UNDERSTANDING ASYLUM OUTCOMES FOR ZIMBABWEANS
By Prof. Norma Kriger (Independent scholar and consultant)
A quarter of Zimbabwe’s population now lives in other southern African countries, the UK, the US, or elsewhere. To avoid being deported to crisis-ridden Zimbabwe, many Zimbabweans have learned to navigate immigration and asylum bureaucracies. My own encounters with these bureaucracies – as a researcher on foreign migrants in SA in 2006-7 and as an occasional expert witness in Zimbabwean asylum cases in the US – made me curious about how Zimbabwean asylum seekers in the US fare compared to their counterparts in South Africa. It turns out that more Zimbabweans have been granted asylum in the US in the past decade than in SA, even though the vast majority of Zimbabweans seek asylum in South Africa – ninety percent of Zimbabweans seeking asylum globally applied in SA in 2009. How can one account for so many more Zimbabweans obtaining asylum in the US despite the overwhelming majority applying in SA? Given that South Africa and the US are the two top destinations for asylum seekers, it is important to understand why applicants from one national group – and Zimbabweans happen to be the leading nationality globally among new individual asylum applicants – have had such different experiences in these two countries.
The US has much higher aggregate asylum grant rates, regardless of nationality, than SA. While both systems have been severely critiqued as broken or dysfunctional, at least three factors would contribute to making the US a preferable system from the point of view of the average asylum seeker: a decision-making process that, despite its flaws, largely respects asylum law, allows applicants the right to legal representation at their own expense in first instance determinations (even if the majority are unable to afford legal counsel), and is not tainted by society-wide xenophobic attitudes (even if anti-immigrant attitudes exist). In contrast, decisions in SA are characterized by extraordinary disregard for asylum law, compounded by the influence of xenophobic attitudes in society, and applicants have no right to legal representation in first instance determinations.
Beyond systemic differences that confront asylum seekers of all nationalities in both the US and SA, special circumstances account for the very low grant rates of Zimbabwean applicants in SA, especially when compared to those for applicants from the Great Lakes region and the Horn of Africa. The pro-Mugabe and pro-ZANU PF bias in SA’s foreign policy appears to have led asylum bureaucrats in SA to perceive Zimbabwe as a country that does not produce legitimate refugees, thereby resulting in discrimination against Zimbabwean applicants. In the US, Zimbabwean asylum applicants have had higher than average grant rates. There is no evidence that US foreign policy toward Zimbabwe – vigorously anti-Mugabe and anti-ZANU PF – has influenced US adjudicators of Zimbabwean asylum cases, though it would be an interesting research question.
Grant rates for Zimbabweans in SA are also low because proximity enables Zimbabweans to enter SA en masse, many Zimbabweans (and other national groups) use the asylum system without a legitimate claim to a well-founded fear of individual persecution as it is the only avenue available to most to legalize their status and obtain the right to work in SA, and the threat of deportation for failed Zimbabwean asylum seekers is not a disincentive to applying for asylum. Since April 2009 there has been a moratorium on deportation of Zimbabweans, and even prior to that date, deportation did not deter Zimbabweans from re-entering SA without authorization. Conversely, Zimbabweans who are able to apply for asylum in the US are a small group whose wealth and education set them apart from most Zimbabwean migrants and arguably contribute to their higher than average grant rates. Also, asylum seekers in the US usually do not have the right to work as soon as they apply for asylum and the cost of failed asylum is nearly always deportation with dim prospects for re-entry – disincentives to entering the asylum system that may reduce the number of fraudulent applications and contribute to higher than average grant rates for all nationalities in the US than in SA.
Comparative analysis of this kind is fraught with challenges. Multiple variables may explain asylum outcomes. Each asylum system operates in a distinct macro-environment, the systems themselves are different, and applicants who make it from Africa to the US likely differ from their compatriots who stay in Africa. Comparisons of grant rates across the two countries’ asylum systems are difficult to make because the official statistics measure asylum grants differently. For example, the SA data on asylum grants refers to individual applicants and does not include their dependants. US immigration court data also refers to individual cases only but asylum grants to individuals who go through an interview process with Asylum Officers refer to the principal applicant and his or her dependants. Thus even within the US system, the official statistics do not permit a straightforward comparison of the numbers of applicants granted asylum in immigration courts as opposed to asylum office interviews. Moreover, for US immigration courts, there is data on the total number of asylum applicants received annually and by nationality. From 2005, comparable data is not available publicly for asylum applicants who interview with the Asylum Office. It is also not possible to compare grant rates for the two asylum routes in the US from 2005. Beyond limitations for comparing cross-country and even intra-country grant rates, the SA government cautions about the flaws in its asylum data which does not even cover all the offices that grant asylum. To add to the challenge, inferring causation from aggregate data analysis has its own potential pitfalls. Nonetheless, the comparative analysis that follows illuminates some crucial differences confronting Zimbabwean asylum seekers in the US and SA, starting with the different asylum systems in the two countries.
US and SA asylum systems
The US and SA asylum processes have significant differences. The SA asylum system is new relative to the US system. While the US asylum system has improved over the years as a result of ongoing interactions between NGOs and the government, the SA asylum system remains at an early stage in its institutional development.
In the US, separate bureaucracies deal with the applications of refugees (who apply from outside the US) and asylum seekers (who apply from inside the US). The US provides two routes for asylum seekers to obtain asylum or become “asylees”: an interview with an asylum officer in the Department of Homeland Security or a removal proceeding in an immigration court in the Department of Justice. In SA, those seeking refuge must enter the asylum process inside the country. Post-apartheid SA considered creating immigration courts but opted not to. Instead, refugee status determination officers (senior administrative clerks) in the Department of Home Affairs grant or deny asylum and successful applicants become “refugees”. Both the US and SA asylum systems have procedures for appeal and judicial review. In the US – but not in SA – asylum applicants have the right to legal representation at their own expense in first instance determinations. For all these differences in the two countries’ asylum processes, in both systems decision-makers must accept or reject each asylum application on its individual merits by taking into account the credibility of asylum claims in the context of broadly similar refugee law. SA is unusual in Africa for its case-by-case approach to refugee status determination. The following brief descriptions of the US and SA decision-making processes for asylum applicants omit many important legal and policy issues but aim to provide the essential similarities and differences.
