Citation
US refugee and asylum and Latin America

Material Information

Title:
US refugee and asylum and Latin America
Creator:
Garcia, Adrie ( author )
Language:
English
Physical Description:
1 electronic resource (60 pages) : ;

Subjects

Subjects / Keywords:
Political refugees ( lcsh )
Immigration enforcement ( lcsh )
Emigration and immigrants -- United States ( lcsh )
Emigration and immigration -- Latin America ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
The United States and Latin America have a very complex history that continues to this day. Many decisions and actions taken by the United States have had extensive implications throughout Latin America. This is particularly true of immigration issues. This study narrows this down and investigates the role of United States refugee and asylum policy and how this has played out in Latin America. It investigates several factors that may have an impact on refugee and asylee admissions from Latin America.
Thesis:
Thesis (M.A.)-University of Colorado at Denver.
Bibliography:
Includes bibliographic references
System Details:
System requirements: Adobe Reader.
Statement of Responsibility:
by Adrie Garcia.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
954759942 ( OCLC )
ocn954759942
Classification:
LD1193.L64 2016m G37 ( lcc )

Downloads

This item has the following downloads:


Full Text
US REFUGEE AND ASYLUM LAW AND LATIN AMERICA
by
ADRIE GARCIA
B.A. Metropolitan State University of Denver 2013
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
Of the requirements for the degree of
Masters of Arts
Political Science Program
2016


This thesis for the Master of Arts degree by
Adrie Anna Garcia
Has been approved for the
Political Science Program
By
Michael Berry, Chair
Lucy McGuffey
Ryan Crewe


Garcia, Adrie Anna, MA, Political Science
US Refugee and Asylum Law and Latin America
Thesis directed by Professor Michael Berry
ABSTRACT
The United States and Latin America have a very complex history that continues to this
day. Many decisions and actions taken by the United States have had extensive implications
throughout Latin America. This is particularly true of immigration issues. This study narrows
this down and investigates the role of United States refugee and asylum policy and how this has
played out in Latin America. It investigates several factors that may have an impact on refugee
and asylee admissions from Latin America.
The form and content of this abstract are approved. I recommend its publication.
Approved: Michael Berry
m


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION..........................................................1
A Brief History of United States Refugee Law......................4
II. LITERATURE REVIEW....................................................8
Heavy Burden of Proof and Subjective Judgement....................9
Domestic Politics and Foreign Policy.............................12
Deterrence Through Development...................................15
The Tug of War Between Norms and Interests.......................16
Why the Actors Matter: State or Non State........................21
Discussion and Critique..........................................24
III. METHODOLOGY AND EXPLANATORY FRAMEWORK..............................26
Human Rights and National Interests..............................27
Type of Threat...................................................28
Politics.........................................................29
Hypotheses.......................................................30
Data and Sample..................................................31
Measurement, Methodology, and Expected Results...................32
Research Design and Discussion...................................35
IV. RESULTS.............................................................38
Regression Analysis..............................................39
Accounting for Eras..............................................41


The End of the Cold War
42
NAFTA..........................................................44
September 11, 2001 ............................................45
Conclusion.....................................................48
BIBLIOGRAPHY.........................................................50
APPENDIX
A. List of States in Data Set........................................53
B. Sum of all data...................................................54
C. Correlation Matrix................................................55
v


CHAPTER I
INTRODUCTION
The relations between the United States and Latin America have always been
complex. In recent decades this complexity continues to grow. From the Monroe Doctrine
to the Good Neighbor Policy, from the communist tensions of the Cold War to the North
American Free Trade Agreement (NAFTA), United States foreign policy has had highly
consequential implications throughout Latin America. This region has endured many civil
wars and brutal dictatorships that have displaced countless people (Kay 2010). In 1954 The
United States protected its United Fruit Company interests in Guatemala which led to years
of civil war (Schoultz 1998). In 1960 the United States imposed an embargo on Cuba which
put Cuba at a devastating economic disadvantage that has lasted decades (Haas 1998). 1973
the CIA helped orchestrate a coup against the democratically elected president of Chile,
Salvador Allende, and the Unites States supported the brutal military regime of Augusto
Pinochet, which lasted until 1990 (Kombluh 2003) The United States invaded Panama in
1989 to remove Manuel Norriega from power, an operation that cost up to 1,000 Panamanian
casualties (Lindsay-Poland 2003) These are just a few examples of how United States
Foreign Policy has directly affected Latin American states.
Another example of foreign policy that has had a significant impact on Latin
America is the United States trade policies. Development and industrialization have
contributed to the displacement and migration of populations from Latin America (Anghotti
1


1996). Free trade has brought with it an era of, urbanization, maquiladoras, and race to the
bottom tactics by exploitative multinational corporations. These implications and countless
others are among the factors that have contributed to migration flows throughout the region.
According to Lees (1966) Push-Pull theory these conditions create push factors such as
lack of opportunity and a daily threat of violence that essentially spawn migration.
The other side of this coin is the pull factors of the United States such as better
economic conditions and an escape from political violence that attract migrants (Lee 1966).
While many immigrants pushed from Latin America are migrants fleeing poverty and
seeking greater opportunities, there are also many who are fleeing persecution and violence
in their home state.
These migrants are considered refugees and asylees. This determination is made
based on the following specifications: people who have been persecuted or fear they will be
persecuted on account of race, religion, nationality, and/or membership in a particular social
group or political opinion (US Citizenship and Immigration Services 2015). This is a
definition that has developed over time, first with the 1951 Convention on the Status of
Refugees which originally only addressed European refugees fleeing persecution during
World World II (UNHCR 2015). Later this definition evolved to remove the time and
geographic constraints with the 1967 Protocol Relating to the Status of Refugees (UNHCR
2015). Throughout this timeframe the United Nations High Council on Refugees (UNHCR)
has been a major driving force behind this development and has helped to address many
2


refugee crises around the world (UNHCR 2015). While the UNHCR deals with refugee
crises at the international level, individual states must also address these issues as they are
called upon to help. The United States has repeatedly been faced with refugee flows and has
responded in various ways. An example of this variation can be found in comparing Cuba
and Haiti. In the case of Cuba the United States has been very open to accepting refugees
and asylees whereas in Haiti the United States has not been so generous despite many periods
of violence and unrest (COHA 2010)..
The goal of this research project is to examine specifically how United States policy
has dealt with refugees and asylees from Latin America. Why have refugee and asylum
numbers varied from state to state over time? Have United States interests or relations in the
region affected the response of the United States to these populations, or has the United
States focused on a human rights based approach? In order to do so it is necessary to find
measures for such variables and collect and analyze the available data. This will help to
investigate the correlations that exist among these factors, if any correlations exists at all.
This study will show that there are correlations between these variables and that even more is
revealed when the data is broken out over specific time frames. Before addressing these
specific results it is first necessary to discuss how the literature has approached the issues in
question and to provide a brief overview the development of United States refugee law.
3


A Brief History of United States Refugee Law
As previously mentioned, the Second World War led the international community to
increased recognition of the need to protect individuals fleeing persecution from their home
states. This human rights concern was one of many stemming from the war, and the result
was the emergence of many international institutions to address these concerns. From
President Roosevelt presiding over the initial United Nations (UN) Charter to Eleanor
Roosevelts role in the development of the Universal Declaration of Human Rights (UDHR),
the United States has self-proclaimed a central role in the development of an international
human rights regime. As this regime began to grow, human rights became a touted United
States value. Consequently, a new body of law addressing asylum and refugee issues was
born in the United States with the passage of the Displaced Persons Act in 1948 (Public Law
82-60 65 stat 96). This legislation directly addressed European immigrants fleeing
persecution after World War II. A few years later, the UN brought forth the 1951 Refugee
Convention which defined the rights of refugees and specified who was considered a refugee
(UNHCR 2015). This was later broadened with the 1967 Protocol Relating to the Status of
Refugees as it removed the geographic and time restrictions that had originally been in place
to serve only those fleeing Europe after World War II (UNHCR 2015).
With this new global focus, and with the Cold War heating up, the United States
accepted hundreds of thousands of refugees from countries such as Cuba, China, Vietnam,
Poland, Korea, Hungary, among others. Decades later, as a response to these large influxes of
4


refugees, came the Migration and Refugee Assistance Act and the United States Refugee Act
of 1980 ( Public Law 87-510, 76 Stat. 121 ;Public Law, 96-212 94 Stat 102). While the
Migration and Refugee Assistance Act expanded programs and funding to accommodate
influxes of refugees, the United States Refugee Act took measures to limit refugee flows by
imposing annual limits. The stated purpose of this act is explicitly linked to the United States
human rights values. The precise wording states that the laws were to create a consistent
procedure for the admission to this country of refugees of special humanitarian concern to
the United States (Public Law 96-212 94 stat 102). The combination of United States
involvement in the development of this protection through international organizations and
through the passage of domestic legislation show the United States certainly subscribes to the
international norm regarding the protection of refugees. According to Finnemore and
Sikkinks (1998) model of the norm life cycle, it is evident that the United States considered
itself a critical player in both the emergence and the cascade, or proliferation, of the norm
to open doors to those fleeing persecution.
Over this time, the United States has developed a very robust refugee program,
resettling more than three million people since 1975 (Refugee Council 2014). However,
while the United States refugee program is extensive, it is also limited. One of the ways in
which the United States has restricted its flow of refugees is through refugee admission
ceilings as laid out in the Refugee Act of 1980. Each year the president limits refugee
admissions by region through the Refugee Admissions Ceiling (Martin and Yankay 2014).
5


This ceiling pertains to refugees, or individuals seeking refuge from outside the United
States. The ceiling excludes asylum seekers, who are individuals seeking this status from
within United States borders or at a point of entry into the country. There is no dictated limit
for asylum, however the numbers of asylees are generally significantly lower than the
refugee ceiling (Refugee Council 2014). Considering both the limitations within the United
States refugee program procedures and the human rights principles that are depicted within
the legislation, the question arises as to how these determinations are made. Why do refugee
and asylum admissions vary annually? Do they fluctuate in response to humanitarian crises
as they arise, or are there other factors at hand that may explain this variation?
With regards to United States refugee and asylum law, Latin America is an intriguing
region to examine as it consistently admits very few refugees when compared to other
regions (Refugee Council 2014). Latin America also represents a low percentage of those
granted asylum (Immigration Policy Center 2015). Considering the many humanitarian
crises throughout Latin America in recent decades, it is curious that these numbers have not
grown in response. Additionally this calls into question the nature of humanitarian concerns
the United States chooses to address. Many of the contemporary issues facing Latin America
have been involving drug cartels and gang violence. The UNHCR has recognized this
changing face of conflict and acknowledges what has come to be known as generalised
violence and the limits on a states ability to address such violence as factors to consider in
refugee cases (Turk 2011). In Latin America, the 1984 Cartagena Declaration on Refugees
6


took steps towards a more inclusive approach considering refugees in Latin America (Turk
2011). However, of the aforementioned steps international organizations have taken, the
United States remains only party to the 1967 Protocol and has not yet produced any major
domestic legislation regarding generalized violence. This begs the question as to whether US
refugee admissions shows a similar response to this type of humanitarian crisis as it does
others, such as civil wars or repressive dictatorships. Furthermore, considering the
complicated dynamics in the region, could it be that factors outside refugee law influence
admission outcomes in the US? Possibilities include domestic politics, the US governments
relations with the various Latin American states, multilateral trade agreements, among other
factors. Any of these elements could have a bearing on refugee and asylum outcomes, but do
they, and if so what carries the most weight? This inquiry requires an in depth survey of the
existing literature before proceeding.
7


CHAPTER II
LITERATURE REVIEW
The literature on US refugee and asylum policy has considerable breadth and presents
several perspectives, and from these perspectives certain recurrent themes arise. Among
these themes, several predominant arguments stand out. One of these arguments is that the
burden of proof that falls on the applicant for refugee or asylum is exorbitantly high,
especially when having to prove something that has been only vaguely stated within the legal
framework. Furthermore this language is so ambiguous that it leaves much room for
subjective interpretation by immigration court officials. In a seemingly arbitrary judicial
environment, this ambiguity impacts the outcomes of these cases (Aleinkoff 2003;Bonilla-
Mathe 2013; Zolberg et al 1992). Another argument is that refugee and asylum law is both
politically motivated and produces political outcomes (McBride 1999; Rexton-Kan 2011;
Charlton et al 1988; Suhkre 1994; Zolberg et al 1992; Zucker and Zucker 1996). This
perspective emphasizes the influence of both domestic politics and foreign policy in United
States response to refugee flows. Other scholars have argued that there is a struggle
between norms and interests in United States refugee and asylum policy (McBride 1999;
Charlton et al 1988; Rexton-Kan 2011; Rosenblum and Salehyan 2004). This point of view
takes into account how both United States interests and human rights norms associated with
refugees play into this process. Finally there is the argument that the United States refugee
program responds very differently to different types of threats. More specifically, when non-
8


state actors are the principal source of the violence, the response is negligible (Bauer 2004;
Bonilla-Mathe 2013; Estevez 2013; Longmire 2014). It is critical to note that none of these
perspectives are necessarily mutually exclusive, and that a closer look at each may more
comprehensively portray how United States refugee and asylum policies play out in reality.
Furthermore, to consider the case of Latin America, what does this literature tell us about the
consequences of such policies for refugee and asylum seekers from this region?
Heavy Burden of Proof and Subjective Judgement
When applying for refugee status or asylum in the United States, applicants must
demonstrate that they are unable to return to their state of origin due to the fact that they
were persecuted or fear persecution due to race, religion, nationality, political opinion, or
membership in a particular social group and not firmly resettled in another country (USCIS
2013).
While on the surface this definition seems rather concise, there are certain concepts
or circumstances within the language that could be considered open to interpretation.
Persecution, political opinion, and membership in a particular social group are the
most glaring examples of ambiguous concepts. How does one prove persecution or fear of
persecution to an immigration official? If the applicant has been targeted in a highly visible
conflict like a civil war, this may not be so difficult. However this may be much harder to
prove with the daily violence in many Latin American states perpetrated by cartels, corrupt
9


officials, and gangs. As previously mentioned, the Cartagena Declaration directly
addressed this issue by including persons who have fled their country because their lives,
safety or freedom have been threatened by generalized violence, foreign aggression, internal
conflicts, massive violation of human rights or other circumstances which have seriously
disturbed public order (UNHCR 1984, 3). This declaration, however, is not legally binding.
Nor was the United States a participant in its creation. This leaves applicants in the United
States without any type of direct protection on the grounds of generalized violence. Because
of this, the burden of proof in these cases is much more out of reach although the risk in
their daily lives is still very prevalent. As Aleinkoff (2003) points out this process gets so
bogged down in identifying the cause of the persecution that the actual goal of determining
whether or not they face the risk of violence is undermined. In other words, refugee seekers
may have difficulty identifying their persecutors or making a case for the root cause of the
persecution while the primary issue, whether or not they face violence as a result of this
persecution, is neglected.
Assuming the applicants can prove persecution, they must then prove that they fall
into one of the designated classes of race, religion, nationality, political opinion, or
membership in a particular social group (United States Citizenship and Immigration
Services 2013). This is yet another obstacle for applicants, especially if they fall into one of
the latter categories as these require proof of a characteristic (Aleinkoff 2003). Political
10


opinion and membership in a particular social group are the claims made by Latin American
refugees, but these claims are often grounded in characteristics that are very difficult to
prove such as the refusal to join a gang or the inability to carry out ones occupation due to
extortion (Bonilla-Mathe 2013). These examples demonstrate how blurry these definitions
can become and how the focus can fall away from the question of whether or not the
applicants face a risk of violence and towards whether statutory regulations have been met.
While the UN expressed the legitimacy behind both of these circumstances by
offering them as examples in a handbook on refugees, the broad language of domestic
refugee law causes the fate of such claims to lie in the interpretation of immigration officials
(Bonilla-Mathe 2013). Consequently, as can be seen by the decisions of the Board of
Immigration Appeals, these officials tend to favor very narrow interpretations of such
definitions (Aleinkoff 2003). The applicant may have been able to present evidence of their
circumstances to an immigration judge that subjectively decides otherwise. These
subjective decisions are facilitated by the laws ambiguity, making consistency in refugee
and asylum cases a challenge. It is also relevant to note that these broad definitions are not
merely prerequisites in a legal process, they also carry moral, political, and legal
implications (Zolberg et al 1992). The applicants may be in a struggle with the language to
prove their individual case, while in reality the political environment could highly
11


exacerbate these difficulties. Immigration flows invoke political responses. Refugee issues
as a source of immigration become politicized in the process.
Domestic Politics and Foreign Policy
The increase of refugees in recent decades has created the widespread belief that
being overly sympathetic to the plight of refugees will spawn a refugee crisis and they
will come pouring in in masses (Zolberg et al 1992). It is this type of sentiment that has
domestic political salience. This is particularly acute in the case of Latin America, being
situated so near to the United States. The logic behind the fear in this case is that if the
United States starts validating the case that the generalized violence in Latin America in
sufficient to begin accepting refugees, then it will open the floodgates for mass immigration
across the Mexican border or through the Caribbean (Rexton-Kan 2011). This fear and the
political significance behind it were clearly illustrated during the unaccompanied minor
refugee waves that entered the United States during 2013 and 2014. There were intense
political debates and protests on both sides of the issue. This debate is one that is typical of
immigration issues in domestic politics, with one side urging the humanitarian acceptance
and the other pushing for greater restrictions (Zucker and Zucker 1992). The level of
political influence goes far beyond this domestic debate however. Global politics factor into
this entire process, from the root of the conflicts to the international response, to the
direction of the refugee flows (Suhrke 1994).
12


