perhaps the central recommendation being that all asylum determinations on the merits should include some brief written opinion.