US asylum system
The US Immigration and Nationality Act, 2000, and the Refugees Act, 1980, distinguish between refugees and asylum seekers. An applicant for refugee status is outside the US, while an applicant seeking asylum status is in the US or at a US port of entry. During 2009, more than 73,000 persons were admitted to the US as refugees and over 21,000 individuals were granted asylum. This discussion focuses only on the adjudication process for asylum seekers in the US.
In the US, asylum may be obtained in two ways: affirmatively through an interview with an Asylum Officer in the Department of Homeland Security’s Office of Citizenship and Immigration Services (USCIS) or defensively through removal proceedings in immigration court in the Executive Office for Immigration Review (EOIR) of the Department of Justice. Removal proceedings are administrative processes: neither immigration courts nor immigration judges are part of the judiciary. In 2010, over 11,000 people were granted asylum affirmatively and less than 10,000 individuals were granted asylum defensively by an immigration judge or the Board of Immigration Appeals.
An affirmative applicant takes the initiative to seek asylum, voluntarily submitting an application to one of the eight urban-based Regional Asylum Offices in the DHS. Several weeks later, the applicant receives a date for an interview with an Asylum Officer. The interview is non-adversarial. Usually a few weeks after the interview, the applicant will be informed of the outcome. If an applicant in a valid immigration status (e.g. foreign student, visitor’s visa) fails to establish eligibility for asylum during the interview by the Asylum Officer, the application will be denied and the applicant will retain his or her lawful status. If the applicant does not have a valid status and the Asylum Officer finds the applicant’s claim to be ineligible, the Asylum Office places the applicant in removal proceedings before an immigration judge. The Asylum Office serves the applicant with a “Notice to Appear” in that court on a specific date. The notice to appear is the equivalent of a summons in a civil case.
Noncitizens may also be put directly in removal proceedings by immigration enforcement officials because they are undocumented, are in violation of their status when apprehended, or are caught seeking entry into the US without proper documentation. Applicants put directly in removal proceedings are usually detained (jailed) by DHS after being apprehended and most remain detained through their hearings and any subsequent appeal. Removal proceedings are adversarial, with DHS trial attorneys opposing asylum claims before the immigration courts. The immigration judge may grant asylum or deny asylum and issue an order of removal. An asylum applicant may appeal a denial to the Board of Immigration Appeals, also within the Department of Justice, and seek further review in a US Court of Appeals.
Asylum decisions made by immigration court judges and asylum officers are discretionary. Adjudicators weigh all the evidence and other factors and decide purely on the basis of their judgment, albeit within a framework of law and judicial opinions interpreting the law. Applicants must prove that they have “a well-founded fear of persecution” in their home country on account of at least one of five internationally recognized grounds: their race, religion, nationality, political opinion, or membership in a particular group. These grounds for asylum incorporate the definition of refugees in the 1951 UN Convention relating to the Status of Refugees and the 1967 Protocol. Moreover, the applicant must have applied for asylum within one year after last entering the US, and must not be subject to any of the bars to asylum. The applicant’s claim must be credible to the adjudicator. Applicants have the right to legal representation at their own expense both during asylum office interviews and immigration court proceedings. Applicants who cannot afford representation in the US can in some instances find pro bono counsel or reduced rate lawyers to represent them in first instance proceedings and on appeal.
SA asylum system
The asylum system in SA is governed by the Refugees Act, 1998, which came into operation in 2000 and is administered by the Department of Home Affairs. The only route to become a refugee in SA is to apply for asylum from inside the country.
An asylum seeker who enters the country at an official border post is legally entitled to request asylum from the border official and to receive a temporary transit permit. The transit permit is intended to enable the applicant to reach a Refugee Reception Office (RRO) where the applicant must lodge a claim for asylum. One may enter the country irregularly and therefore not obtain a transit permit but still have the right to apply for asylum. All four permanent Refugee Reception Offices are located in major cities – the fifth closed in June 2011 following a High Court order. There are also two relatively new interim offices, one located in Musina, a town near the Zimbabwe-South Africa border, and another in Pretoria/Tshwane.
The Refugees Act and Regulations lay out the duties of the refugee reception officer at the RRO to assist the applicant in completing the application form, issue the applicant an asylum seeker permit that includes written notice to appear before a refugee status determination officer on a specified date, and renew the permit whenever its validity period is set to expire by the refugee reception officer (generally every three months) until the applicant has been awarded or denied refugee status.
The refugee status determination officer must interview the applicant and process his/her application in accordance with rules on just administration, as outlined in the SA Constitution. Generally, the interview must take place within one month and a written decision must be issued within 6 months of submitting the claim. If approved, the Department of Home Affairs issues the asylum seeker with a refugee permit or formal refugee identity document. In contrast to the US, refugee status in SA is not permanent and is also not an easy pathway to citizenship.
If the application is rejected because the claim is made on grounds other than those specified in the Refugees Act, the decision is automatically reviewed by a Standing Committee on Refugee Affairs. An asylum seeker does not appear before the Committee but has the right to make written representations. If the claim is rejected because it is considered inadequate within the terms of the Refugees Act, the applicant may appeal to the Refugee Appeal Board and has the right of legal representation at his/her expense. This is the first time in the process that the applicant has the right to legal representation. If the board upholds the refugee status determination officer’s negative decision, the asylum seeker may seek judicial review by a High Court.