The international political component can be easily seen from the time of the Cold
War, as the United States and the Soviet Union were highly involved in the competition for
allies in the developing world at a time of widespread conflict including many liberation
movements and civil wars (Suhrke 1994; Forsythe 1988). As the United States and the
Soviet Union helped to arm both sides of these conflicts, the resulting violence created
refugee outflows. The same Cold War politics that created such outward flows likewise
dictated the direction of the movements of refugees (Suhrke 1994). Examples of Cold War
politics dictating refugee flows would be the mass relocation of Vietnamese to the United
States and the political refugee resettlement extended to Cubans, both states that were Cold
War hot spots. The case of Cuba is particularly critical in this discussion. The United States
absorbed massive amounts of political asylees from Cuba during the Cold War without the
major domestic political pushback that was previously discussed. This speaks to both the
salience of the international politics of the Cold War and its influence on domestic politics.
Additionally there was no major catastrophe as a result of such a massive refugee flow
which serves as an example of how currently rhetoric regarding the fear of opening the
floodgates is likely unfounded.
As the Cold War came to an end, however, there was little strategic incentive for the
United States to remain so highly active in these conflicts, and so began the withdrawal of
aid (Suhrke 1994). This left states scrambling quickly for solutions, and in many places was
13


coupled with the de-stabilizing aftermath of colonialism and a strain on the developing
worlds resources. This post-Cold War geopolitical environment proved quite different for
refugees, where superpowers are much less likely to get involved directly in these conflicts
and are no longer likely to launch major refugee programs (Suhrke 1994). Haiti is a prime
example of this shift.
In 1991 and 1992 approximately 44,000 Haitians fled after the overthrow of
President Artiside. The first response was a court order halting the processing of these cases
in immigration court followed by the rerouting of 34,000 of these individuals to
Guantanamo Bay. In the end, only 10,000 were approved for asylum status. The United
States simply had no foreign policy interest invested in Haiti. When compared to the mass
approval of political asylum seekers fleeing Cuba, this case shows the extent to which the
international political stage influences refugee and asylum outcomes (McBride 1999). A
deeper look at the way the United States responded to Haiti shows that in place of accepting
mass quantities of refugees, it is a strategy of deterrence that has taken hold. Rather than
admitting refugees, the United States shifted towards investing in and encouraging
development within these countries. The idea here is to deter refugee flows by investing in
creating better economic conditions in the country of origin (Suhrke 1994; Xenos 1993).
14


Deterrence through Development
The deterrence through development concept implies that in a modem environment
focused on free trade and global markets the United States will more readily take actions to
avert the flow of refugees than admit refugees (Zolberg et al 1992). In the case of Haiti, the
United States refused to denounce support of the oppressive Duvalier regime but created an
export processing zone in hope of sparking development and thereby avert the flight of
Haitians while at the same time admitting few Haitian refugees (Suhrke 1994; Xenos 1993).
This process ultimately failed as Haiti spun into further violence and refugee seekers
increased. Despite the failure of such actions, it appears as though the United States has not
abandoned this approach, especially with regards to Latin America.
Recent United States-Latin American relations have shown that the United States
continues to engage in many development projects while at the same time remaining very
conservative in its refugee program there. A major justification for this type of action is that
projects like these will raise all boats and both the United States and the developing states
economies and conditions will improve. This has yet to be seen in Latin America. While
some states like Brazil and Chile have undergone intense development and have improved
their macro-economies, there remain many marginalized populations that have been
displaced or neglected as this process has occurred. An example specific to Brazil would be
the tens of thousands evicted from slums of Rio just before it hosted the World Cup
15


(Waldron 2014). In the case of Chile the displaced Mapuche were forcibly removed from
their native land to accommodate a water plant (Barbera-Hemandez 2005).
Examples such as these can be found throughout Latin America where development
initiatives are pushed forward despite human rights concerns. Regardless of any question of
intention or motivation behind these actions, these cases indicate that development has not
been the most effective vehicle for limiting refugee flows through societal advancement.
Perhaps this is simply an example of how the United States has prioritized its economic
interests over its self proclaimed identity as a human rights leader. These are questions some
scholars have addressed through the lens of norms versus interests.
The Tug of War Between Norms and Interests
The United States Refugee Act of 1980 explicitly states that the refugee admissions
ceilings, shall be such number as the President determines, before the beginning of the
fiscal year and after appropriate consultation, is justified by humanitarian concerns or is
otherwise in the national interest (Public Law 96-212 94 stat 102). These two options for
justification set the stage for a struggle between refugee admissions based upon human
rights norms and United States interests. Rexton-Kan (2011) emphasizes this struggles in
the case of Mexico and to whom he refers as the narco-refugees. On the one hand, if the
United States qualifies the cases of those fleeing drug violence in Mexico as refugees, the
proximity and volume of the problem create a perceived national security threat of a major
16


refugee flow at a time of contentious political debate over illegal immigration (Rexton-Kan
2011). On the other hand, the United States dismissing these claims as illegitimate and
sending the asylum seekers back to the violent circumstances strikes at the heart of
American values of justice and humanitarianism (Rexton-Kan 2011, vi). This example
highlights how prioritizing human rights would mean the United States should accept narco-
refugees, but that doing this may run counter to national interests. If the United States takes
a human rights approach, this introduces the risk of invoking a refugee crisis. Additionally,
close trade relations with Mexico puts the United States in an awkward position, for to begin
to accept high volumes of Mexican refugees makes an international statement that the
Mexican government is unable to handle the situation. In this struggle between norms and
interests, which has more pull in refugee and asylum law? This case is similar to other
Central American states that struggle with cartel and gang violence such as Nicaragua, El
Salvador and Guatemala, to name a few. The literature reflects that the complexity of the
relations between states and the ways refugee policy can reflect these relations make this
comparison difficult to determine.
While there is evidence that both norms and interests have an impact on refugee and
asylum outcomes, some scholars have argued that in the United States the national interest
seems to carry more weight (Rosenblum and Salehyan 2004). Furthermore over time as
interests have shifted, so has its refugee policy. During the Cold War the United States
17


favored states with whom it had military ties, and therefore accepted fewer refugees from
those states so as not to indicate that the regimes in question were committing human rights
violations or were unable to address domestic concerns. As interests shifted towards
economics, the United States began to favor states that were major trading partners, and
therefore accepted fewer refugees from those states to maintain strong economic ties
(Rosenblum and Salhyan 2004). This concept implies that the reputation of a state matters
when it comes to refugee policy and interstate relations. In other words, relations between
the United States and a trading partner would become strained if the United States were to
accept large amounts of refugees from the partner state as this would imply that the state in
question cannot handle its internal conflicts.
Another point of strategic interest to the United States refers back to the salience of
illegal immigration issues in domestic politics, a factor referred to as migration control.
This implies that the growth of illegal immigration will create a negative impact on refugee
and asylum applications from the state of origin of the undocumented immigration
(Rosenblum and Salhyan 2004). This restrictionist approach is supported by the Presidents
ability to set the refugee ceilings, and is only furthered by negligible rates of asylum
acceptance (Charlton et al 1988). Given the way refugee and asylum operate, the
presidents ceiling would only be effective if asylum numbers reflected these limits. Quite
simply, given a huge illegal immigrant population, the maximization of both executive
18


discretion and ultimate political control is secured only by reluctantly granting any
regularizations of immigration status (Charlton et al 1988, 257).
Another example is the recent influx of immigrants, most notably the
unaccompanied minors, fleeing violence in Central America. The asylum approval rates
from these countries over this time has been very low, and ... only about 4 percent of
asylum claims from Guatemala, Honduras and El Salvador are successful, a jarringly low
number for countries with some of the highest murder rates in the world (Cabrera 2014, 1).
The proximity to the United States paired with the miniscule asylum approval rate shows
that the national interest may play a role in the response to these asylum seekers.
While a shift can clearly be seen in policy response due to the national interests, a
similar shift can be seen from human rights factors. As the Cold War came to an end the
trends began to shift for refugee determinations, favoring democratic values and protecting
individuals against states with higher levels of violence; this shows a possible commitment
to human rights (Rosenblum and Salhyan 2004). The growth and evolution of both these
factors make this contest difficult to analyze as the world rapidly changed during this time
and both factors showed growth. What is essential to note about the interest factors is that
they function with little to no regard towards the human rights conditions in the states of
origin. In other words, national issues such as trade or illegal immigration trump the real
19


risks these individuals face, which is in opposition to both the purpose of refugee law and of
the self-proclaimed values of the United States.
This is a critical point with regards to many Latin American states in todays context
as there is a high level of violent conflict while at the same time a high percentage of the
undocumented immigrants in the United States. Furthermore, many of these states having
close trade relations with the United States. While it is also true that norms and interests are
not necessarily mutually exclusive and these factors may align at times, the literature points
towards national interest influencing United States policy outcomes more than human rights
norms (Rosenblum and Salhyan 2004). How would the United States justify such a
preference for interests over norms when valuing human rights is so central to its identity?
One answer may lie in the way the United States defines conflict. If refugee
admissions have implications for interstate relations, the conflicts within these states would
be defined in a way that reflects these relations. In the case of an amicable relationship, it
would be beneficial if the United States were to define the conflict in a way that absolves the
state from culpability in a conflict and, in doing so, prevent the United States from being
compelled to protect refugee seekers from this state. One way to do this is to claim that the
violence is generalized as opposed to targeted, and an easy way to make this claim is when
the perpetrators of the violence cannot easily be identified by a flag, tank, or uniform.
20


Why the Actors Matter: State or Non-State
While the wording of United States refugee law makes no explicit reference to the
actors committing the violence, all of the conflicts refugees flee from fall into one of the
following categories: a conflict where the violence is state sponsored, a conflict that is
perpetrated by non-state actors, or a conflict in which this distinction is unclear. Does the
United States respond differently to these scenarios? Some scholars say that it does,
especially in the cases of Central American gangs and drug cartels (Bauer 2004; Estevez
2013; Bonilla-Mathe 2013; Longmire 2014). While the burden of proof requires the
applicants to prove their own persecution based upon political opinion, the state then negates
many of these arguments due to the fact that the opinions or characteristics brought forth are
difficult to attribute to their political significance, especially when a non-state actor is
involved. Another challenge applicants face when non-state actors are involved is the
obstacle of proving that they were the target of such violence. Young people in specific
areas may be the targets of gang recruitment or business owners may be targeted to
cooperate with cartels, and the refusal of such actions certainly could produce a reasonable
fear of persecution. This form of targeted persecution from a gang or cartel is much harder
to verify or prove than when a mobilized military is targeting a specific religion (Bonilla-
Mather 2013).
21


Despite the fact that the UN specifically addresses this caveat by stating that
violence may in fact come from actors that do not heed state law, the United States generally
focuses on state sponsored violence when considering refugee claims (Bonilla-Mathe 2013).
Bonilla-Mathe (2013) cited several cases in which the courts inconsistently defined
membership in a social group to deny refugee status to individuals fleeing gang violence
in direct conflict to UNHCR guidelines while widely accepting claims based on violence
perpetrated by states actors. The UNHCR has directly addressed this very issue of
organized gang violence in a report titled Guidance Note on Refugee Claims Relating to
Victims of Organized Gangs, and and within this report has laid out a legal analysis to help
navigate such cases (UNHCR 2010) The United States inconsistency in validating these
claims shows a lack of recognition of UNHCR guidelines. Bauer stresses the implications
of such impractical standards, The law fails to recognize, however, that the actions or
inactions of a civilian caught in the midst of generalized political violence can be construed
as having a political significance for which there are very real-and sometimes deadly-
consequences (2004, 1116).
A major problem with this type of oversight goes back to definitions. The way in
which the United States views the notion of ongoing armed conflict is antiquated and
refers almost exclusively to state actors while ignoring criminal organizations (Longmire
2014, 1). This is precisely why the UNHCR has moved towards a position that specifically
22


addresses generalized violence. Equally important is the fact that the United States often
aids these governments at the federal level to combat such organizations while corruption at
the municipal level often leads to local government consorting with the criminal
organizations (Longmire 2014). In the case of Mexico, their federal government is
committed to working with the United States through the Merida Initiative, which is a
bilateral agreement between the United States and Mexico to cooperate when fighting drug
trafficking and organized crime. This situation is difficult to assess considering that many
local police and governments have ties to the drug cartels, hence making it difficult to
establish if Mexico is unable or unwilling to protect the refugee seekers (Estevez 2013).
This implies the United States may deny that certain violence is political or overestimate the
actions of the federal government, meanwhile ignoring the lower level government actors
involvement in light of efforts to stymie the conflict. In doing so, this allows the United
States to avert refugee flows and maintain friendly relations with the state.
It is also important that the above examples illustrate that the distinction between
state and non-state actors is not always clear. In some cases, state actors may be acting in
cooperation with non-state actors. Another possibility is that state and non-state actors
could be engaged in conflict with one another, and both could be committing acts of targeted
acts of violence towards certain communities. Both of these examples highlight the
difficulties of analyzing whether persecution comes from state and/or non-state actors.
23


However, just as in the norms and interests debate, a closer look could show whether policy
outcomes lean towards response to state or non-state persecutors.
Discussion and Critique
This survey of the literature shows consensus leaning toward the notion that the
United States focuses its refugee and asylum on politics and national interest over human
rights considerations. The desire to hold back the possibility of any major refugee flow
while creating a tension with Latin American governments appears to take precedence over
human rights concerns of violent conflict in these states. The ambiguous nature of the
wording in refugee and asylum law presents a monumental challenge for many individuals
fleeing gang and drug war related violence. This language also allows for judicial
interpretation of many refugee claims, which in the case of Latin America often end up
being narrow interpretations of the language and consequently invalidate the seekers
claims.
Another contemporary trend the literature has yet to fully explore is the move towards
securitization, especially within the United States. If immigration is perceived as a threat to
national security, then the notion of risk gets turned on its head. From this perception
refugees pose a risk to the host state rather than focusing on the vulnerability and risk the
refugees take on (Ibrahim 2005). This is a particularly critical post in the post 9-11
environment where immigration discourse more often than not includes rhetoric of border
security. Ideologically speaking, terrorism is the new evil of the day, replacing
24


anticommunist sentiments of decades past. The War on Terror has replaced the Cold War as
a major component of the United States position in geopolitics. Again, definitions here are
of paramount importance. For example, these criminal organizations are not framed as
terrorists in either the domestic news media or in domestic political discourse. Considering
the widespread violence Mexico and Central America has experienced in recent years, the
United States news coverage has been negligible. As Rexton-Kan (2011) discusses with
regards to Mexico, the United States should be paying attention to the nature of the violence
at its borders. Not only is the violence widespread and enduring, it is also gruesome. It is so
gruesome, in fact, that they have developed an entire vocabulary to categorize the ways in
which cartels dismember the bodies of their victims (Rexton-Kan 2011).
This begs the question whether emphasizing the terrorist nature of these crimes would
change the way these issues are viewed. Cartels and gangs being framed as terrorists could
potentially create more support for the United States responsibility to protect refugees
fleeing this type of violence. The opposite could also occur, and this redefinition could swing
support towards more intense securitization with a fear of the violence crossing over.
Considering the literature, would this even make a difference as they are often non-state
actors? The many questions left unanswered with regards to United States refugee and
asylum policy and the ongoing violence in many Latin American states presents a ripe
opportunity for research especially in regards to why refugee and asylee admissions vary
annually? What factors have an impact on this variation?
25