Each asylum application for refugee status is decided by Department of Home Affairs’ personnel on its individual merits in terms of the criteria provided in the Refugees Act. To be granted refugee status, an asylum seeker must show persecution or a well-founded fear of persecution based on race, religion, nationality, political opinion, membership in a social group – the same five grounds as in the US – or tribe. In addition, SA refugee law grants refugee status to a person compelled to leave his or her country as a result of external aggression, occupation, foreign domination, or events seriously disturbing or disrupting public order in either a part or the whole of his/her country of origin. This provision draws on the 1969 Organization of African Unity Refugee Convention. As in the US, SA law denies asylum status to certain categories of people, including those who have committed serious crimes.
The numbers: Zimbabwean asylum seekers in the US and SA
South Africa was the main destination for new individual asylum applications worldwide in 2009 with over 222,000 applicants – over 25% of all such applications globally. The US, with less than 48,000 new applications – roughly one-fifth of the number of new individual asylum claims as SA – was in second place. By nationality, the highest number of new asylum claims in SA in 2009 was lodged by individuals originating from Zimbabwe (158,000), followed by applicants from Myanmar (48,600), Eritrea (43,300), and Ethiopia (42,500). Ninety per cent of Zimbabwean asylum claims globally were lodged in South Africa in 2009. Despite this, South Africa has granted refugee status to fewer Zimbabwean asylum seekers than the US. Asylum grant rates for Zimbabweans in SA are very low compared to aggregate immigration court grant rates – overall immigration court grant rates for all nationalities – and immigration court grant rates for Zimbabweans in the US. [Affirmative asylum data on aggregate grant rates and grant rates by nationality are not available.] Acceptance rates in SA for Zimbabwean asylum applicants are also much lower than both aggregate grant rates and grant rates for applicants from the Horn of Africa and the Great Lakes region.
Zimbabwean asylum seekers in SA
Zimbabwean asylum seekers have put tremendous strain on the already fragile asylum system in South Africa and have grown – exponentially since 2008 – as a percentage of the total number of new applicants per year. Between 2000 and 2004, there were 136,000 new asylum claims lodged in SA; roughly 10,000 (or over 7%) of these were submitted by Zimbabweans. In 2006 and 2007, new Zimbabwean asylum applicants constituted over one-third of total new asylum applicants. Zimbabweans became the single largest group by nationality of new asylum seekers in 2006.  The number of new Zimbabwean asylum applicants per annum then soared from under 18,000 in 2007 to nearly 112,000 in 2008 and then almost 150,000 in 2009. Total annual new asylum applicants increased from almost 46,000 (2007) to over 207,000 (2008) and more than 220,000 (2009). Consequently, Zimbabweans represented 54% of all new asylum seekers in SA in 2008 and two-thirds of all new asylum applicants in 2009.
Zimbabweans represent a very small minority of refugees in SA. The total number of asylum seekers granted refugee status in SA has grown: 14,530 (1999), 36,700 (2007), some 37,000 (2008), almost 48,000 (2009). According to UNHCR, 710 Zimbabweans became refugees between 2000 and 1 April 2008. Another 200 Zimbabweans were granted refugee status in the refugee determination process in 2009. If one makes a generous assumption that another 200 Zimbabweans were granted asylum in the last nine months of 2008, then around 1,100 Zimbabweans would have been granted refugee status between 2000 and 2009.
It is difficult to obtain time series data on grant rates by nationality but Zimbabweans appear to have not fared well compared to the aggregate asylum grant rate in SA. The grant rate for all new asylum applications in 2006 that were decided in the first instance that year was about 16%; the corresponding figure for 2007 was around 29%. Grant rates hovered around 10% in 2008 and 2009. The acceptance rates for Zimbabweans in 2006, 2007, and 2009 were around 1.5%.
The outcomes of refugee determination interviews for Zimbabwean asylum applicants in 2009 were also much lower than for many other nationalities. Of the top ten countries from which asylum applicants originated in 2009, Zimbabwean grant rates in refugee status determinations clustered with applicants from Malawi, Bangladesh, India, China, Pakistan, Nigeria, and Mozambique, all of which had either zero or effectively zero percent acceptance rates. Outside of the top ten countries from which asylum applicants originated in 2009, grant rates in refugee status determinations were around 22% for Burundians, 30% for Ethiopians, Congolese, and DRC citizens, 65% for Somalis, and 74% for Eritreans.
Zimbabwean asylum seekers in the US
The number of Zimbabwean asylum applicants in the US is a very tiny percentage of all asylum applicants. Numbers are only available for applicants received in immigration courts and not for applications received in Asylum Offices. Between 2003 and 2010, immigration courts received over 400,000 asylum applications; only some 1,750 were from Zimbabweans.
Between 2000 and 2009 1,922 Zimbabwean individuals were granted asylum and at least 32 entered the US as refugees (see table 1). However, the data for Zimbabweans granted asylum affirmatively (through an interview with an asylum officer) includes the spouses and children of the applicant, whereas the SA data does not include dependents (spouses and children) although they are automatically granted asylum if the applicant is accepted. Applying aggregate data on the percentage of principal applicants to total applicants granted asylum affirmatively from 2002 to 2009 in the US, one can estimate the number of Zimbabwean principal applicants who obtained asylum, the more appropriate number with which to compare the number of Zimbabwean asylum grants in SA. About 70-74% of applicants granted asylum affirmatively in the US between 2002 and 2009 were principal applicants. Therefore one can estimate that between 2000 and 2009, the number of successful Zimbabwean principal applicants granted asylum affirmatively ranged between 993 and 1050. Adding in the 503 Zimbabwean principal applicants granted asylum through immigration court grants, a total of between 1496 and 1553 Zimbabwean principal applicants were granted asylum between 2000 and 2009. Even excluding the 32 Zimbabwean refugees admitted into the US, the number of Zimbabwean principal applicants granted asylum in the US over the past decade has exceeded the number of Zimbabwean applicants who were accepted as refugees in SA.
|Table 1||Zimbabwean individuals granted (G) and denied (D) asylum, 2000-2010|
The data to calculate the grant rate by nationality in asylum interviews is not available after 2004, so a comparison of total grant rates for Zimbabweans in the US and SA is not possible. However, aggregate immigration court asylum grant rates and immigration court asylum grant rates by nationality can be calculated. The grant rate for Zimbabweans – the number of grants as a percentage of the total number of grants and denials (see Table 1) – in US immigration courts was 43 % in 2003, 60% in 2004, 41% in 2005, 47% in 2006, 61% in 2007, 63% in 2008, 58% in 2009, and 52% in 2010 (see Table 2). Zimbabwean grant rates in the immigration courts have exceeded aggregate immigration court grant rates in each fiscal year from 2003 to 2010, and in four years (2004, 2006, 2008, 2009) quite significantly (see Table 2).
|Table 2||Immigration courts: Aggregate and Zimbabwean asylum grant rates, 2003-2010|
Those nationalities in SA with more favorable asylum grant rates than Zimbabweans, notably Burundians, Congolese from the DRC and the Republic of the Congo, Eritreans, Ethiopians and Somalis, also have almost uniformly higher grant rates in US immigration courts than Zimbabweans between 2006 and 2010. For example, in 2010 grant rates were 70% for Burundians (though the number of applicants was too small to be statistically meaningful), 35% for DRC citizens (one of the two rates for all these nationalities in the four year period that was lower than Zimbabwean grant rates), 70% for Republic of the Congo citizens, 86% for Eritreans, 79% for Ethiopians and 78% for Somalis. However, the aggregate grant rate for Zimbabweans was lower than for these other nationalities (except for DRC citizens) in 2010 but never by more than 34% – a much smaller differential than between the Zimbabwean asylum grant in SA between 2006 and 2009 and these other nationality grant rates in the same period.
Making sense of the numbers
To make sense of the higher aggregate grant rates for Zimbabwean asylum seekers in the US than in SA, the following factors are pertinent: adjudicators’ respect for the law, asylum seekers’ access to legal representation (albeit at their own expense), and lower level anti-immigrant attitudes. The expense of getting to the US means that Zimbabwean asylum applicants in the US will be wealthier and more educated, and therefore arguably better placed to win asylum than their less privileged counterparts who constitute the majority of Zimbabweans entering the SA asylum system. To account for Zimbabweans’ lower than average grant rates in SA, it is important to consider the influence of SA’s hostile foreign policy to Zimbabwe on asylum decisions, the proximity of Zimbabwe to SA and therefore Zimbabweans’ easy physical access to SA’s asylum system, the high incentives (a work permit and legal status) to enter the asylum system and low risks of asylum denials for Zimbabweans (deportation is not a high bar to re-entry).
Both countries’ asylum systems are deeply flawed in ways that critics complain have contributed to asylum seekers being denied due process. Yet the SA system, a much more recently established system, alone seems to routinely flout its own law. Asylum applicants often have to queue up overnight(s) to obtain access to a Refugee Reception Office where they must submit an asylum application within a prescribed period of time or face arrest, detention and deportation. To get access to the refugee reception officer, asylum applicants often have to pay bribes to security officials managing the lines. Once inside the office, refugee reception officers often violate their legal obligations to help asylum seekers fill out their eligibility forms and persist in two illegal practices: the issue of appointment slips to return at a later date to apply for asylum and the imposition of work and study restrictions on temporary asylum permits. Most disconcertingly, and central to an assessment of the decision-making process itself, a study of the written rejection letters for over 300 asylum applicants concluded that refugee determination officers almost always made decisions without any attempt and/or ability to assess the individual merits of applicants as prescribed in the law. In the US, the significant disparities in rates of favorable decisions in immigration courts are a source of great concern because it indicates that respondents’ chances of success are highly dependent upon the judges before whom they appear rather than the merits of the case. However, disparities seem to be associated with a judge’s gender, prior work experience, and length of time on the bench rather than with flagrant disregard for the law.
To make matters worse, asylum applicants in SA have no right to legal representation (until appeal) to mitigate endemic unlawful decisions. Studies on US immigration court decisions have found that whether or not asylum seekers are represented “is the single most important factor affecting the outcome of her case.” A US Government Accountability Office study found that between 1995 and 2007, legal representation generally doubled the likelihood of an immigration court granting asylum, after statistically controlling for a number of variables. Between 2006 and 2010, the percentage of represented noncitizens in immigration court removal proceedings ranged from 35 to 43 percent. The data are not available broken down according to legal representation for removal proceedings in which asylum is sought. However, assuming a similar rate of representation for asylum applicants, at least a significant minority have representation, thereby doubling their chances of acceptance.
Anti-immigrant attitudes appear to affect decisions in both asylum systems but pervade decision-making only in SA. In the US, anti-immigrant attitudes expressed by immigration judges are a concern. Critics of the US asylum process refer to individual immigration court judges sometimes abusing asylum applicants and behaving unprofessionally. In SA, anti-immigrant sentiment appears to permeate the system. Following the outbreak of countrywide xenophobic violence in May 2008, Amnesty International recommended an inquiry be held to examine the role of possible contributory factors, including “the longstanding pattern of discriminatory practices and attitudes shown towards asylum-seekers, refugees and migrants by officials, including from the Department of Home Affairs (DHA), the police services and also the magistracy.”
In the US, asylum seekers get the right to work only 150 days after they have submitted their applications, providing less incentive to use the asylum system to gain the right to work. When the US repealed the rules for asylum seekers in 1995 that had allowed them to work while their applications were being processed and introduced a system under which most applicants had to win asylum before they could be employed, the annual number of applications fell from about 140,000 to about 35,000. Most observers attributed the drop in the number of applications to the new rules. Moreover, almost all failed asylum seekers are given removal orders and cannot return to the US. These factors, by reducing the numbers of asylum seekers who abuse the system to stay in the country, may contribute to higher US grant rates than in SA. In SA, as soon as one enters the asylum system, one obtains a work permit – a factor that surely encourages foreigners who have no well-founded fear of persecution in their home countries to seek asylum and likely contributes to lower aggregate grant rates than in the US. The large number of Zimbabweans who apply and do not get asylum also suppresses the aggregate grant rates in SA.