CHAPTER III
METHODOLOGY AND EXPLANATORY FRAMEWORK
In order to embark on any research on this topic, it is necessary to operationalize
some of these complex factors. Effective conceptualization of the many variables in
question is instrumental in determining possible correlations within refugee and asylum law
and its response to Latin Americans fleeing violence. For the purposes of this proposed
study, the dependent variables will be refugee and asylee admissions (RAA) (y). In order to
explore the variance in outcomes throughout Latin America, the unit of analysis will be
state/year. The measurement of this dependent variable is straightforward as the number of
annual admits is available through the annually published Yearbook of Immigration
Statistics (Department of Homeland Security). Refugee and asylum admissions are a ratio
variable as the value of zero means there would be no refugees or asylees allowed from a
region. These values would fall between zero and the maximum admissions for that year.
Because there are specific states that are notably different from the rest additional
models will be presented to exclude these outlying cases in order to account for their effect
on the data. The most obvious example of this is Cuba since there was a program designed
specifically for Cuban refugees as discussed earlier. Because these numbers are likely to be
much larger than for other states, it is necessary to parse this out for an accurate analysis of
the region as a whole and of Cubas impact on the data.
This study will be conducted with bivariate and multivariate regression analyses
using independent variables grouped according to some of the main theoretical debates
presented in the existing literature as follows: human rights concerns vs national interest,
26


types of threat (state or non-state actors), and political factors. The measures for each of
these are described in more detail in the following sections.
Human Rights and National Interests
Interests will be measured by United States relations with each state using the following
variables:
Interests-United States trade relations which is hypothesized to have a
negative effect on RAA
The more closely we are economically tied with a home state, the less likely the
United States is to accept refugees from that state because this would imply that the
government in incapable of effectively handling the issue.
Interests-United States military aid which is hypothesized to have a negative
effect on RAA
The more the United States invests in the military operations of the home state, the
less likely the United States is to accept refugees from that state because again that would
imply the government the United States is attempting to assist is incapable of handling the
issue.
Human rights concerns will be measured by indicators that show where states fall
concerning humanitarian issues (also including the question of states that share democratic
ideals as discussed in the literature review) using the following variables:
How high the state scores on the Freedom House index is hypothesized to
have a positive effect on RAA
The freedom index is on a scale from 1 to 7 with 7 being the least free, therefore the
lower the score on the freedom index, the United States would be less likely to accept
27


refugees because these would be states that have more political rights and civil liberties.
The higher the score on the freedom index would indicate that these states have less political
rights and civil liberties and it is hypothesized that the United States would accept more
refugees and asylees from these states.
How high the state scores on the Political Terror Scale is hypothesized to
have a positive impact on RAA
The higher the score on the Political Terror Scale would implicate poor human rights
conditions and based on these hypotheses would therefore imply a higher level of refugee
and asylees accepted.
The higher the number of incidents on the Global Terrorism Database is
hypothesized to have a positive impact on RAA
The more incidents on the Global Terrorism Database would likewise implicate poor human
rights conditions, and would include those not accounted for on the Political Terror Scale,
and it is hypothesized that this would lend to a higher level of refugee and asylees accepted.
Type of Threat
The type of threat will be measured by a comparison of the following variables:
The higher the rate of incidents that include state sponsored violence is
hypothesized to have a positive impact on RAA, as published by the Political
Terror Scale.
The PTS accounts only for violence committed by state actors and it is hypothesized
that this type of violence would be more likely to be considered and accepted in refugee and
asylum cases.
28


The higher the rate of incidents that include non-state actors is hypothesized
to have a negative impact on RAA as published by the Global Terrorism
Database
The GTD accounts only for terrorist activity that is not committed by governments
and it is hypothesized that this type of violence would be less likely to be considered and
approved in refugee and asylum cases.
Politics
United States domestic politics impact on Refugee and Asylum outcomes will be
measured using the following variables:
If the state of origin is a major source of illegal immigration, this is
hypothesized to have a negative impact on RAA
As the number of undocumented immigrants from a state increases, this will imply a
lower number of refugees and asylees from that state as a response to domestic political
concerns of mass migration or of a refugee crisis.
The more the state of origin is mentioned in medias political immigration
discourse this is hypothesized to have a negative impact on RAA
As the media gives more coverage to immigration from specific states, this would
imply a lower number of refugees and asylees from that state in response to domestic
political concerns.
29


Hypotheses
Considering the above variables that will be explored and the desired focus for the
analysis of this study as Latin America, the most appropriate central hypotheses are as
follows:
Hypothesis One As the violence involving non state actors increases, the RAA in those regions will decrease.
Hypothesis Two The more the state of origin is mentioned in political immigration discourse this is hypothesized to have a negative impact on RAA
National
Interests
Explanatory
Framework
(---------------\
Unit of Analysis:
State/Year
V_______________J
Human Rights
Conditions
non-state actors
Figure 1 Explanatory Framework
30


While the results pertaining to these hypotheses will certainly produce useful results,
the results of the entire study will surely be more constructive in creating a more complete
understanding of how these policies play out in reality. These principal hypotheses have
been chosen to address both the nature of the violence many in Latin Americans are fleeing,
and whether or not there is any correlation between domestic politics and the United States
response to these circumstances. The independent variable of conflicts involving non-state
actors has been chosen to draw the distinction between cartel violence and other organized
crime from state sponsored violence. The second independent variable, being a recurrent
point of discussion in United States immigration politics, has been chosen to evaluate the
relationship between domestic immigration politics and the outcomes of refugee and asylum
cases.
Data and Sample
The data for the dependent variable, refugee and asylum admissions are available
through the Department of Homeland Securitys Yearbook of Immigration Statistics, with
earlier editions appearing in their archives (US Department of Homeland Security 2015).
Within this document, there is a table with a breakdown of immigrants admitted by class of
admission and nationality. Refugees and Asylees combined are one of these classes. The
entire population in this case would be every state and year since 1980, when the refugee
admission ceilings program began. The sample chosen for this study will be each state
within Latin America from 1980 forward. This sample has been chosen to avoid the pre-
1980, pre-ceiling refugee numbers from throwing off the data. This does limit the study as it
excludes many conflicts in Latin America during the 1970s as well as major waves of Cuban
migration after the Cuban Revolution. This sample is, however, the most appropriate sample
31


for this study because it covers every year since the implementation of Refugee Admissions
Ceilings. Data collection will occur by plotting out each case (state/year) then recording the
refugee and asylee admits from each state/year. Data collection pertaining to the independent
variables will also be discussed after a clarification of the measurement of these variables.
Measurement, Methodology, and Expected Results
The measurement of the dependent variable in this study is straightforward. The
concept of United States refugee and asylum policy outcomes are operationalized through
refugee and asylum admissions. Refugee and asylum admissions are a ratio variable as the
value of zero means there would be no refugees or asylees allowed from a region. These
values would fall between zero and the maximum admissions for that year. This measure is
valid as these are the only two routes a refugee or asylee can take to obtain such status within
the United States, therefore it covers those who enter to go through this legal process
annually.
Despite the strengths of the study, there is a possible threat to the measures validity
which may require further attention. The primary issue here concerns the distinction
between affirmative asylum and defensive asylum. Affirmative asylum is the case applicants
bring to court who are not facing deportation while defensive asylum pertains to those who
are also facing deportation proceedings. As this measure only considers those admitted
under this class of admission, it omits the cases in which someone has already entered the
United States illegally and afterwards applies for defensive asylum. Without including these
cases, this could potentially threaten the validity of the study, especially considering the
highly politicized nature of deportation of undocumented Latinos in contemporary domestic
politics. This is a particularly relevant concern with regards to the second hypothesis which
32


relates directly to United States immigration politics. Within this sample however, there is
no uniformity across the timeframe with the data available to parse out defensive and
affirmative asylum. This is why the chosen variable best fits the study despite this obstacle.
The first hypothesis states that as the conflicts involving non-state actors increase, the
refugee and asylum admissions will decrease. The concept of conflicts involving non-state
actors is captured using the Global Terrorism Database (National Consortium for the Study
of Terrorism and Responses to Terrorism 2015). The Global Terrorism Database provides a
comprehensive data set for violence perpetrated by non-state actors for each state within a
region and is available throughout the entire sample. This fits the non-state actor
requirement as this data set applies the following definition of terrorism: "the threatened or
actual use of illegal force and violence by a non-state actor to attain a political, economic,
religious, or social goal through fear, coercion, or intimidation (National Consortium for the
Study of Terrorism and Responses to Terrorism 2015).
This is an appropriate mechanism in this context because this data set accounts for
drug cartel and other organized criminal activity, both of which are essential in any analysis
of contemporary Latin America. The data for this variable will consist of non-state actor
terrorist events in each state for each year. This would also be a ratio variable with zero
indicating zero terrorist incidents and increasing as the incidents increase.
The testing of this first hypothesis could also be taken further by juxtaposing the GTD
with the Political Terror Scale (PTS). Data from these sources are also available throughout
the entire timeframe of this study and are provided on a state-by-state basis. The PTS
focuses primarily on state sponsored violence, which would provide as a useful comparison
to those who endure non-state violence. The PTS is an index which ranks countries ordinally
33


from 1 to 5, with 5 being widespread and daily state sponsored violence (Gibney et al 2012).
This is an ordinal variable and would provide where each state falls on the scale. It is also
important to keep in mind that state perpetrated violence and violence by non-state actors
often happen in tandem and that these two variables are not mutually exclusive.
The second hypothesis tests the presence of a state as a part of United States political
discourse on immigration causing the refugee admissions ceilings and percent of asylum
applications granted to decrease. In order to effectively operationalize the independent
variable here being a political immigration issue, it is most effective to analyze it through
news media analysis. More specifically, through using the New York Times Article Archive,
the number of times each country was brought up in articles pertaining to immigration can be
easily determined. Furthermore the search engine available allows the narrowing down of
these articles within a given year. For example, a search with key terms Mexico and
Immigration between January 1, 1997 and December 31, 1997 yields 181 results (New
York Times). This is a feasible number of records to scan for relevance to United States
immigration politics. This allows invalid cases, such as results that discuss foreign
immigration issues rather than domestic, to be thrown out by viewing the content of each
article. Again, this would be a ratio variable with zero indicating no mentions of that region
in immigration discourse. Some may argue that this is not a valid representation of United
States politics, but as the Times is the newspaper of record it is the most consistent and
available method to collect data on salient political issues in domestic politics. Using the
Times rather than presidential speeches or congressional hearings or bills is advantageous in
that it speaks to what is salient enough in domestic politics to be published in a national
34


paper, whereas political actors may avoid these issues intentionally as they may be thought of
as a political minefield.
Research Design and Discussion
This study is a large-N, quantitative, non-experimental research project. This is the
best approach for the purposes of this project because in order to assess United States
refugee and asylum policy, which covers multitudes of cases, it is most fitting to take into
account as many cases as possible. The central research question of why does RAA vary
annually is best answered by looking at the whole population, which is why a quantitative
study is most the most appropriate approach.
Upon collection of the data as described above, a statistical multivariate linear
regression model will be employed to test these hypotheses along with alternative
independent variables to account for alternative explanations. A more in-depth analysis of
these results, especially with regards to what they imply for United States refugee and
asylum policy towards Latin America will then be discussed at length.
If the hypotheses are supported, the expected results would reveal a negative
relationship with both independent variables the dependent variable and would imply that
being a topic of discussion in immigration politics and/or having increased violence
committed by non-state actors would correlate with more restrictive outcomes in asylum and
refugee cases.
Much of the research in this particular area of study has been qualitative, and in many
cases has come from legal scholarship (Aleinkoff 2003; Bauer 2003; Bonilla-Mathe 2013;
Estevez 2013; Longmire 2014; Rexton-Kan 2011). While these studies certainly discussed
the trends in general terms and used specific cases to support their arguments, there lacked
35


any statistical analysis of the variables they discussed. This study will take into account many
of the variables they brought up, only in a quantifiable fashion so as to provide a means for
statistical analysis. As opposed to a legal, semantic, or theoretical critiques as others have
done, this study aims to observe the relationships between policy outcomes and possible
explanatory factors.
The methodology in this project is similar to that of Rosenblum and Salehyan (2004)
as theirs was a quantitative assessment of the norms versus interests debate with regards to
United States asylum outcomes. This research, however, is distinct from this study in that it
focuses on two major factors that could potentially impact the ways in which the United
States carries out refugee and asylum law towards Latin America specifically, that is
domestic immigration politics and the spread of violence by non-state actors in Latin
America.
Testing the hypotheses of this research project effectively analyzes United States
Refugee and Asylum outcomes in Latin America by looking at two issues that pertain to
Latin America refugee flows: violence committed by non-state actors and United States
immigration politics. This study is limited in that large n quantitative studies like this often
overlook substantive aspects that smaller, more qualitative studies may be able to capture.
Taking that into consideration, however, a quantitative analysis provides a more accurate
portrayal of the trends taking place. An individual case may illuminate the possibilities
within refugee law whereas a quantitative study can statistically analyze the relationship
between two variables and determine if a correlation exists.
The strongest aspect of this research design is that it distinguishes between violence
committed by state actor and that committed by non-state actors. This is an essential
36


distinction to make when focusing the analysis on Latin America considering the cartel and
gang violence this region has endured. Rosenblum and Salehyan (2004) used only the
Political Terror Scale in their analysis which excludes violence committed by non-state
actors. By incorporating the Global Terrorism Database and including non-state actors into
the analysis, this study accounts for cartel and gang violence as a potential factor in this
equation.
37