If these factors cast light on aggregate asylum grant rates being higher in the US than SA, one must still account for the particular asylum outcomes for Zimbabweans: in SA, their grant rates have been lower than aggregate rates and lower than many other nationalities’ grant rates; in the US, their grant rates have been more favorable than aggregate grant rates. A compelling case can be made that in SA, Zimbabwean grant rates have been negatively affected by SA’s policy toward Zimbabwe. In the US, their acceptance rates are consistent with having been favorably influenced by the hostile US policy to President Mugabe and ZANU PF.  Though the data do not exist to establish causation, it is plausible to hypothesize that the stridently anti-Mugabe US policy makes asylum adjudicators more likely to accept the credibility of Zimbabwean asylum seekers’ claims than they might, absent the official demonization of Mugabe.
Since at least 2002, three consecutive SA presidents have demonstrated a pro-ZANU PF bias in the ongoing crisis in Zimbabwe. South Africa’s support for ZANU PF has included accepting key elements of President Mugabe and his party’s contentious narrative about Zimbabwe’s crisis, partisanship as a mediator and facilitator between ZANU PF and its key party rivals who have been in a coalition government together since February 2009, voting against efforts to condemn ZANU PF’s human rights record and obstructing attempts to place ZANU PF’s human rights violations on the agenda in multilateral institutions and to impose an international arms embargo, refusing to impose any form of sanctions against the Mugabe regime and continuing to supply it with electricity and oil – even when it did not pay its bills – and arms and ammunition.
Many analysts have recognized the impact of SA’s policy to Zimbabwe on asylum outcomes for Zimbabweans. Asylum bureaucrats initially denied asylum seekers from Zimbabwe entry to the asylum system because “there is no war in Zimbabwe” and therefore they had no right to apply for asylum. In June 2002 the University of Witwatersrand Law Clinic tested the legality of this practice in a case involving five Zimbabwean exiles. The case was settled when the DHA conceded that any Zimbabwean had a right to apply for asylum. Nonetheless, a 2006 report found that the perception that Zimbabweans were not legitimate asylum seekers persisted among asylum officials and the police. A 2009 research report remarked on the large discrepancy between the proportion of Zimbabwean asylum applicants in the survey at the refugee reception office phase and their much lower proportion at the refugee determination phase. The report suggested one possible explanation for this change is variation in treatment based on nationality, and recommended research “to investigate whether those migrant populations viewed with alarm by South Africans, or perceived (often incorrectly) to be economic rather than humanitarian migrants, confront greater obstacles in the application process.” In its report on the May 2008 xenophobic violence in SA, Amnesty International captured the negative attitudes of asylum bureaucrats to Zimbabweans when it proposed that the “implicit official denial that Zimbabweans in South Africa may be in need of international protection” may have been a factor contributing to the hostility toward Zimbabweans.
In SA, another factor contributing to low grant rates of Zimbabweans relative to those of some other nationalities has to do with its physical proximity to Zimbabwe. Most Zimbabweans enter SA by crossing the Limpopo River, a hazardous journey because of risks of seasonal flooding, as well as assault and rape, often by those paid to assist them to make the crossing. With very limited options to reside and work legally in SA, Zimbabweans have inundated the nascent asylum system because it at least enables them to stay in the country legally while their case awaits determination in a heavily backlogged caseload and permits them to work. Many Zimbabweans who seek asylum are economic migrants who do not have a “well-founded fear of persecution”. The risk for failed asylum seekers has been low in that deported Zimbabweans frequently re-enter the country without authorization. Since April 2009, there has been a moratorium on deporting Zimbabweans, which is due to end in July 2011. With low or non-existent risks associated with failed asylum, there is little or no disincentive to apply for asylum. The high incentives and negligible costs for applying for asylum mean that many unqualified Zimbabweans enter the asylum system, contributing to low grant rates.
Additionally, the SA asylum system would likely want to keep grant rates low for Zimbabweans for fear of encouraging even more Zimbabweans crossing the border as asylum seekers. In the US, the imperfect analogy is with the low grant rates for asylum seekers from Mexico. Asylum applications by Mexicans received by US immigration courts in 2010 were the second largest number by nationality – only Chinese had a higher number. In 2010, only 49 Mexicans were granted asylum and 508 were denied – a grant rate of less than 9%. The analogy is imperfect because Mexico has not been experiencing widespread state-sponsored human rights abuses as Zimbabwe has.
Systemic factors present in SA but not in the US help to account for the higher aggregate asylum grant rates (including for Zimbabweans) in the US than SA. In SA, asylum decision-makers fail to adhere to asylum law, and in particular, to even assess asylum applications on the basis of their individual merits and credibility; applicants do not have a right to legal representation in first instance asylum determination that might facilitate challenging unlawful decisions; and widespread xenophobia makes negative decisions more likely. Moreover, asylum seekers in the US who come from geographically distant countries, like Zimbabweans, are a self-selected group which arguably has better resources to maneuver in the asylum bureaucracy than most of their less privileged compatriots who seek asylum in SA.
Specific factors contribute to Zimbabweans in SA having lower asylum grant rates than aggregate grant rates. Zimbabweans in SA are discriminated against by asylum adjudicators who see them as coming from a country that does not produce refugees – a perception undoubtedly shaped by SA’s policy of support to President Mugabe and ZANU PF during its years of challenge by a popular but beleaguered opposition party. Moreover, physical proximity to SA has made Zimbabweans the largest asylum-seeking national group, even though probably most Zimbabwean applicants are economic migrants rather than individuals with well-founded fears of persecution. Deportation for failed asylum seekers does not operate as a major disincentive to enter the asylum system for Zimbabweans as, even absent the current moratorium, Zimbabweans can and have re-entered SA without authorization. Reluctance to encourage even more Zimbabweans to cross the border and seek asylum has likely also contributed to asylum adjudicators’ low grant rates for Zimbabweans.
This comparison of asylum grant rates for Zimbabweans in the US and SA demonstrates not only inter- and intra-country disparities but the many influences that impact on asylum outcomes. These influences include the design of the asylum system (the right to work and the right to legal representation at one’s own expense), the ability or will of asylum adjudicators to understand and apply asylum law, foreign policy (at least in SA), and asylum seekers’ calculus of the risk of deportation – itself a product of geography and effectiveness of border control.
 UNHCR, 2009: Global Trends. Refugees, Asylum-seekers, Returnees, Internally Displaced and Stateless Persons, 15 June 2010.
 Grant rates are the number of asylum grants divided by the sum of the number of asylum grants and denials.
 I focus on first instance determinations where the vast majority of asylum decisions are made in both systems as I do not have data for appeal decisions.
 Grant rates are calculated by dividing the number of asylum grants by the total number of asylum grants and denials.
 Asylum interview outcome records distinguish between “denials” and “referrals”. Applicants who have a legal status and are not granted asylum in an interview are recorded as having been “denied” asylum. Applicants who are not in a valid status and are not granted asylum in an interview are counted as having been “referred” to immigration court proceedings rather than as having been “denied” asylum.
 Report on 2009 Asylum Statistics, Government of SA, 2010. The SA government admits problems that “have a serious impact on the quality dimension of information printed in the report.” The report does not include data from the Musina office near the Zimbabwe border. Nonetheless, the report claims that it “provides an indication on the extent of new arrivals and refugee status determination during the year 2009.”
 In South Africa, the Refugees Act provides an alternative to determining refugee status on a case-by-case basis in the event of a mass influx of refugees. The Minister of Home Affairs may declare “any group or category of persons” to be refugees and impose conditions provided they are in conformity with the constitution and international law. In the US, the Secretary of Homeland Security may designate a foreign country for Temporary Protected Status (TPS) due to temporary conditions in the country, including ongoing armed conflicts or environmental disasters. http://www.uscis.gov/portal/site/uscis/menuitem.eb1d4c2a3e5b9ac89243c6a7543f6d1a/?vgnextoid=848f7f2ef0745210VgnVCM100000082ca60aRCRD&vgnextchannel=848f7f2ef0745210VgnVCM100000082ca60aRCRD (accessed 13 May 2011) After Operation Murambatsvina in 2005 in Zimbabwe, there was discussion of granting Zimbabweans TPS in the US (interview, Hugh Byrne, April 15 2010, consultant, US Office of Refugees, Asylees, and International Operations, Washington DC).
 United States Committee for Refugees and Immigrants, U.S. Committee for Refugees World Refugee Survey 2004 – South Africa, 25 May 2004, available at: http://www.unhcr.org/refworld/docid/40b459474.html (accessed 13 May 2011)
 For example, the rights of asylum seekers and refugees to work, accommodation, social assistance, health and educational services, permanent residence and citizenship; the rights of asylum seekers in detention; and the rights of their dependants (spouses and children) to legal status and health and educational services are not discussed.
 Daniel C. Martin, Refugees and Asylees: 2010, Office of Immigration Statistics: 2011, Department of Homeland Security. The year refers to the fiscal year, October 1 to September 30.
 Daniel C. Martin, Refugees and Asylees: 2010, Office of Immigration Statistics: 2011, Department of Homeland Security.
 Only 7% of asylum seekers applied while they still had a lawful immigration status in 2004. See Jaya Ramji-Nogales, Andrew I. Schoenholtz & Phillip G. Schrag, “Refugee Roulette, Disparities in Asylum Adjudication”, Stanford Law Review, November 2007, v.60, n.295, footnote 22, p.307.
 Jaya Ramji-Nogales, Andrew I. Schoenholtz & Phillip G. Schrag, “Refugee Roulette, Disparities in Asylum Adjudication”, Stanford Law Review, November 2007, v.60, n.295, footnote 22, p.308.
Daniel C. Martin, Refugees and Asylees: 2010, Office of Immigration Statistics: 2011, Department of Homeland Security.
 Jaya Ramji-Nogales, Andrew I. Schoenholtz & Phillip G. Schrag, “Refugee Roulette, Disparities in Asylum Adjudication”, Stanford Law Review, November 2007, v.60, n.295, footnote 22, p.309.
 The judge may deny the applicant asylum but award another type of relief, such as withholding of removal or protection under the Convention against Torture. These types of relief come with fewer benefits than asylum. For example, applicants granted asylum also secure asylum for their dependants (spouses and children).
 The US Attorney General appoints the members Board of Immigration Appeals; they serve at his pleasure and exercise his delegated authority rather than statutory authority.
 The one-year deadline was imposed by Congress in 1996, effective April 1 1998. If an asylum officer believes an applicant filed more than a year after entering the US, the asylum officer will refer it to an immigration court unless the applicant proves the existence of “changed circumstances” or “extraordinary circumstances”. US law bars from asylum include those who have committed serious crimes, pose a national security threat, engage in persecution of others, or firmly settle in another country before coming to the US. For the one-year deadline and other bars to asylum, see Asylum Officer Training http://www.uscis.gov/portal/site/uscis/menuitem.5af9bb95919f35e66f614176543f6d1a/?vgnextoid=2a1d1a877b4bc110VgnVCM1000004718190aRCRD&vgnextchannel=3a82ef4c766fd010VgnVCM1000000ecd190aRCRD
 Most asylum seekers do not cross the border legally or are ignorant of their right to request asylum at the border or, in the case of Zimbabweans, fear to do so as they believe intelligence agents are located at the border posts. Protecting Refugees, Asylum Seekers and Immigrants in South Africa, CoRMSA (Consortium for Refugees and Migrants in South Africa), Johannesburg, 2008, p.34. At present, individuals are being pre-screened at borders and turned back if they have not sought asylum in the first safe country they passed through, even though this practice is unlawful. Press release: LHR criticizes DHA’s use of unlawful measures to exclude refugees from SA. http://www.lhr.org.za/news/2011/press-release-lhr
 The Tshwane Interim Refugee Reception Office in Pretoria was opened on April 15 2009 to try to reduce the pressure on the other Gauteng province offices and initially served SADC nationals. The Musina office opened in July 2008. Since January 2010, the Marabastad office, also in Pretoria, services Zimbabweans and the Tshwane RRO services all other nationals. (See Protecting Refugees, Asylum Seekers and Immigrants in South Africa, CoRMSA, 19 June 2009, p.84; Roni Amit, Protection and Pragmatism: Addressing Administrative Failures in South Africa’s Refugee Status Determination Decisions, Forced Migration Studies Report, Forced Migration Studies Project, April, 2010, p.15 footnote 3.).