CHAPTER IV
RESULTS
One of the first steps to complete the analysis was to set the data as panel data,
combining state (appendix A) and year between 1980 and 2013. Then a summary was run on
all of the data collected. This data set resulted in 1,155 cases total with very few cases of
missing data. The exception here is the independent variable measuring illegal immigration.
Beyond the fact that this variable is based on estimation, there was very little available data
on this variable. Using the Office of Yearbook Statistics, only 40 observations were
available during the entire timeframe so unfortunately this variable did not produce many
results. For this reason, this variable has been left out of much of the analysis.
The range of the dependent variable, refugee and asylee admissions, fell between 0
and 43,455 with a mean of 819.21. As predicted both the higher end of this range and the
mean are thrown off by Cuba. When Cuba is excluded, the range is between 0 and 25, 625
and the mean is 226.51. Cuba alone has a range of 3,885 and 43,455 and a mean of
18,402.61. Two other states that showed large numbers of admits were Haiti and Venezuela.
These, along with four other states summaries are shown below in table 1.
Summaries of the independent variables were also run (appendix B). The
independent variables from the two hypotheses showed that for HI measuring violence from
non-state actors, the annual terrorist incidents reported from the GTD fall between 0 and 710.
As for H2 and domestic media attention, the New York Times articles that discussed the state
in question and immigration ranged from 0 to 428 annually. As a final preliminary step,
correlations were run with each independent variable and the dependent variable to examine
the direction of the relationship for each. A correlation matrix was also run to investigate the
38


direction of the relationship of all independent variables with RAA (appendix C). The
relationships were all relatively weak with the strongest relationship at 0.539 between civil
liberties and RAA. Once the preliminary overview was complete, I moved on to both
bivariate and multivariate regressions which will lead to a much more in depth analysis
Table One: Refugee and Asylee Admit Summary
State Range Minimum Range Maximum Mean
Total-All states 0 43,455 819.21
Cuba 3,885 43,455 18,402.61
Mexico 6 605 159.81
Haiti 0 25,625 2,184.69
Colombia 13 12,591 1,403.3
Venezuela 3 2,497 517.18
El Salvador 0 1,249 304.2
Guatemala 2 1,559 341.96
Total-excluding Cuba 0 25,625 226.51
Regression Analysis
In order to provide the most complete analysis, I separately ran bivariate regressions
with each independent variable (Appendix E) then a multivariate regression with all of the
variables include (Table 2). Table 2 reports the coefficient and P-value for the multivariate
regression. Two models were also initially used to analyze the data accounting for Cuba
through use of a dummy variable. While this distinction made a difference in the
coefficients, it did not change which variables were statistically significant. The illegal
39


immigration variable has been excluded from these tests due to such minimally available
data.
As hypothesized, domestic media attention towards immigration from the state in
question had a negative relationship with refugee and asylee admits. The freedom index was
split with political rights showing a statistically significant negative relationship and civil
liberties showing a positive relationship and no statistical significance. This is particularly
interesting because not only do these two measures come from the same source, but they are
highly correlated with one another (.873). This indicates that political rights have a negative
statistically significant relationship to RAA while civil liberties do not. As political rights
become more repressed RAA goes up.
The other variables that turned up significant were trade and military aid though both
of these turned out to have the opposite relationship to what was hypothesized. Whereas
both of these had been hypothesized to have a negative impact on RAA, they both turned out
positive. As trade goes up so does RAA. This is also true of United States military aid. The
more aid that was allocated as military aid towards a state, RAA went up from that state also.
This indicates that close economic relations and the United States providing military aid
positively affect RAA rather than the negative reputation effect that had been hypothesized
as the literature had discussed.
Another interesting aspect of these results is that, as demonstrated earlier, Cuba is an
extreme outlier, yet when Cuba was controlled for, the direction and the significance of each
variable remained unchanged. While Cubas circumstances are certainly unique and its
refugee and asylee admits are consistently higher than all other Latin American states, Cuba
40


did not throw off the data as expected. The r-squared measure does however jump from
.0036 to .7676 which indicates that controlling for Cuba does create a much closer model fit.
Table 2 Results:
Model 1-whole data set Model 2-data set with
added Cuba dummy
variable
N=663 N=663
R-Sq=.0036 R-Sq=7676
Coef. P-Value Coef. P-Value
Trade-Trade balance with US by partnering state .0599* .002 .04* .004
Political Rights-Freedom Index -237.1028* .083 -277.82* .021
Civil Liberties-Freedom Index 168.2124 .334 30.72 .82
Violence by state actors- Political Terror Scale 97.27 .485 80.57 .467
Violence by non-state actors- Global Terrorism Database -1.3 .377 -1.14 .325
Domestic Media-New York Times articles -20.81* .000 -14.94* .000
US military aid-USAID Greenbook 7.99* .001 8.35* .000
Constant 835.56 .22 -456.91 .083
* indicates statistical significance where P<0.1
Accounting for Eras
While the data the provided comprehensive results shown above, one thing that was
lacking was any insight as to the impact of historical events. Between 1980 and 2013 there
41


were several major events that could have influenced United States policy, and three of these
stand out as pivotal moments in United States history particularly regarding Latin America:
the end of the Cold War, NAFTA, and 9/11. Considering this, the data was then organized
into a few different models to investigate the influence of these events under the following
time constraints. The data was divided into the following categories: through and after 1989
to look at the possible impact of the Cold War (table 3), before 1994 and 1994 onward to
examine the possible influence of NAFTA (table 4), and up through 2001 and after to look at
the results pre and post 9/11 (table 5). The tables report the coefficient and P-value for each.
The results show that major events like this do in fact impact the data. A closer look at each
model will reveal the degree to which each of these events may have impacted refugee and
asylee outcomes.
The End of the Cold War
To analyze the possible impact of the Cold War on RAA, the data was split into two
timeframes: 1980-1989 and 1990-2013. Once the timeframe has been narrowed down to
1980 through 1989, the only statistically significant variable is a negative relationship
between political rights and RAA. After 1989 the results change, and the statistically
significant variables become trade, media, and military aid. So pre-1989 the less politically
free the state in question, the more likely refugees and asylees from that country are to
increase. This indicates an impact of Cold War politics on refugee and asylee outcomes.
Again this is not true of civil liberties which only further stresses the salience of the politics
of the Cold War. After the Cold War was over other factors become more important such as
domestic politics in the media and economic factors, such as trade and military aid. These
42


differing results point to the policy shift that occurred as the Cold War came to an end and
provides supporting evidence for how this shift affected RAA coming from Latin America.
Table 3 Results:
1980- 1990-2013
1989
N=80 N=583
R-sq=.12 R-sq=.01
Coefficient P-value Coefficient P-value
Trade-Trade balance -.18 .0612*
with US by partnering P=-.601 P=.003
state
Political Rights-Freedom -770.20* -100.98
Index P=.039 P=. 511
Civil Liberties-Freedom 504.03 58.53
Index P=.319 P=.771
Violence by state actors- 377.89 54.94
Political Terror Scale P=365 P=.711
Violence by non-state -2.74 -1.24
actors-Global Terrorism P=.437 P=.471
Database
Domestic Media-New -.04 -20.72*
York Times articles P=998 P=.000
US military aid-USAID 3.43 8.49*
Greenbook P=.854 P=.001
Constant -943.29 924.4
.397 .219
* Indicates statistical significance where P<0.1
43


NAFTA
Another event that has been pivotal for United States-Latin American relations was
the passage of NAFTA as it deregulated trade barriers. While it was initially a trilateral
agreement between the United States, Canada, and Mexico it marked the beginning of a shift
towards the massive expansion of trade between the United States and Latin America and
was eventually extended to include Guatemala, El Salvador, Honduras, Costa Rica,
Nicaragua, and the Dominican Republic. The process of splitting the data into two
timeframes was also applied here: 1980-1993 and 1994 to 2013 to analyze the possible
changes pre and post NAFTA. Unlike the previous example, it does not appear as though
NAFTA made much of an impact on the outcome of RAA across the timeframe. Just as in
the previous results before 1994 the only statistically significant factor is political rights.
After 1994 this significance is lost. Instead, trade, domestic political media coverage, and
United States military aid become statistically significant and the direction of the
relationships remain consistent with all previous models.
One interesting point of analysis is to look at these results and the deterrence through
development concept that was previously discussed. According to this notion trade should
have a negative relationship with RAA not because the United States would accept fewer
refugees from these places but because more opportunities become available and quality of
life should improve. If that were the case, it should be especially evident after something like
NAFTA that allows for the free flow of goods and services and more export processing
zones. According to this logic and the purported outcomes of free trade, this relationship
should be negative when in fact this relationship turned out to be positive. The more trade
44


between the United States and another state, refugee and asylee admits go up. This implies
that trade could contribute to push factors rather than creating more stability.
Table 4 Results:
Coefficient P-value Coefficient P-value
Trade-Trade balance with US -.06 .06*
by partnering state P=74 P=01
Political Rights-Freedom -578.20* -154.07
Index P=.007 Tf II L
Civil Liberties-Freedom 235.14 1.65
Index P=347 P=99
Violence by state actors- 161.13 75.47
Political Terror Scale Tf II L P=67
Violence by non-state actors- -1.16 -4.49
Global Terrorism Database P=57 P=. 31
Domestic Media-New York -6.8 -22.03*
Times articles P=594 P=.000
US military aid-USAID 1.94 7.33*
Greenbook P=858 P=.014
Constant -647.04 6.83
P=33 P=427
* Indicates statistical significance where P<0.1
September 11, 2001
The final historical event selected to analyze as a possible impact on United States
refugee and asylum policy towards Latin America was 9/11. As national security became a
45


focal point of United States foreign policy, it begs the question as to whether this priority is
reflected in refugee and asylum admits. The timeframes selected were 1980-2001 and 2002-
2013. The results here do show a distinction between pre and post 9/11. When this event
splits the data, the results look very different. The earlier model of 1980-2001 shows only
the only statistically significant variable as a negative relationship with political rights. Post
9/11 the only statistically significant variable is a positive relationship with the Political
Terror Scale which measures state sponsored violence as hypothesized. Trade, media, and
military aid, which had all turned out to be significant in the latter half of each previous
model, now all turn out not to be not statistically significant post 9/11.
Another observation is that from 1980-2001, as in the case of all the earlier parts of
the time-based models, political rights again show to be statistically significant and that this
significance goes away when looking forward from that year. A broader look at these results
show a negative relationship with political rights any time the data include the earlier years
whether it be pre-1989, pre-1994, or pre-2001. Taking this into account shows that the early
Cold War year heavily weighs the data towards a negative relationship with political rights.
The other variables mentioned trade, military aid, and media show up as significant in the
whole data set as well as in the post 1989 and post 1994 models, yet are not significant in the
post 9/11 model. During these middle years, trade, media, and military aid all impact RAA.
Then in the post 9/11 era, state sponsored violence becomes the only statistically significant
variable. Furthermore, this is not the case with non-state actors although these two are
positively correlated with one another. While the central hypothesis of non-state actor
violence having a negative relationship with RAA did not turn out to be statistically
significant, the fact that state sponsored violence becomes the only significant variable is
46


important. Just as political rights were prioritized during the Cold War, it seems as though
after 9/11 another major shift has occurred and political violence rather than political rights
are now impacting RAA.
Table 5 Results:
Coefficient P-value Coefficient P-value
Trade-Trade balance with US .009 -.022
by partnering state P=837 P=. 631
Political Rights-Freedom -402.56* -169.65
Index P=.006 P=.621
Civil Liberties-Freedom -73.71 29.009
Index P=682 P=.947
Violence by state actors- 2.83 483.08*
Political Terror Scale .986 P=.049
Violence by non-state actors- -.019 -25.14
Global Terrorism Database P=989 P=. 115
Domestic Media-New York -2.04 3.61
Times articles P=642 P=.674
US military aid-USAID -1.75 9.56
Greenbook P=698 P=.843
cons -533.07 26.17
P=. 321 P=983
* Indicates statistical significance where P<0.1
47


Conclusion
An overview of these results produce some significant takeaways. First while Cuba
was a major outlier, it did not impact the data as expected. However, Cuba being a Soviet
ally in the Western Hemisphere is a major part of what makes Cuba an outlier, and the Cold
War years included in the data from 1980-1989 do affect the data. All of the models that
include these years show political rights to be positively related at a significant level and
anytime these years are excluded, this significance goes away. Civil liberties which is highly
correlated with political rights and is taken from the same source (the Freedom Index) never
results as being statistically significant. This highlights the heavy influence of Cold War
politics over almost any other variable.
Another interesting point is the relationship between trade, military aid, and media
with RAA. All of these are significant in the whole dataset and from 1990 forward and 1994
forward. Trade and military aid both resulted as having positive relationships with RAA.
Rather than the idea that the United States will not take refugees from states with close
economic ties, the opposite has resulted. While it may be that the United States is taking
more refugees from states it is economically tied to, it may also be the case that increased or
deregulated trade and military aid have a destabilizing effect and create push factors. Further
research could investigate that as a possibility.
The decision to exclude illegal immigration data due to lack of availability is an
aspect of the study that could have drastically altered the results. Had this information been
available it would have provided more insight into the role of this variable on RAA. This
could possibly be investigated in future research by focusing on the states that are primary
sources of illegal immigration and their asylum approval rates.
48


Finally the post 9/11 shift towards political violence as the only statistically
significant variable show another shift in United States policy. This shift shows that the
current US foreign policy with its focus on securitization has a potential impact on refugee
and asylum policy. This too could be a future avenue for research especially with the rise of
new security threats and refugee crises. Another application of this research would be to
investigate how these results compare to other regions of the world. Considering the unique
characteristics of Latin America such as proximity and high levels of undocumented
immigrants there could be many differences. There are also variables that play out very
different in other regions such as non-state actor violence. It would be very interesting to
apply this research to other regions to see what kind of results these variables produce.
49


BIBLIOGRAPHY
Aleinikoff, Alexander. 2003. "Protected characteristics and social perceptions: an analysis
of the meaning of membership of a particular social group." Refugee Protection in
International Law. Cambridge University Press Accessed at
http://www.refworld.org/docid/470a33b30.html
Angotti, Thomas 1996. "Latin American urbanization and planning: inequality and
unsustainability in North and South." Latin American Perspectives) 12-34.
Barbera-Hernandez, Lila 2005. "Indigenous peoples, human rights and natural resource
development: Chile's Mapuche people and the right to water Annual Survey of
International & Comparative Law. 11-1.
Bauer, Laura Isabel. 2003. "They beg our protection and we refuse: US asylum law's failure
to protect many of today's refugees." Notre Dame Law Review 79: 1081.
Bonilla-Mathe, Isabel. 2013. "To group or not to group; refugee status for Central American
victims of gang violence." Tulane Journal of International & Comparative Law 22: 145.
Cabrera, Jorge. 2014. Why arent Central Americans getting asylum in the US?
International Business Times. Retrieved from http://www.ibtimes.com/why-arent-central-
americans-getting-asvlum-us-1653494
Charlton, Roger, Lawrence T. Farley, and Ronald Kaye 1988. "Identifying the mainsprings
of US refugee and asylum policy: A contextual interpretation." Journal of Refugee Studies 1,
no. 3-4: 237-259.
COHA June 24th 2010 Disparities in US immigration policy toward Jaiti and Cuba
accessed at:http://www.coha.org/disparities-in-u-s-immigration-policy-toward-haiti-and-
cuba-a-legacy-to-be-continued/
Displaced Persons Act. 1948. Public Law 82-60. United States Statutes at Large. Volume 65
Stat. 96
Estevez, Ariadna. 2013. "The biopolitics of asylum law in Texas: The case of Mexicans
fleeing drug violence in Juarez." Norteamerica, Revista Academica del CISAN-UNAM 8.
Gibney, M., Cornett, L., Wood, R., & Haschke, P., Political Terror Scale 1976-2012. Date
Retrieved, from the Political Terror Scale Web site:http://www.politicalterrorscale.org/
Haass, Richard N. 1998. Economic Sanctions and American Diplomacy. Council on
Foreign Relations.
50


Immigration Policy Center, 2015. Mexican and Central American asylum and credible fear
claims: background and context Accessed at: http://www.immigrationpolicv.org/
Kornbluh, Peter 2003. The Pinochet File: A Declassified Dossier on Atrocity and
Accountability. New York: The New Press, p. 28. ISBN 1-56584-936-1.
Lindsay-Poland, John 2003. Emperors in the Jungle: The Hidden History of the U.S. in
Panama. Duke University Press, p. 118.
Longmire, Sylvia. June 26,2014. "Should Migrants Fleeing Gang Violence in Central
America Be Accorded Refugee Status?." Small Wars Journal, Accessed at:
http://www.isn.ethz.ch/Digital-Library/Articles/Detail/7idM81403
Martin, Daniel and James Yankay. 2014. Refugees and Asylees: 2013. Office of
Immigration Statistics.
McBride, Michael J 1999. "Migrants and asylum seekers: policy responses in the United
States to immigrants and refugees from Central America and the Caribbean." International
Migration 37-1:289-317.
Migration Policy Institute, 2015. Accessed at: http://www.migrationpolicy.org/
Migration and Refugee Assistance Act. Public Law 87-510. 1962 United States Statutes at
Large. Volume 76 stat 121.
National Consortium for the Study of Terrorism and Responses to Terrorism (START).
(2013). Global Terrorism Database [Data file]. Retrieved from http://www.start.umd.edu/gtd
New York Times. New York Times Article Archive 1981 to present.
http://www.nytimes.com/ref/membercenter/nytarchive.html Accessed April 17th, 2015
Refugee Council, 2014. History of the US Refugee Resettlement Program, Presidential
Determinations, and Refugee Admissions Levels Accessed at
http ://www. refugeecouncil. org.uk/
Rexton-Kan, Paul 2011. "Mexico's1 narco-refugees': the looming challenge for US national
security." The Strategic Studies Institute. The US Army War College.
Rosenblum, Marc R., and Idean Salehyan. 2004. "Norms and interests in US asylum
enforcement." Journal of Peace Research 41, no. 6: 677-697.
Finnemore, Martha, and Kathryn Sikkink. 1998 "International norm dynamics and political
change." International organization 52-04: 887-917.
51