 UNHCR, 2009 Global Trends, p.17. The provisional global total of 922,500 individual asylum applications includes 836,100 initial applications, filed in first instance procedures, and 86,400 claims submitted on appeal or review. The SA asylum claims for 2009 represent first instance claims, according to Report on 2009 Asylum Statistics, Government of SA, 2010.
 UNHCR, 2009 Global Trends, p.17
 UNHCR, 2009 Global Trends, p.18
 UNHCR, 2009 Global Trends, p.18
 Report on 2009 Asylum Statistics, 2010, gives the total number of new asylum applicants as 12,226 (2000), 16,325 (2001), 24,187 (2002) and 41,369 (2004).
 IRIN, “South Africa: Rethinking asylum”, 15 August 2008, http://www.irinnews.org/PrintReport.aspx?ReportID=79850
 In 2006, there were almost 19,000 new Zimbabwean applicants out of the total new applicant pool of less than 54,000 (Protecting Refugees and Asylum Seekers in South Africa, CoRMSA, 2007, p.13); in 2007, there were 17,700 new Zimbabwean asylum applicants and a total of over 46,000 new asylum applicants (CoRMSA, 2008, p.17).
 Report on 2009 Asylum Statistics, Government of SA, 2010
 Report on 2009 Asylum Statistics, Government of SA, 2010
 Report on 2009 Asylum Statistics, Government of SA, 2010
 US Committee for Refugees and Immigrants (USCRI), World Refugee Survey (WRS) 2000: South Africa
 USCRI, WRS 2008: South Africa
 USCRI, WRS 2009: South Africa
 UNHCR, 2009 Global Trends
 IRIN, “South Africa: Rethinking asylum”, 15 August 2008.
 Report on 2009 Asylum Statistics, Government of SA, 2010
 Roughly 5,000 of the new applicants in 2006 were processed, and nearly 800 applicants were granted refugee status. CorRMSA, 2007, pp.4, 13.
 CoRMSA, 2008, p.17. Almost 6,000 of the new applicants in 2007 were processed, and 29% were granted refugee status.
 For the 2008 grant rate, see CoRMSA, 2009, p.30 Report on 2009 Asylum Statistics, Government of SA, 2010 states that 4,567 applications were approved and over 46,000 rejected in initial decisions in 2009; hence a grant rate of about 10%.
 The grant rate for Zimbabweans in 2006 and 2007 was about 1.5% (USCRI, World Refugee Survey 2009: South Africa). In 2009, 200 Zimbabwean asylum seekers were granted refugee status by refugee status determination officers, while 15,370 were rejected in the same process (Report on 2009 Asylum Statistics, Government of SA, 2010).
 The following grant rates for first instance decisions are calculated from data provided in Report on 2009 Asylum Statistics, Government of SA, 2010.
 Asylum cases received by immigration courts for 2003-2010 are available in FY 2007-2010 Statistical Yearbooks, US Department of Justice, Executive Office for Immigration Review (EOIR), appendix.
 Yearbook of Immigration Statistics: 2009. Table 14. Refugee arrivals by region and country of nationality: Fiscal Years 2000 to 2009. US Department of Homeland Security, USCIS, Refugee, Asylum, and Parole System (RAPS).
 Yearbook of Immigration Statistics: 2009, Table 18, Fiscal years 2002-2009 makes it possible to calculate the percentage of total individual affirmative grants that were for principal applicants. Comparable data for 2000-1 were not available online.
 Yearbook of Immigration Statistics: 2009, Table 17. Individuals granted asylum affirmatively by region and country of nationality: fiscal years 2000 to 2009. US Department of Homeland Security, USCIS, Refugee, Asylum, and Parole System (RAPS).
 Affirmative denial data by nationality are not publicly available but various efforts are underway to obtain this data.
 Yearbook of Immigration Statistics: 2009, Table 19.
 E = Data withheld to limit disclosure: the numbers are so small that it might be possible to identify the individuals.
 FY 2010 Statistical Yearbook, US Department of Justice, EOIR, January 2011. Note a minute difference in asylum grant numbers between Yearbook of Immigration Statistics: 2009 and 2011 EOIR data for numbers of Zimbabwean asylum grants in 2006 and 2008. According to the EOIR data, 56 Zimbabwean individuals won asylum grants in 2006 and 68 in 2006.
 FY 2010 Statistical Yearbook, US Department of Justice, EOIR, January 2011, appendix provides data on immigration court denials by nationality, 2006-2010. For 2003-5 defensive denials, see FY 2007, 2008, and 2009 Statistical Yearbooks.
 FY 2010 Statistical Yearbook, US Department of Justice, EOIR, January 2011, Figure 16, K1 for 2006-2010. For 2003-5, see FY 2007-9, Statistical Yearbooks, Figure 16. The Statistical Yearbook offers no explanation for the sharp increase in asylum grant rates in immigration courts between 2005 and 2006.
 FY Statistical Yearbooks, 2007-9, appendix for Zimbabwe data in 2003-5. For 2006-2010, see FY 2010 Statistical Yearbook, January 2011, appendix.
 These grant rates were calculated from the same US official sources as the Zimbabwe asylum grant rates.
 Both the US and SA asylum systems have been severely criticized on a number of counts, including: lack of resources (e.g. insufficient numbers of asylum adjudicators, lack of training of immigration court judges in the US and of status refugee determination officers in SA), huge backlogs, lack of qualified interpreters, absence of government-paid legal representation for asylum applicants, and arbitrary decisions. For criticisms of the US asylum system, see E.g. Assembly Line Injustice: Blueprint to Reform America’s Immigration Courts (Appleseed, May 2009); Reforming the Immigration System: Proposals to Promote Independence, Fairness, Efficiency, and Professionalism in the Adjudication of Removal Cases, American Bar Association, February 2010.
 Roni Amit with assistance from Tamlyn Monson, Darshan Vigneswaran and George Mukundi Wachira, National Survey of the refugee reception and status determination system in South Africa, FMSP, February 2009, p.25-34; see also CoRMSA, 2009, p.86. The law stipulates that asylum seekers are supposed to submit their application when they access the RRO; they should not have to make an appointment to apply. The appointment slips have no legal validity and expose asylum seekers to arrest, detention and deportation.
 Roni Amit, Protection and Pragmatism: Addressing Administrative Failure in South Africa’s Refugee Status Determination Decisions, FMSP Report, April 2010, see especially pp.7-10, 12-13, p.53; see also CoRMSA, 2009, p.86; Amit et al, FMSP, 2009.
 Reforming the Immigration System: Proposals to Promote Independence, Fairness, Efficiency, and Professionalism in the Adjudication of Removal Cases, American Bar Association, February 2010,
 Jaya Ramji-Nogales et al, “Disparities in Asylum Adjudication”, Stanford Law Review, November 2007, p.340, citing findings of studies.
 US GAO, Report to Congressional Requesters, US Asylum System. Significant Variation Existed in Asylum Outcomes across Immigration Courts and Judges, p.30, September 2008. GAO-08-940.
 FY 2010 Statistical Yearbook, US Department of Justice, EOIR, January 2011, G2.
 .e.g. Appleseed, pp.12, 14.
 AI, “South Africa Must Protect Those At Risk of “˜Xenophobic Attack'”, 23 May 2008. http://www.unhcr.org/refworld/docid/4843fd2f17.html [accessed 2 May 2011]
 David Ngaruri Kenney and Philip G. Schrag, Asylum Denied: A Refugee’s Struggles for Safety in America, University of California Press, 2008, p.2
 Lauren Ploch, Zimbabwe: The Transitional Government and Implications for U.S. Policy, Congressional Research Service, July 7 2010. http://www/fas.org/sgp/crs/row/RL34509.pdf (accessed May 3, 2011), pp.26-37. Beginning in 2001, the Bush administration introduced punitive measures against ranking ZANU PF officials and these remain in force as the Obama administration believes that the coalition government has not made sufficient progress toward economic and political reforms. For much of the past eleven years, the US has banned transfers of defense items and services (but not other bilateral trade), suspended non-humanitarian assistance to the government, and imposed targeted sanctions against the government. Targeted sanctions include the prohibition of direct assistance to the government, financial and visa sanctions against high ranking ZANU PF officials who undermine democratic institutions and processes in Zimbabwe and people who benefit financially from business dealing with such individuals as well as the spouses of people in either group. Under the Zimbabwe Democracy and Economic Recovery Act of 2001, US support for financial assistance to Zimbabwe by international financial institutions is prohibited until the President is able to certify that specific conditions relating to the rule of law, democratic elections, and land reform have been met. Current US policy is to support “reformers” in the transitional government to build on the provisions of the power-sharing agreement for democratic and economic reforms and to provide humanitarian assistance to help the poor and strengthen civil society. The US remains one of the largest donors of humanitarian assistance to Zimbabwe but almost all of it bypasses the government of Zimbabwe.
 David Ngaruri Kenney and Philip G. Schrag, Asylum Denied: A Refugee’s Struggles for Safety in America, University of California, p.2 states that allegations of foreign policy influences on asylum officers were made in the 1980s but that professionally trained asylum officers now make decisions based on law rather than foreign policy preferences. The suggestion that US foreign policy might influence asylum decisions for Zimbabweans is confined to adjudicators’ inevitably subjective assessment of credibility rather than application of the law itself.
 Norma Kriger, “The Politics of Legal Status for Zimbabweans in South Africa”, pp. 77-100 in JoAnn McGregor and Ranka Primorac (eds.), Zimbabwe’s New Diaspora: Displacement and the Cultural Politics of Survival, Berghahn Books, New York, 2010, pp.88-9; R.W. Johnson, South Africa’s Brave New World: the beloved Country since the End of Apartheid (Overlook, 2010) The contentious claims in ZANU PF’s narrative of the crisis include that Zimbabwe’s problems began when Mugabe decided to redress the historically injustice of sharply skewed racially-based land distribution and that punitive sanctions imposed by the West on Mugabe’s government and subsequently on the inclusive government were unfair, have had a devastating impact on the economy, and must be lifted.
 Lawyers for Human Rights (South Africa), The Documented Experiences of Refugees, Deportees and Asylum Seekers in South Africa: A Zimbabwean Case Study, April 2006
 Amit et al, National Survey of the refugee reception and status determination system in South Africa, FMSP, February 2009, p18, also pp.16-17.
 Amnesty International, “South Africa Must Protect Those At Risk of “˜Xenophobic Attack'”, 23 May 2008. http://www.unhcr.org/refworld/docid/4843fd2f17.html [accessed 2 May 2011]
 It is also the case that some Zimbabweans who would in theory qualify for asylum in SA choose not to apply because they want to continue to visit their families in Zimbabwe, which would be against the SA law.
 FY 2010 Statistical Yearbook, US Department of Justice, EOIR, January 2011, appendix – immigration courts, FY2010 asylum statistics. Mexico and Central American countries are consistently among the predominant nationalities of immigration court completions. For 2006 to 2010, Mexico, Guatemala, El Salvador, and Honduras were the top 4 nationalities (E1).