Ibrahim, Maggie 2005. "The Securitization of Migration: A Racial
Discourse l." International migration 43-5: 163-187.
Kay, Cristobal 2001. "Reflections on rural violence in Latin America." Third World
Quarterly 22-5: 741-775.
Lee, Everett S. 1966 "A theory of migration." Demography 3-1: 47-57.
Schoultz, Lars 1998. Beneath the United States. Harvard University Press.
Suhrke, Astri. 1994. "Towards a comprehensive refugee policy: Conflict and refugees in the
post-cold war world." in Aid in place of migration, 13-38. eds W.R. Bohning and M-
.L.Schlueter- Pareyes. International Labor Office.
United States Citizenship and Immigration Services .2013. Refugees. Accessed at
http://www.uscis.gov/humanitarian/refugees-asylum/refugees
United States Department of Homeland Security, Yearbook of Immigration Statistics
Refugees and Asylees. Data Tables, http://www.dhs.gov/yearbook-immigration-statistics.
Accessed April 13th, 2015.
United States Department of Justice. Executive Office for Immigration Review. Statistics
and Publications, http://www.justice.gov/eoir/efoia/foiafireq.htm. Accessed April 13th,
2015.
United States Refugee Act of 1980. Public Law 96-212. United States Statutes at Large.
Volume 94 stat 102.
Zolberg, Aristide R., Astri Suhrke, and Sergio Aguayo. 1992. Escape from violence:
Conflict and the refugee crisis in the developing world. New York: Oxford University Press.
Zucker, Norman L., and Naomi Flink Zucker. 1992. "From Immigration to Refugee
Redefinition: A history of refugee and asylum policy in the United States. "Journal of Policy
History 4, no. 01: 54-70.
52


APPENDIX A
List of States in Data Set
Antigua/Barbuda, Argentina, Bahamas, Barbados, Belize, Bolivia, Brazil, Cayman Islands,
Chile, Colombia, Costa Rica, Cuba, Dominica, Dominican Republic, Ecuador, El Salvador,
Granada, Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexico, Nicaragua, Panama,
Paraguay, Peru, Saint Kitts and Nevis, Saint Lucia, Saint Vincents and Grenadines,
Suriname, Trinidad and Tobago, Uruguay, Venezuela.
53


APPENDIX B
Sum of All Data
Variable Obs Mean Std. Dev. Min Max
state 0
stateid 1155 17.49884 9.813364 1 34
year 1155 1995.848 38.66581 1889 2813
admit 1012 819.2185 4815.421 8 43455
trade 888 2214.874 7455.363 .9 74795.6
freepr 1112 2.559353 1.62565 1 7
freed 1112 2.815647 1.489498 1 7
gtd 1895 24.79323 87.64664 8 718
pts 1811 2.588495 1.181855 1 12
illegal 48 1567325 2488736 118888 7838888
media 1155 12.62857 35.19113 8 428
milaid 1155 8315388 4.34e+87 8 5.61e+B8
cubadum 1155 .8294372 .1691819 8 1
54


APPENDIX C
Correlation Matrix
year admit trade freepr freed gtd pts illegal media
year 1.0000
admit 0.4969 1.0000
trade 0.1991 0.0422 1.0000
freepr 0.2559 0.0846 -0.3939 1.0000
freed 0.2792 0.5389 -0.3552 0.6268 1.0000
gtd -0.0502 -0.0365 0.5995 -0.2631 -0.2048 1.0000
pts 0.1017 0.1556 0.5380 0.1696 0.0744 0.1684 1.0000
illegal 0.1920 0.0656 0.9810 -0.4415 -0.3546 0.5234 0.5263 1.0000
media 0.1702 0.0913 0.9545 -0.3882 -0.2952 0.6381 0.4939 0.9260 1.0000
milaid 0.2854 -0.0213 0.6058 -0.0587 -0.2451 0.1877 0.4646 0.5977 0.5138
milai
1.000
55


Full Text

PAGE 1

US REFUGEE AND ASYLUM LAW AND LATIN AMERICA by ADRIE GARCIA B.A. Metropolitan State University of Denver 2013 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment O f the requirements for the degree of Masters of Arts Political Science Program 2016

PAGE 2

ii This thesis for the Master of Arts degree by Adrie Anna Garcia H as been approved for the Political Science Program B y Michael Berry, Chair Lucy McGuffey Ryan Crewe April 28th 2016

PAGE 3

iii Garcia, Adrie Anna MA, Political Science US Refugee and Asylum Law and Latin America Thesis dire cted by Professor Michael Berry ABSTRACT The United States and Latin America have a very complex history that continues to this day. Many decisions and actions taken by the United States have had extensive implications throughout Latin America. This is particularly true of immigration issues. This study narrows this down and investigates the role of United States refugee and asylum policy and how this has played out in Latin America. It investigates several factors that may have an impact on refugee and asylee admissions from Latin Ame rica. The form and content of this abstract are approved. I recommend its publication. Approved: Michael Berry

PAGE 4

iv TABLE OF CONTENTS CHAPTER I. INTRODUCTION ................................ ................................ ................................ ............ 1 A Brief History of United States Refugee Law ................................ ....................... 4 II LITERATURE REVIEW ................................ ................................ ................................ 8 Heavy Burden of Proof and Su bjective Judgement ................................ ................. 9 Domestic Politics and Foreign Policy ................................ ................................ .... 12 Dete rrence Through Development ................................ ................................ ......... 1 5 The Tug of War Between Norms and Interests ................................ ..................... 16 Why the Actor s Matter: State or Non State ................................ ........................... 2 1 Discussion and Critique ................................ ................................ ......................... 24 III METHODOLOGY AND EXPLANATORY FRAMEWORK ................................ ... 26 Human Rights and National Interests ................................ ................................ .... 27 Type of Threat ................................ ................................ ................................ ........ 28 Politics ................................ ................................ ................................ .................... 29 Hypotheses ................................ ................................ ................................ ............. 30 Data and Sample ................................ ................................ ................................ .... 31 Measurement, Methodology, and Expected Results ................................ .............. 32 Research Design and Discussion ................................ ................................ ........... 35 IV RESULTS ................................ ................................ ................................ .................... 38 Regression Analysis ................................ ................................ ............................... 39 Accounting for Eras ................................ ................................ ............................... 41

PAGE 5

v The End of the Cold War ................................ ................................ ....................... 42 NAFTA ................................ ................................ ................................ .................. 44 September 11, 2001 ................................ ................................ ............................... 45 Conclusion ................................ ................................ ................................ ............. 48 BIBLIOGRAPHY ................................ ................................ ................................ .............. 50 APPENDIX A. List of States in Data Set ................................ ................................ ............................... 53 B. Sum of all data ................................ ................................ ................................ ............... 54 C. Correlatio n Matrix ................................ ................................ ................................ ......... 55

PAGE 6

1 CHAPTER I INTRODUCTION The relations between the United States and Latin America have always been complex. In recent decades this complexity continues to grow. From the Monroe Doctrine to the Good Neighbor Policy, from the communist tensions of the Cold War to the North American Free Trade Agreement (NAFTA), United States foreign policy has had highly consequential implications throughout Latin America. This region has endured many civil wars and brutal dictatorships that have displac ed countless people (Kay 2010). In 1954 The United States protected its United Fruit Company interests in Guatemala which led to years of civil war (Schoultz 1998). In 1960 the United States imposed an embargo on Cuba which put Cuba at a devastating econ omic disadvantage that has lasted decades (Haas 1998). 1973 the CIA helped orchestrate a coup against the democratically elected president of Chile, Salvador Allende, and the Unites States supported the brutal military regime of Augusto Pinochet, which las ted until 1990 (Kornbluh 2003) The United States invaded Panama in 1989 to remove Manuel Norriega from power, an operation that cost up to 1,000 Panamanian casualties (Lindsay Poland 2003) These are just a few examples of how United States Foreign Policy has directly affected Latin American states. Another example of foreign policy that has had a significant im pact on Latin America is the contributed to the displacement and migration o f populations from Latin America (Anghotti

PAGE 7

2 1996). Free trade has brought with it an era of, urbanization, maquiladoras, and race to the bottom tactics by exploitative multinational corporations. These implications and countless others are among the facto rs that have contributed to migration flows throughout the region. lack of opportunity and a daily threat of violence that essentially spawn migration. The othe economic conditions and an escape from political violence that attract migrants (Lee 1966). While many immigrants pushed from Latin America are migrants fleeing poverty and seek ing greater opportunities, there are also many who are fleeing persecution and violence in their home state. These migrants are considered refugees and asylees. This determination is made people who have been pers ecuted or fear they will be persecuted on account of race, religion, nationality, and/or membership in a particular social This is a definition that has developed over time, first with the 1951 Convention on the Status of Refugees which originally only addressed European refugees fleeing persecution during World World II (UNHCR 2015). Later this definition evolved to remove the time and geographic constraints with the 1967 Protoco l Relating to the Status of Refugees (UNHCR 2015). Throughout this timeframe the United Nations High Council on Refugees (UNHCR) has been a major driving force behind this development and has helped to address many

PAGE 8

3 refugee crises around the world (UNHCR 2 015). While the UNHCR deals with refugee crises at the international level, individual states must also address these issues as they are called upon to help. The United States has repeatedly been faced with refugee flows and has responded in various ways An example of this variation can be found in comparing Cuba and Haiti. In the case of Cuba the United States has been very open to accepting refugees and asylees whereas in Haiti the United States has not been so generous despite many periods of violen ce and unrest (COHA 2010).. The goal of this research project is to examine specifically how United States policy has dealt with refugees and asylees from Latin America. Why have refugee and asylum numbers varied from state to state over time? Have Unite d States interests or relations in the region affected the response of the United States to these populations, or has the United States focused on a human rights based approach? In order to do so it is necessary to find measures for such variables and col lect and analyze the available data. This will help to investigate the correlations that exist among these factors, if any correlations exists at all. This study will show that there are correlations between these variables and that even more is revealed when the data is broken out over specific time frames. Before addressing these specific results it is first necessary to discuss how the literature has approached the issues in question and to provide a brief overview the development of United States ref ugee law.

PAGE 9

4 A Brief History of United States Refugee Law As previously mentioned, the S econd World War led the international community to increased recognition of the need to protect individuals fleeing per secution from their home states This human right s concern was one of many stemming from the war, and the result was the emergence of many international institutions to address these concerns. From President Roosevelt presiding over the initial United Nations (UN) Charter to Eleanor the development of the Universal Declaration of Human Rights (UDHR), the United States has self proclaimed a central role in the development of an international human rights regime. As this regime began to grow, human rights became a touted United States value. Consequently, a new body of law addressing asylum and refugee issues was born in the United States with the passage of the Displaced Persons Act in 1948 ( Public Law 82 60 65 stat 96). This legislation directly addressed European immigrants fleeing persecution after World War II. A few years later, the UN brought forth the 1951 Refugee Convention which defined the rights of refugees and specified who was considered a refugee (UNHCR 2015). This was later broadened with the 1967 Protocol Relating to the Status of Refugees as it removed the geographic and time restrictions that had originally been in place to serve only those fleeing Europe after World War II (UNHCR 2015). With this new global focus, and with the Cold War heating up, the United States accepted hundreds of thousands of refugees from countries such as Cuba, China, Vietnam, Poland, Korea, Hungary, among others. Decades later, as a response to these large influxes of

PAGE 10

5 refugees, came the Migration and Refugee Assistance Act and the United Sta tes Refugee Act of 1980 ( Public Law 87 510, 76 Stat. 121 ;Public Law, 96 212 94 Stat 102). While the Migration and Refugee Assistance Act expanded programs and funding to accommodate influxes of refugees, the United States Refugee Act took measures to lim it refugee flows by imposing annual limits. The stated purpose of this act is explicitly linked to the United States human rights values. The precise wording states that the laws were to create a consistent efugees of special humanitarian concern to 212 94 stat 102). The combination of United States involvement in the development of this protection through international organizations and through the passage of domestic legis lation show the United States certainly subscribes to the international norm regarding the protection of refugees. According to Finnemore and itself a critical to open doors to those fleeing persecution. Over this time, the United States has developed a very robust refugee program, resettling more than three million people since 1975 ( Refugee Council 2014). However, while the United States refugee program is extensive, it is also limited. One of the ways in which the United States has restricted its flow of refugees is through refugee admission ceilings as laid out in the Refugee Act of 1980. Each year the president limits refugee admissions by region through the Refugee Admissions Ceiling (Martin and Yankay 2014).

PAGE 11

6 This ceiling pertains to refugees, or individuals seeking refuge from outside the United States. The ceiling excludes asylum seekers, who are individuals seeking this status from within United States borders or at a point of entry into the country. There is no dictated limit for asylum, however the numbers of asylees are generally significantly lower than the refugee ce iling (Refugee Council 2014). Considering both the limitations within the United States refugee program procedures and the human rights principles that are depicted within the legislation, the question arises as to how these determinations are made. Why do refugee and asylum admissions vary annually? Do they fluctuate in response to humanitarian crises as they arise, or are there other factors at hand that may explain this variation? With regards to United States refugee and asylum law, Latin America is a n intriguing region to examine as it consistently admits very few refugees when compared to other regions (Refugee Council 2014). Latin America also represents a low percentage of those granted asylum (Immigration Policy Center 2015). Considering the man y humanitarian crises throughout Latin America in recent decades, it is curious that these numbers have not grown in response. Additionally this calls into question the nature of humanitarian concerns the United States chooses to address. Many of the con temporary issues facing Latin America have been involving drug cartels and gang violence. The UNHCR has recognized this ddress such violence as factors to consider in refugee cases (Trk 2011). In Latin America, the 1984 Cartagena Declaration on Refugees

PAGE 12

7 took steps towards a more inclusive approach considering refugees in Latin America (Trk 2011). However, of the aforemen tioned steps international organizations have taken, the United States remains only party to the 1967 Protocol and has not yet produced any major domestic legislation regarding generalized violence. This begs the question as to whether US refugee admissio ns shows a similar response to this type of humanitarian crisis as it does others, such as civil wars or repressive dictatorships. Furthermore, considering the complicated dynamics in the region, could it be that factors outside refugee law influence adm relations with the various Latin American states, multilateral trade agreements, among other factors. Any of these elements could have a bearing on refugee and asylum outcomes, but do they, and if so what carries the most weight? This inquiry requires an in depth survey of the existing literature before proceeding.

PAGE 13

8 CHAPTER II LITERATURE REVIEW The literature on US refugee and asylum policy has considerable breadth and presents several perspectives, and from these perspectives certain recurrent themes arise. Among these themes, several predominant arguments stand out. One of these arguments is that the burden of proof that falls on the applicant for refugee or asylum is exorbitantly high, especially when having to prove something that has been only vaguely stated within the legal framework. Furthermore this language is so ambiguous that it leaves much room for subjective interpretation by immigration court o fficials. In a seemingly arbitrary judicial environment. this ambiguity impacts the outcomes of these cases (Aleinkoff 2003;Bonilla Mathe 2013; Zolberg et al 1992). Another argument is that refugee and asylum law is both politically motivated and produc es political outcomes (McBride 1999; Rexton Kan 2011; Charlton et al 1988; Suhkre 1994; Zolberg et al 1992; Zucker and Zucker 1996). This perspective emphasizes the influence of both domestic politics and foreign policy in United e flows. Other scholars have argued that there is a struggle between norms and interests in United States refugee and asylum policy (McBride 1999; Charlton et al 1988; Rexton Kan 2011; Rosenblum and Salehyan 2004). This point of view takes into account h ow both United States interests and human rights norms associated with refugees play into this process. Finally there is the argument that the United States refugee program responds very differently to different types of threats. More specifically, when non

PAGE 14

9 state actors are the principal source of the violence, the response is negligible (Bauer 2004; Bonilla Mathe 2013; Estevez 2013; Longmire 2014). It is critical to note that none of these perspectives are necessarily mutually exclusive, and that a c loser look at each may more comprehensively portray how United States refugee and asylum policies play out in reality. Furthermore, to consider the case of Latin America, what does this literature tell us about the consequences of such policies for refugee and asylum seekers from this region? Heavy Burden of Proof and Subjective Judgement When applying for refugee status or asylum in the United States, applicants must they were persecuted or fear persecution due to race, religion, nationality, political opinion, or 2013). While on the surface this definition seems rather concis e, there are certain concepts or circumstances within the language that could be considered open to interpretation. most glaring examples of ambiguous concepts. Ho w does one prove persecution or fear of persecution to an immigration official? If the applicant has been targeted in a highly visible conflict like a civil war, this may not be so difficult. However this may be much harder to prove with the daily violen ce in many Latin American states perpetrated by cartels, corrupt

PAGE 15

10 officials, and gangs. As previously mentioned, the Cartagena Declaration directly safety or freed om have been threatened by generalized violence, foreign aggression, internal conflicts, massive violation of human rights or other circumstances which have seriously g. Nor was the United States a participant in its creation. This leaves applicants in the United States without any type of direct protection on the grounds of generalized violence. Because of this, the burden of proof in these cases is much more out of re ach although the risk in their daily lives is still very prevalent. As Aleinkoff (2003) points out this process gets so bogged down in identifying the cause of the persecution that the actual goal of determining whether or not they face the risk of violen ce is undermined. In other words, refugee seekers may have difficulty identifying their persecutors or making a case for the root cause of the persecution while the primary issue, whether or not they face violence as a result of this persecution, is negle cted. Assuming the applicants can prove persecution, they must then prove that they fall ration Services 2013) This is yet another obstacle for applicants, especially if they fall into one of the latter categories as these require proof of a characteristic (Aleinkoff 2003). Political

PAGE 16

11 opinion and membership in a particular social group are the claims made by Latin American refugees, but these claims are often grounded in characteristics that are very difficult to extortion (Bonilla Mathe 2013). Th ese examples demonstrate how blurry these definitions can become and how the focus can fall away from the question of whether or not the applicants face a risk of violence and towards whether statutory regulations have been met. While the UN expressed the legitimacy behind both of these circumstances by offering them as examples in a handbook on refugees, the broad language of domestic refugee law causes the fate of such claims to lie in the interpretation of immigration officials (Bonilla Mathe 2013). Cons equently, as can be seen by the decisions of the Board of Immigration Appeals, these officials tend to favor very narrow interpretations of such definitions (Aleinkoff 2003). The applicant may have been able to present evidence of their circumstances to a n immigration judge that subjectively decides otherwise. These and asylum cases a challenge. It is also relevant to note that these broad definitions are not merel y prerequisites in a legal process, they also carry moral, political, and legal implications (Zolberg et al 1992). The applicants may be in a struggle with the language to prove their individual case, while in reality the political environment could highl y

PAGE 17

12 exacerbate these difficulties. Immigration flows invoke political responses. Refugee issues as a source of immigration become politicized in the process. Domestic Politics and Foreign Policy The increase of refugees in recent decades has created the wide spread belief that will come pouring in in masses (Zolberg et al 1992). It is this type of sentiment that has domestic political salience. This is particularly acut e in the case of Latin America, being situated so near to the United States. The logic behind the fear in this case is that if the United States starts validating the case that the generalized violence in Latin America in sufficient to begin accepting ref ugees, then it will open the floodgates for mass immigration across the Mexican border or through the Caribbean (Rexton Kan 2011). This fear and the political significance behind it were clearly illustrated during the unaccompanied minor refugee waves tha t entered the United States during 2013 and 2014. There were intense political debates and protests on both sides of the issue. This debate is one that is typical of immigration issues in domestic politics, with one side urging the humanitarian acceptanc e and the other pushing for greater restrictions (Zucker and Zucker 1992). The level of political influence goes far beyond this domestic debate however. Global politics factor into this entire process, from the root of the conflicts to the international response, to the direction of the refugee flows (Suhrke 1994).

PAGE 18

13 The international political component can be easily seen from the time of the Cold War, as the United States and the Soviet Union were highly involved in the competition for allies in the devel oping world at a time of widespread conflict including many liberation movements and civil wars (Suhrke 1994; Forsythe 1988). As the United States and the Soviet Union helped to arm both sides of these conflicts, the resulting violence created refugee out flows. The same Cold War politics that created such outward flows likewise dictated the direction of the movements of refugees (Suhrke 1994). Examples of Cold War politics dictating refugee flows would be the mass relocation of Vietnamese to the United St ates and the political refugee resettlement extended to Cubans, both states that were Cold War hot spots. The case of Cuba is particularly critical in this discussion. The United States absorbed massive amounts of political asylees from Cuba during the C old War without the major domestic political pushback that was previously discussed. This speaks to both the salience of the international politics of the Cold War and its influence on domestic politics. Additionally there was no major catastrophe as a r esult of such a massive refugee flow As the Cold War came to an end, however, there was little strategic incentive for the United Stat es to remain so highly active in these conflicts, and so began the withdrawal of aid (Suhrke 1994). This left states scrambling quickly for solutions, and in many places was

PAGE 19

14 coupled with the de stabilizing aftermath of colonialism and a strain on the dev eloping Cold War geopolitical environment proved quite different for refugees, where superpowers are much less likely to get involved directly in these conflicts and are no longer likely to launch major refugee programs (Suhr ke 1994). Haiti is a prime example of this shift. In 1991 and 1992 approximately 44,000 Haitians fled after the overthrow of President Artiside. The first response was a court order halting the processing of these cases in immigration court followed by the rerouting of 34,000 of these individuals to Guantanamo Bay. In the end, only 10,000 were approved for asylum status. The United States simply had no foreign policy interest invested in Haiti. When compared to the mass approval of political asylum s eekers fleeing Cuba, this case shows the extent to which the international political stage influences refugee and asylum outcomes (McBride 1999). A deeper look at the way the United States responded to Haiti shows that in place of accepting mass quantitie s of refugees, it is a strategy of deterrence that has taken hold. Rather than admitting refugees, the United States shifted towards investing in and encouraging development within these countries. The idea here is to deter refugee flows by investing in creating better economic conditions in the country of origin (Suhrke 1994; Xenos 1993).

PAGE 20

15 Deterrence through Development The deterrence through development concept implies that in a modern environment focused on free trade and global markets the United States will more readily take actions to avert the flow of refugees than admit refugees (Zolberg et al 1992). In the case o f Haiti, the United States refused to denounce support of the oppressive Duvalier regime but created an export processing zone in hope of sparking development and thereby avert the flight of Haitians while at the same time admitting few Haitian refugees (S uhrke 1994; Xenos 1993). This process ultimately failed as Haiti spun into further violence and refugee seekers increased. Despite the failure of such actions, it appears as though the United States has not abandoned this approach, especially with regard s to Latin America. Recent United States Latin American relations have shown that the United States continues to engage in many development projects while at the same time remaining very conservative in its refugee program there. A major justification f or this type of action is that economies and conditions will improve. This has yet to be seen in Latin America. While some states like Brazil and Chile have un dergone intense development and have improved their macro economies, there remain many marginalized populations that have been displaced or neglected as this process has occurred. An example specific to Brazil would be the tens of thousands evicted from s lums of Rio just before it hosted the World Cup

PAGE 21

16 (Waldron 2014). In the case of Chile the displaced Mapuche were forcibly removed from their native land to accommodate a water plant (Barbera Hernandez 2005). Examples such as these can be found throughout L atin America where development initiatives are pushed forward despite human rights concerns. Regardless of any question of intention or motivation behind these actions, these cases indicate that development has not been the most effective vehicle for limit ing refugee flows through societal advancement. Perhaps this is simply an example of how the United States has prioritized its economic interests over its self proclaimed identity as a human rights leader. These are questions some scholars have addressed through the lens of norms versus interests. The Tug of War Between Norms and Interests The United States Refugee Act of 1980 explicitly states that the refugee admissions the fiscal year and after appropriate consultation, is justified by humanitarian concerns or is 212 94 stat 102). These two options for justification set the stage for a struggle between refugee admission s based upon human Kan (2011) emphasizes this struggles in United States qualifies the cases of those fleeing drug violence in Mexico as refugees, the proximity and volume of the problem create a perceived national security threat of a major

PAGE 22

17 refugee flow at a time of contentious political debate over illegal immigration (Rexton Kan 2011). On the other hand, the United States dismissing these claims as illegitimate and Kan 2011, vi). This example highlights how prioritizing human r ights would mean the United States should accept narco refugees, but that doing this may run counter to national interests. If the United States takes a human rights approach, this introduces the risk of invoking a refugee crisis. Additionally, close tra de relations with Mexico puts the United States in an awkward position, for to begin to accept high volumes of Mexican refugees makes an international statement that the Mexican government is unable to handle the situation. In this struggle between norms and interests, which has more pull in refugee and asylum law? This case is similar to other Central American states that struggle with cartel and gang violence such as Nicaragua, El Salvador and Guatemala, to name a few. The literature reflects that the c omplexity of the relations between states and the ways refugee policy can reflect these relations make this comparison difficult to determine. While there is evidence that both norms and interests have an impact on refugee and asylum outcomes, some schol ars have argued that in the United States the national interest seems to carry more weight (Rosenblum and Salehyan 2004). Furthermore over time as interests have shifted, so has its refugee policy. During the Cold War the United States

PAGE 23

18 favored states wit h whom it had military ties, and therefore accepted fewer refugees from those states so as not to indicate that the regimes in question were committing human rights violations or were unable to address domestic concerns. As interests shifted towards econom ics, the United States began to favor states that were major trading partners, and therefore accepted fewer refugees from those states to maintain strong economic ties (Rosenblum and Salhyan 2004). This concept implies that the reputation of a state matter s when it comes to refugee policy and interstate relations. In other words, relations between the United States and a trading partner would become strained if the United States were to accept large amounts of refugees from the partner state as this would imply that the state in question cannot handle its internal conflicts. Another point of strategic interest to the United States refers back to the salience of T his implies that the growth of illegal immigration will create a negative impact on refugee and asylum applications from the state of origin of the undocumented immigration (Rosenblum and Salhyan 2004). This restrictionist approach is supported by the Pre ability to set the refugee ceilings, and is only furthered by negligible rates of asylum acceptance (Charlton et al 1988). Given the way refugee and asylum operate, the Quite simply, given a huge illegal immigrant population, the maximization of both executive

PAGE 24

19 discretion and ultimate political control is secured only by reluctantly granting any Another example is the recent influx of immigrants, most notably the unaccompanied minors, fleeing violence in Central America. The asylum approval rates asylum claim s from Guatemala, Honduras and El Salvador are successful, a jarringly low The proximity to the United States paired with the miniscule asylum approval rate shows t hat the national interest may play a role in the response to these asylum seekers. While a shift can clearly be seen in policy response due to the national interests, a similar shift can be seen from human rights factors. As the Cold War came to an end th e trends began to shift for refugee determinations, favoring democratic values and protecting individuals against states with higher levels of violence; this shows a possible commitment to human rights (Rosenblum and Salhyan 2004). The growth and evolutio n of both these factors make this contest difficult to analyze as the world rapidly changed during this time and both factors showed growth. What is essential to note about the interest factors is that they function with little to no regard towards the hum an rights conditions in the states of origin. In other words, national issues such as trade or illegal immigration trump the real

PAGE 25

20 risks these individuals face, which is in opposition to both the purpose of refugee law and of the self proclaimed values of the United States. as there is a high level of violent conflict while at the same time a high percentage of the undocumented immigrants in the United States. Furtherm ore, many of these states having close trade relations with the United States. While it is also true that norms and interests are not necessarily mutually exclusive and these factors may align at times, the literature points towards national interest infl uencing United States policy outcomes more than human rights norms (Rosenblum and Salhyan 2004). How would the United States justify such a preference for interests over norms when valuing human rights is so central to its identity? One answer may lie in the way the United States defines conflict. If refugee admissions have implications for interstate relations, the conflicts within these states would be defined in a way that reflects these relations. In the case of an amicable relationship, it would be b eneficial if the United States were to define the conflict in a way that absolves the state from culpability in a conflict and, in doing so, prevent the United States from being compelled to protect refugee seekers from this state. One way to do this is t o claim that the violence is generalized as opposed to targeted, and an easy way to make this claim is when the perpetrators of the violence cannot easily be identified by a flag, tank, or uniform.

PAGE 26

21 Why the Actors Matter: State or Non State While the wording of United States refugee law makes no explicit reference to the actors committing the violence, all of the conflicts refugees flee from fall into one of the following categories: a conflict where the violence is state sponsored, a confli ct that is perpetrated by non state actors, or a conflict in which this distinction is unclear. Does the United States respond differently to these scenarios? Some scholars say that it does, especially in the cases of Central American gangs and drug cart els (Bauer 2004; Estevez 2013; Bonilla Mathe 2013; Longmire 2014). While the burden of proof requires the applicants to prove their own persecution based upon political opinion, the state then negates many of these arguments due to the fact that the opini ons or characteristics brought forth are difficult to attribute to their political significance, especially when a non state actor is involved. Another challenge applicants face when non state actors are involved is the obstacle of proving that they were the target of such violence. Young people in specific areas may be the targets of gang recruitment or business owners may be targeted to cooperate with cartels, and the refusal of such actions certainly could produce a reasonable fear of persecution. This form of targeted persecution from a gang or cartel is much harder to verify or prove than when a mobilized military is targeting a specific religion (Bonilla Mather 2013).

PAGE 27

22 Despite the fact that the UN specifically addresses this caveat by stating that vio lence may in fact come from actors that do not heed state law, the United States generally focuses on state sponsored violence when considering refugee claims (Bonilla Mathe 2013). Bonilla Mathe (2013) cited several cases in which the courts inconsistentl y defined in direct conflict to UNHCR guidelines while widely accepting claims based on violence perpetrated by states actors. The UNHCR has directly addressed this very issue of stency in validating these claims shows a lack of recognition of UNHCR guidelines. Bauer stresses the implications The law fails to recognize, however, that the actions or inactions of a civilian caught in the midst of gene ralized political violence can be construed as having a political significance for which there are very real and sometimes deadly A major problem with this type of oversight goes back to definitions. The way in which the Unit refers almost exclusively to state actors while ignoring criminal organizations (Longmire 2014, 1). This is precisely why the UNHCR has moved towards a position that specifically

PAGE 28

23 add resses generalized violence. Equally important is the fact that the United States often aids these governments at the federal level to combat such organizations while corruption at the municipal level often leads to local government consorting with the cr iminal organizations (Longmire 2014). In the case of Mexico, their federal government is committed to working with the United States through the Merida Initiative, which is a bilateral agreement between the United States and Mexico to cooperate when fight ing drug trafficking and organized crime. This situation is difficult to assess considering that many local police and governments have ties to the drug cartels, hence making it difficult to establish if Mexico is unable or unwilling to protect the refugee seekers (Estevez 2013). This implies the United States may deny that certain violence is political or overestimate the involvement in light of efforts to stymie the conflict. In doing so, this allows the United States to avert refugee flows and maintain friendly relations with the state. It is also important that the above examples illustrate that the distinction between state and non state actors is not always cl ear. In some cases, state actors may be acting in cooperation with non state actors. Another possibility is that state and non state actors could be engaged in conflict with one another, and both could be committing acts of targeted acts of violence towa rds certain communities. Both of these examples highlight the difficulties of analyzing whether persecution comes from state and/or non state actors.

PAGE 29

24 However, just as in the norms and interests debate, a closer look could show whether policy outcomes lea n towards response to state or non state persecutors. Discussion and Critique This survey of the literature shows consensus leaning toward the notion that the United States focuses its refugee and asylum on politics and national interest over human rights considerations. The desire to hold back the possibility of any major refugee flow while creating a tension with Latin American governments appears to take precedence over human rights concerns of violent conflict in these states. The ambiguous nature of the wording in refugee and asylum law presents a monumental challenge for many individuals fleeing gang and drug war related violence. This language also allows for judicial interpretation of many refugee claims, which in the case of Latin America often e nd up claims Another contemporary trend the literature has yet to fully explore is the move towards securitization, especially within the United States. If immigratio n is perceived as a threat to national security, then the notion of risk gets turned on its head. From this perception refugees pose a risk to the host state rather than focusing on the vulnerability and risk the refugees take on (Ibrahim 2005). This is a particularly critical post in the post 9 11 environment where immigration discourse more often than not includes rhetoric of border security. Ideologically speaking, terrorism is the new evil of the day, replacing

PAGE 30

25 anticommunist sentiments of decades past. The War on Terror has replaced the Cold War as of paramount importance. For example, these criminal organizations are not framed as terrorists in either the dome stic news media or in domestic political discourse. Considering the widespread violence Mexico and Central America has experienced in recent years, the United States news coverage has been negligible. As Rexton Kan (2011) discusses with regards to Mexico, the United States should be paying attention to the nature of the violence at its borders. Not only is the violence widespread and enduring, it is also gruesome. It is so gruesome, in fact, that they have developed an entire vocabulary to categorize the ways in which cartels dismember the bodies of their victims (Rexton Kan 2011). This begs the question whether emphasizing the terrorist nature of these crimes would change the way these issues are viewed. Cartels and gangs being framed as terrorists could fleeing this type of violence. The opposite could also occur, and this redefinition could swing support towards more intense securitization with a fear of the violen ce crossing over. Considering the literature, would this even make a difference as they are often non state actors? The many questions left unanswered with regards to United States refugee and asylum policy and the ongoing violence in many Latin American states presents a ripe opportunity for research especially in regards to why refugee and asylee admissions vary annually? What factors have an impact on this variation?

PAGE 31

26 CHAPTER III METHODOLOGY AND EXPLANATORY FRAMEWORK In order to embark on any research on this topic, it is necessary to operationalize some of these complex factors. Effective conceptualization of the many variables in question is instrumental in determining possible correlations within refugee and asylum law and its response to Latin Americans fleeing violence. For the purposes of this proposed study, the dependent variables will be refugee and asylee admissions (RAA) (y). In order to explore the variance in outcomes throughout Latin America, the unit of analysis will be state/year. The measurement of this dependent variable is straightforward as the number of annual admits is available through the annually published Yearbook of Immigration Statistics (Department of Homeland Security). Refugee and asylum admissions are a ratio variable as the value of zero means there would be no refugees or asylees allowed from a region. These values would fall between zero and the maximum admissions for that year. Because there are specific states that are notably di fferent from the rest additional models will be presented to exclude these outlying cases in order to account for their effect on the data. The most obvious example of this is Cuba since there was a program designed specifically for Cuban refugees as discu ssed earlier. Because these numbers are likely to be much larger than for other states, it is necessary to parse this out for an accurate analysis of This study will be conducted with bivariate and multivariate regression analyses using independent variables grouped according to some of the main theoretical debates presented in the existing literature as follows: human rights concerns vs national interest,

PAGE 32

27 types of threat (state or non state actors ), and political factors. The measures for each of these are described in more detail in the following sections. Human Rights and National Interests Interests will be measured by United States relations with each state using the following variables: Int erests United States trade relations which is hypothesized to have a negative effect on RAA The more closely we are economically tied with a home state, the less likely the United States is to accept refugees from that state because this would imply that the government in incapable of effectively handling the issue. Interests United States military aid which is hypothesized to have a negative effect on RAA The more the United States invests in the military operations of the home state, the less likely the United States is to accept refugees from that state because again that would imply the government the United States is attempting to assist is incapable of handling the issue. Human rights concerns will be measured by indicators that show where states fall concerning humanitarian issues (also including the question of states that share democratic ideals as discussed in the literature review) using the following variables: How high the state scores on the Freedom House index is hypothesized to have a pos itive effect on RAA The freedom index is on a scale from 1 to 7 with 7 being the least free, therefore the lower the score on the freedom index, the United States would be less likely to accept

PAGE 33

28 refugees because these would be states that have more politic al rights and civil liberties. The higher the score on the freedom index would indicate that these states have less political rights and civil liberties and it is hypothesized that the United States would accept more refugees and asylees from these states How high the state scores on the Political Terror Scale is hypothesized to have a positive impact on RAA The higher the score on the Political Terror Scale would implicate poor human rights conditions and based on these hypotheses would therefore imply a higher level of refugee and asylees accepted. The higher the number of incidents on the Global Terrorism Database is hypothesized to have a positive impact on RAA The more incidents on the Global Terrorism Database would likewise implicate poor human rig hts conditions, and would include those not accounted for on the Political Terror Scale, and it is hypothesized that this would lend to a higher level of refugee and asylees accepted. Type of Threat The type of threat will be measured by a comparison of the following variables: The higher the rate of incidents that include state sponsored violence is hypothesized to have a positive impact on RAA, as published by the Political Terror Scale. The PTS accounts only for violence committed by state actors and it is hypothesized that this type of violence would be more likely to be considered and accepted in refugee and asylum cases.

PAGE 34

29 The higher the rate of incidents that include non state actors i s hypothesized to have a negative impact on RAA as published by the Global Terrorism Database The GTD accounts only for terrorist activity that is not committed by governments and it is hypothesized that this type of violence would be less likely to be co nsidered and approved in refugee and asylum cases. Politics measured using the following variables: If the state of origin is a major source of illegal immigration, this is hypothesized to have a negative impact on RAA As the number of undocumented immigrants from a state increases, this will imply a lower number of refugees and asylees from that state as a response to domestic political concerns of mass migratio n or of a refugee crisis. discourse this is hypothesized to have a negative impact on RAA As the media gives more coverage to immigration from specific states, this would imply a lower number of refugees and asylees from that state in response to domestic political concerns.

PAGE 35

30 Hypotheses Considering the above variables that will be explored and the desired focus for the analysis of this study as Latin America, the most appropriate central hypotheses are as follows: Hypothesis One As the violence involving non state actors increases, the RAA in those regions will decrease. Hypothesis Two The more the state of origin is mentioned in political immigration discourse this is hypothesized to have a negative impact on RAA Figure 1 Explanatory Framework

PAGE 36

31 While the results pertaining to these hypotheses will certainly produce useful results, the results of the entire study will surely be more constructive in creating a more complete understanding of how these policies play out in reality. These principal hypotheses have been chosen to address both the nature of the violence many in Latin Americans are fleeing, and whether or not there is any c response to these circumstances. The independent variable of conflicts involving non state actors has been chosen to draw the distinction between cartel violence and other organized crime from st ate sponsored violence. The second independent variable, being a recurrent point of discussion in United States immigration politics, has been chosen to evaluate the relationship between domestic immigration politics and the outcomes of refugee and asylum cases. Data and Sample The data for the dependent variable, refugee and asylum admissions are available earlier editions appearing in their archives (US Department of Homeland Security 2015). Within this document, there is a table with a breakdown of immigrants admitted by class of admission and nationality. Refugees and Asylees combined are one of these classes. The entire population in this case would be every state and year since 1980, when the refugee admission ceilings program began. The sample chosen for this study will be each state within Latin America from 1980 forward. This sample has been chosen to avoid the pre 1980, pre ceiling refugee numbers from throwi ng off the data. This does limit the study as it excludes many conflicts in Latin America during the 1970s as well as major waves of Cuban migration after the Cuban Revolution. This sample is, however, the most appropriate sample

PAGE 37

32 for this study because it covers every year since the implementation of Refugee Admissions Ceilings. Data collection will occur by plotting out each case (state/year) then recording the refugee and asylee admits from each state/year. Data collection pertaining to the independen t variables will also be discussed after a clarification of the measurement of these variables. Measurement, Methodology, and Expected Results The measurement of the dependent variable in this study is straightforward. The concept of United States refuge e and asylum policy outcomes are operationalized through refugee and asylum admissions. Refugee and asylum admissions are a ratio variable as the value of zero means there would be no refugees or asylees allowed from a region. These values would fall be tween zero and the maximum admissions for that year. This measure is valid as these are the only two routes a refugee or asylee can take to obtain such status within the United States, therefore it covers those who enter to go through this legal process a nnually. which may require further attention. The primary issue here concerns the distinction between affirmative asylum and defensive asylum. Affirmative asylum i s the case applicants bring to court who are not facing deportation while defensive asylum pertains to those who are also facing deportation proceedings. As this measure only considers those admitted under this class of admission, it omits the cases in wh ich someone has already entered the United States illegally and afterwards applies for defensive asylum. Without including these cases, this could potentially threaten the validity of the study, especially considering the highly politicized nature of depor tation of undocumented Latinos in contemporary domestic politics. This is a particularly relevant concern with regards to the second hypothesis which

PAGE 38

33 relates directly to United States immigration politics. Within this sample however, there is no uniformi ty across the timeframe with the data available to parse out defensive and affirmative asylum. This is why the chosen variable best fits the study despite this obstacle. The first hypothesis states that as the conflicts involving non state actors increase the state of Terrorism and Responses to Terrorism 2015). The Global Terrorism Database provides a comprehensive data set for violence perpetrated by non state actors for each state within a region and is available throughout the entire sample. This fits the non state actor requirement as this data set applies the following definit ion of terrorism: "the threatened or actual use of illegal force and violence by a non state actor to attain a political, economic, Study of Terrorism and Respo nses to Terrorism 2015). This is an appropriate mechanism in this context because this data set accounts for drug cartel and other organized criminal activity, both of which are essential in any analysis of contemporary Latin America. The data for this variable will consist of non state actor terrorist events in each state for each year. This would also be a ratio variable with zero indicating zero terrorist incidents and increasing as the incidents increase. The testing of this first hypothesis cou ld also be taken further by juxtaposing the GTD with the Political Terror Scale (PTS). Data from these sources are also available throughout the entire timeframe of this study and are provided on a state by state basis. The PTS focuses primarily on state sponsored violence, which would provide as a useful comparison to those who endure non state violence. The PTS is an index which ranks countries ordinally

PAGE 39

34 from 1 to 5, with 5 being widespread and daily state sponsored violence (Gibney et al 2012). This is an ordinal variable and would provide where each state falls on the scale. It is also important to keep in mind that state perpetrated violence and violence by non state actors often happen in tandem and that these two variables are not mutually exclu sive. The second hypothesis tests the presence of a state as a part of United States political discourse on immigration causing the refugee admissions ceilings and percent of asylum applications granted to decrease. In order to effectively operationaliz e the independent variable here being a political immigration issue, it is most effective to analyze it through news media analysis. More specifically, through using the New York Times Article Archive the number of times each country was brought up in ar ticles pertaining to immigration can be easily determined. Furthermore the search engine available allows the narrowing down of d December 31, 1997 yields 181 results (New immigration politics. This allows invalid cases, such as results that discuss foreign immigration issues rather than do mestic, to be thrown out by viewing the content of each article. Again, this would be a ratio variable with zero indicating no mentions of that region in immigration discourse. Some may argue that this is not a valid representation of United States poli tics, but as the Times is the newspaper of record it is the most consistent and available method to collect data on salient political issues in domestic politics. Using the Times rather than presidential speeches or congressional hearings or bills is adva ntageous in that it speaks to what is salient enough in domestic politics to be published in a national

PAGE 40

35 paper, whereas political actors may avoid these issues intentionally as they may be thought of as a political minefield. Research Design and Discussio n This study is a large N, quantitative, non experimental research project. This is the refugee and asylum policy, which covers multitudes of cases, it is most fittin g to take into account as many cases as possible. The central research question of why does RAA vary annually is best answered by looking at the whole population, which is why a quantitative study is most the most appropriate approach. Upon collection of the data as described above, a statistical multivariate linear regression model will be employed to test these hypotheses along with alternative independent variables to account for alternative explanations. A more in depth analysis of these results, esp ecially with regards to what they imply for United States refugee and asylum policy towards Latin America will then be discussed at length. If the hypotheses are supported, the expected results would reveal a negative relationship with both independent variables the dependent variable and would imply that being a topic of discussion in immigration politics and/or having increased violence committed by non state actors would correlate with more restrictive outcomes in asylum and refugee cases. Much of the research in this particular area of study has been qualitative, and in many cases has come from legal scholarship (Aleinkoff 2003; Bauer 2003; Bonilla Mathe 2013; Estevez 2013; Longmire 2014; Rexton Kan 2011). While these studies certainly discussed the trends in general terms and used specific cases to support their arguments, there lacked

PAGE 41

36 any statistical analysis of the variables they discussed. This study will take into account many of the variables they brought up, only in a quantifiable fashion so as to provide a means for statistical analysis. As opposed to a legal, semantic, or theoretical critiques as others have done, this study aims to observe the relationships between policy outcomes and possible explanatory factors. The methodology in this proj ect is similar to that of Rosenblum and Salehyan (2004) as theirs was a quantitative assessment of the norms versus interests debate with regards to United States asylum outcomes. This research, however, is distinct from this study in that it focuses on t wo major factors that could potentially impact the ways in which the United States carries out refugee and asylum law towards Latin America specifically, that is domestic immigration politics and the spread of violence by non state actors in Latin America. Testing the hypotheses of this research project effectively analyzes United States Refugee and Asylum outcomes in Latin America by looking at two issues that pertain to Latin America refugee flows: violence committed by non state actors and United Stat es immigration politics. This study is limited in that large n quantitative studies like this often overlook substantive aspects that smaller, more qualitative studies may be able to capture. Taking that into consideration, however, a quantitative analy sis provides a more accurate portrayal of the trends taking place. An individual case may illuminate the possibilities within refugee law whereas a quantitative study can statistically analyze the relationship between two variables and determine if a corr elation exists. The strongest aspect of this research design is that it distinguishes between violence committed by state actor and that committed by non state actors. This is an essential

PAGE 42

37 distinction to make when focusing the analysis on Latin Americ a considering the cartel and gang violence this region has endured. Rosenblum and Salehyan (2004) used only the Political Terror Scale in their analysis which excludes violence committed by non state actors. By incorporating the Global Terrorism Database and including non state actors into the analysis, this study accounts for cartel and gang violence as a potential factor in this equation.

PAGE 43

38 CHAPTER IV RESULTS One of the first steps to complete the analysis was to set the data as panel data, combining state (appendix A) and year between 1980 and 2013. Then a summary was run on all of the data collected. This data set resulted in 1,155 cases total with very few cases of missing data. The exception here is the independent variable measur ing illegal immigration. Beyond the fact that this variable is based on estimation, there was very little available data on this variable. Using the Office of Yearbook Statistics, only 40 observations were available during the entire timeframe so unfortu nately this variable did not produce many results. For this reason, this variable has been left out of much of the analysis. The range of the dependent variable, refugee and asylee admissions, fell between 0 and 43,455 with a mean of 819.21. As predicte d both the higher end of this range and the mean are thrown off by Cuba. When Cuba is excluded, the range is between 0 and 25, 625 and the mean is 226.51. Cuba alone has a range of 3,885 and 43,455 and a mean of 18,402.61. Two other states that showed l arge numbers of admits were Haiti and Venezuela. ries are shown below in table 1. Summaries of the independent variables were also run (appendix B). The independent variables from the two hypotheses showed that fo r H1 measuring violence from non state actors, the annual terrorist incidents reported from the GTD fall between 0 and 710. As for H2 and domestic media attention, the New York Times articles that discussed the state in question and immigration ranged fro m 0 to 428 annually. As a final preliminary step, correlations were run with each independent variable and the dependent variable to examine the direction of the relationship for each. A correlation matrix was also run to investigate the

PAGE 44

39 direction of the relationship of all independent variables with RAA (appendix C). The relationships were all relatively weak with the strongest relationship at 0.539 between civil liberties and RAA. Once the preliminary overview was complete, I moved on to both bivariate and multivariate regressions which will lead to a much more in depth analysis Table One: Refugee and Asylee Admit Summary State Range Minimum Range Maximum Mean Total All states 0 43,455 819.21 Cuba 3,885 43,455 18,402.61 Mexico 6 605 159.81 Haiti 0 25,625 2,184.69 Colombia 13 12,591 1,403.3 Venezuela 3 2,497 517.18 El Salvador 0 1,249 304.2 Guatemala 2 1,559 341.96 Total excluding Cuba 0 25,625 226.51 Regression Analysis In order to provide the most complete analysis, I separately ran bivariate regressions with each independent variable ( Appendix E) then a multivariate regression with all of the variables include (Table 2). Table 2 reports the coefficient and P value for the multivariate regression. Two models were also initially used to analyze the data accounting for Cuba through use of a dummy variable. While this distinction made a difference in the coefficients, it did not change which variables were statistically significant. The illegal

PAGE 45

40 immigration variable has been excluded f rom these tests due to such minimally available data. As hypothesized, domestic media attention towards immigration from the state in question had a negative relationship with refugee and asylee admits. The freedom index was split with political rights sho wing a statistically significant negative relationship and civil liberties showing a positive relationship and no statistical significance. This is particularly interesting because not only do these two measures come from the same source, but they are hig hly correlated with one another (.873). This indicates that political rights have a negative statistically significant relationship to RAA while civil liberties do not. As political rights become more repressed RAA goes up. The other variables that turned up significant were trade and military aid though both of these turned out to have the opposite relationship to what was hypothesized. Whereas both of these had been hypothesized to have a negative impact on RAA, they both turned out positive. As trade goes up so does RAA. This is also true of United States military aid. The more aid that was allocated as military aid towards a state, RAA went up from that state also. This indicates that close economic relations and the United States providing militar y aid positively affect RAA rather than the negative reputation effect that had been hypothesized as the literature had discussed. Another interesting aspect of these results is that, as demonstrated earlier, Cuba is an extreme outlier, yet when Cuba was controlled for, the direction and the significance of each refugee and asylee admits are consistently higher than all other Latin American states, Cuba

PAGE 46

41 did not throw off the data as expected. The r squared measure does however jump from .0036 to .7676 which indicates that controlling for Cuba does create a much closer model fit. Table 2 Results : Model 1 whole data set Model 2 data set with added Cuba dummy variable N=663 N=663 R Sq=.0036 R Sq=.7676 Coef. P Value Coef. P Value Trade Trade balance with US by partnering state .0599* .002 .04* .004 Political Rights Freedom Index 237.1028* .083 277.82* .021 Civil Liberties Freedom Index 168.2124 .334 30.72 .82 Violence by state actors Political Terror Scale 97.27 .485 80.57 .467 Violence by non state actors Global Terrorism Database 1.3 .377 1.14 .325 Domestic Media New York Times articles 20.81* .000 14.94* .000 US military aid USAID Greenbook 7.99* .001 8.35* .000 Constant 835.56 .22 456.91 .083 indicates statistical significance where P<0.1 Accounting for Eras While the data the provided comprehensive results shown above, one thing that was lacking was any insight as to the impact of historical events. Between 1980 and 2013 there

PAGE 47

42 were several major events that could have influenced United States policy, and thr ee of these stand out as pivotal moments in United States history particularly regarding Latin America: the end of the Cold War, NAFTA, and 9/11. Considering this, the data was then organized into a few different models to investigate the influence of the se events under the following time constraints. The data was divided into the following categories: through and after 1989 to look at the possible im pact of the Cold War (table 3), before 1994 and 1994 onward to examine the possible influence of NAFTA (ta ble 4), and up through 2001 and after to look at the results pre and post 9/11 (table 5). The tables report the coefficient and P value for each. The results show that major events like this do in fact impact the data. A closer look at each model will re veal the degree to which each of these events may have impacted refugee and asylee outcomes. The End of the Cold War To analyze the possible impact of the Cold War on RAA, the data was split into two timeframes: 1980 1989 and 1990 2013. Once the timeframe has been narrowed down to 1980 through 1989, the only statistically significant variable is a negative relationship between political rights and RAA. After 1989 the results change, and the statistically significant variables become trade, media, and military aid. So pre 1989 the less politically free the state in question, the more likely refugees and asylees from tha t country are to increase. This indicates an impact of Cold War politics on refugee and asylee outcomes. Again this is not true of civil liberties which only further stresses the salience of the politics of the Cold War. After the Cold War was over other factors become more important such as domestic politics in the media and economic factors, such as trade and military aid. These

PAGE 48

43 differing results point to the policy shift that occurred as the Cold War came to an end and provides supporting evidence for how this shift affected RAA coming from Latin America. Table 3 Results : 1980 1989 1990 2013 N=80 N=583 R sq=.12 R sq=.01 Coefficient P value Coefficient P value Trade Trade balance with US by partnering state .18 P= .601 .0612* P=.003 Political Rights Freedom Index 770.20* P=.039 100.98 P=.511 Civil Liberties Freedom Index 504.03 P=.319 58.53 P=.771 Violence by state actors Political Terror Scale 377.89 P=.365 54.94 P=.711 Violence by non state actors Global Terrorism Database 2.74 P=.437 1.24 P=.471 Domestic Media New York Times articles .04 P=.998 20.72* P=.000 US military aid USAID Greenbook 3.43 P=.854 8.49* P=.001 Constant 943.29 .397 924.4 .219 *Indicates statistical significance where P<0.1

PAGE 49

44 NAFTA Another event that has been pivotal for United States Latin American relations was the passage of NAFTA as it deregulated trade barriers. While it was initially a trilateral agreement between the United States, Canada, and Mexico it marked the beginning of a shift towards the massive expansion of trade between the United States and Latin America and was eventually extended to include Guatemala, El Salvador, Honduras, Costa Rica, Nicaragua, and the Dominican Republic The process of splitting the data into two timeframes was also applied here: 1980 1993 and 1994 to 2013 to analyze the possible changes pre and post NAFTA. Unlike the previous example, it does not appear as though NAFTA made much of an impact on the outcome of RAA across the timeframe. Just as in the previous results before 1994 the only statistically significant factor is political rights. After 1994 this significance is lost. Instead, trade, domestic political media coverage, and United States military aid become statistically significant and the direction of the relationships remain consistent with all previous models. One interesting point of analysis is to look at these results and the deterrence through development concept that was previously discussed. According to this notion trad e should have a negative relationship with RAA not because the United States would accept fewer refugees from these places but because more opportunities become available and quality of life should improve. If that were the case, it should be especially ev ident after something like NAFTA that allows for the free flow of goods and services and more export processing zones. According to this logic and the pur ported outcomes of free trade, this relationship should be negative when in fact this relationship tu rned out to be positive. The more trade

PAGE 50

45 between the United States and another state, refugee and asylee admits go up. This implies that trade could contribute to push factors rather than creating more stability. Table 4 Results : 1980 1993 1994 2013 N=492 N=171 R sq=.0952 R sq .0013 Coefficient P value Coefficient P value Trade Trade balance with US by partnering state .06 P=.74 .06* P=.01 Political Rights Freedom Index 578.20* P=.007 154.07 P=.45 Civil Liberties Freedom Index 235.14 P=347 1.65 P=.99 Violence by state actors Political Terror Scale 161.13 P=.45 75.47 P=.67 Violence by non state actors Global Terrorism Database 1.16 P=.57 4.49 P=.31 Domestic Media New York Times articles 6.8 P=.594 22.03* P=.000 US military aid USAID Greenbook 1.94 P=.858 7.33* P=.014 Constant 647.04 P=.33 6.83 P=.427 *Indicates statistical significance where P<0.1 September 11, 2001 The final historical event selected to analyze as a possible impact on United States refugee and asylum policy towards Latin America was 9/11. As national security became a

PAGE 51

46 this priority is reflected in refugee and asylum ad mits. The timeframes selected were 1980 2001 and 2002 2013. The results here do show a distinction between pre and post 9/11. When this event splits the data, the results look very different. The earlie r model of 1980 2001 shows only the only statistically significant variable as a negative relationship with political rights. Post 9/11 the only statistically significant variable is a positive relationship with the Political Terror Scale which measures s tate sponsored violence as hypothesized. Trade, media, and military aid, which had all turned out to be significant in the latter half of each previous model, now all turn out not to be not statistically significant post 9/11. Another observation is tha t from 1980 2001, as in the case of all the earlier parts of the time based models, political rights again show to be statistically significant and that this significance goes away when looking forward from that year. A broader look at these results show a negative relationship with political rights any time the data include the earlier years whether it be pre 1989, pre 1994, or pre 2001. Taking this into account shows that the early Cold War year heavily weighs the data towards a negative relationship wi th political rights. The other variables mentioned trade, military aid, and media show up as significant in the whole data set as well as in the post 1989 and post 1994 models, yet are not significant in the post 9/11 model. During these middle years, tr ade, media, and military aid all impact RAA. Then in the post 9/11 era, state sponsored violence becomes the only statistically significant variable. Furthermore, this is not the case with non state actors although these two are positively correlated with one another. While the central hypothesis of non state actor violence having a negative relationship with RAA did not turn out to be statistically significant, the fact that state sponsored violence becomes the only significant variable is

PAGE 52

47 important. Jus t as political rights were prioritized during the Cold War, it seems as though after 9/11 another major shift has occurred and political violence rather than political rights are now impacting RAA. Table 5 Results : 1980 2001 2002 2013 N=362 N=301 R sq=.19 R sq=.12 Coefficient P value Coefficient P value Trade Trade balance with US by partnering state .009 P=.837 .022 P=.631 Political Rights Freedom Index 402.56* P=.006 169.65 P=.621 Civil Liberties Freedom Index 73.71 P=.682 29.009 P=.947 Violence by state actors Political Terror Scale 2.83 .986 483.08* P=.049 Violence by non state actors Global Terrorism Database .019 P=.989 25.14 P=.115 Domestic Media New York Times articles 2.04 P=.642 3.61 P=.674 US military aid USAID Greenbook 1.75 P=.698 9.56 P=.843 _cons 533.07 P=.321 26.17 P=.983 *Indicates statistical significance where P<0.1

PAGE 53

48 Conclusion An overview of these results produce some significant takeaways. First while Cuba was a major outlier, it did not impact the data as expected. However, Cuba being a Soviet ally in the Western Hemisphere is a major part of what makes Cuba an outlier, and the Cold War years included in the data from 1980 1989 do affect the data. All of the models that include these years show political rights to be positively related at a significant level and anytime these years are excluded, this significance goes away. Civil liberties which is highly correlated with political rights and is taken from the same source (the Freedom Index) never results as being statistically significant. This highlights the heavy influence of Cold War politics over almost any other variable. Another interesting point is the relationship between trade, military aid, and media with RAA. All of these are significant in the whole dataset and from 1990 forward and 1994 forward. Trade and military aid both resulted as having positive relationships with RAA. Rather than the idea that the United States will not take refugees from states with close economic ties, the opposite has resulted. While it may be that the United States is taking more refugees from states it is economically tied to, it may a lso be the case that increased or deregulated trade and military aid have a destabilizing effect and create push factors. Further research could investigate that as a possibility. The decision to exclude illegal immigration data due to lack of availabil ity is an aspect of the study that could have drastically altered the results. Had this information been available it would have provided more insight into the role of this variable on RAA. This could possibly be investigated in future research by focusi ng on the states that are primary sources of illegal immigration and their asylum approval rates.

PAGE 54

49 Finally the post 9/11 shift towards political violence as the only statistically significant variable show another shift in United States policy. This shift shows that the current US foreign policy with its focus on securitization has a potential impa ct on refugee and asylum policy This too could be a future avenue for research especially with the rise of new security threats and refugee crises. Another application of this research would be to investigate how these results compare to other regions of the world. Considering the unique characteristics of Latin America such as proximity and high levels of undocumented immigrants there could be many differen ces. There are also variables that play out very different in other regions such as non state actor violence. It would be very interesting to apply this research to other regions to see what ki n d of results these variables produce.

PAGE 55

50 BIBLIOGR APHY Aleinikoff, Alexander. 2003. "Protected characteristics and social perceptions: an analysis Refugee Protection in International Law Cambridge University Press Accessed at http://www.refworld.org/docid/470a33b30.html Angotti, Thomas 1996 "Latin American urbanization and planning: inequality and unsustainability in North and South." Latin American Perspectives ) 12 34. Barbera Hernandez, Lila 2005. "Indigenous peoples, human rights and natural resource d evelopment: Chile's Mapuche people and the right to water Annual Survey of International & Comparative Law : 11 1. Ba uer, Laura Isabel. 2003. "They beg our protection and we refuse: US asylum law's failure to protect many of today's r efugees." Notre Dame Law Review 79: 1081. Bonilla Math e, Isabel. 2013. "To group or not to group; refugee status for Central American victims of gang v iolence." Tulane Journal of International & Comparative Law 22: 145. International Business Times. Retrieved from http://www.ibtimes.com/why arent central americans getting asylum us 1653494 Charlton, Roger, Lawrence T. Farley, and Ronald Kaye 1988. "Identifying the mainsprings of US refugee and asylum policy: A contextual interpretation." Journal of Refugee Stud ies 1, no. 3 4 : 237 259. COHA June 24th 2010 aiti and Cuba accessed at:http://www.coha.org/disparities in u s immigration policy toward haiti and cuba a legacy to be continued/ Displaced Persons Act. 1948. Public Law 82 60. United States Statutes at Large. Volume 65 Stat. 96 Estvez, Ariadna. 2013. "The biopolitics of asylum l aw in Te xas: The case of Mexicans fleeing drug v iolence in Jurez." Norteamrica, Revista Acadmica del CISAN UNAM 8. Gibney, M., Cornett, L., Wood, R., & Haschke, P., Political Terror Scale 1976 2012. Date Retrieved, from the Political Terror Scale Web site: http://www.politicalterrorscale.org/ Haass, Richard N. 1998. Economic Sanctions and American Diplomacy Council on Foreign Relations

PAGE 56

51 Mexican an d Central American asylum and credible fear claims: background and c ontex Accessed at: http://www.immigrationpolicy.org/ Kornbluh, Peter 2003 The Pinochet File: A Declassified Dossier on Atrocity and Accountability New York: The New Press. p. 28. ISBN 1 56584 936 1 Lindsay Poland, Joh n 2003 Emperors in the Jungle: The Hidden History o f the U.S. in Panama Duke University Press. p. 118. Longmire, Sylvia. June 26, 2014. "Should Migrants Fleeing Gang Violence in Central America Be Accorded Refugee Status ?." Small Wars Journal Accessed at: http://www.isn.ethz.ch/Digital Library/Articles/Detail/?id=181403 Immigration Statistics. McBride, Michael J 1999 "Migrants and asylum seekers: policy responses in the United States to immigrants and refugees from Central America and the Caribbean." International Migration 37 1: 289 317. Migration Policy Institute, 2015. Accessed a t: http://www.migrationpolicy.org/ Migration and Refugee Assistance Act. Public Law 87 510. 1962 United States Statutes at Large. Volume 76 stat 121. National Consortium for the Study of Terrorism and Responses to Terrorism (START). (2013). Global Terrorism Database [Data file]. Retrieved from http://www.start.umd.edu/gtd New York Times. New York Times Article Archive 1981 to present. http://www.nytimes.com/ref/membercenter/nytarc hive.html Accessed April 17th, 2015 http://www.refugeec ouncil.org.uk/ Rex ton Kan, Paul 2011. "Mexico's' narco r ef ugees': the looming challenge for US national s ecurity." The Strategic Studies Institute The US Army War College. Rosenblum, Marc R., and Idean Salehyan. 2004. "Norms and interests in US asylum enf orcement." Journal of Peace Research 41, no. 6: 677 697. Finnemore, Martha, and Kathryn Sikkink. 1998 "International norm dynamics and political change." International organization 52 04: 887 917.

PAGE 57

52 Ibrahim, Maggie 2005 "The Securitization of Migration: A Racial Discourse1." International migration 43 5 : 163 187. Kay, Cristbal 2001 "Reflections on rural violence in Latin America." Third World Quarterly 22 5 : 741 775. Lee, Everett S. 1966 "A theory of migration." Demography 3 1 : 47 57. Schoultz, Lars 1998 Beneath the United States Harvard University Press Suhrke, Astri. 1994. "Towards a comprehensive refugee policy: Conflict and refugees in the post cold war world." in Aid in place of migration 13 38. eds W.R. Bohning and M .L.Schlueter Pareyes. International Labor Office. http://www.uscis.gov/humanitarian/refugees asylum/refugees Refugees and Asylees. Data Tables. http://www.dhs.gov/yearbook immigration statistics Accessed April 13th, 2015. http://www.justice.gov/eoir/efoia/foiafreq.htm Accessed April 13t h, 2015. United States Refugee Act of 1980. Public Law 96 212. United States Statutes at Large. Volume 94 stat 102. Zolberg, Aristide R., Astri Suhrke, and Sergio Aguayo. 1992. Escape from violence: Conflict and the refugee crisis in the developing world New York: Oxford University Press. Zucker, Norman L., and Naomi Flink Zucker. 199 2. "From Immigration to Refugee Redefinition: A history of refugee and asylum policy in the United States." Journal of Policy History 4, no. 01: 54 70.

PAGE 58

53 APPENDIX A List of States in Data Set Antigua/Barbuda, Argentina, Bahamas, Barbados, Belize, Bolivia, Brazil, Cayman Islands, Chile, Colombia, Costa Rica, Cuba, Dominica, Dominican Republic, Ecuador, El Salvador, Granada, Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexi co, Nicaragua, Panama, Paraguay, Peru, Saint Kitts and Nevis, Saint Lucia, Saint Vincents and Grenadines, Suriname, Trinidad and Tobago, Uruguay, Venezuela.

PAGE 59

54 APPENDIX B Sum of All Data

PAGE 60

55 APPENDIX C Correlation Matrix