Artículos

 

Risk of Death and Aggressions Encountered while Illegally Crossing the U.S.-Mexico Border

 

Riesgo de muerte y agresiones al cruzar ilegalmente la frontera Estados Unidos-México

 

Pedro Paulo Orraca Romano* y Francisco de Jesús Corona Villavicencio**

 

*University of Sussex. Dirección electrónica: p.orraca-romano@sussex.ac.uk.

**Centro de Investigación en Matemáticas, unidad Monterrey. Dirección electrónica: fcorona@est-econ.uc3m.es.

 

Date of receipt: August 31, 2012.
Date of acceptance: April 30, 2013.

 

Abstract

This article analyzes the type of migrant who is most likely to be exposed to the risk of death or be a victim of aggression by U.S. authorities while illegally crossing the U.S.-Mexico border, and the impact this has on the future migration intentions of repatriated or deported individuals. On the basis of a logistic regression analysis, the study shows that those most vulnerable to risk or aggressions are first-time migrants who hire the services of human smugglers. Furthermore, migrants who were previously exposed to the risk of death or whose belongings were confiscated and never returned are less likely to engage in subsequent migration attempts. Finally, being a victim of physical or verbal abuse at the hands of U.S. Border Patrol agents is unrelated to future migration intentions.

Keywords: illegal migration, migrant deaths, aggression, Mexico, United States.

 

Resumen

Este artículo estudia qué tipo de migrantes tienen mayores probabilidades de enfrentarse al riesgo de muerte y ser víctimas de agresiones por parte de autoridades estadounidenses al cruzar ilegalmente la frontera estados unidos-México, y el impacto que esto tiene sobre las intenciones migratorias futuras de los individuos repatriados o deportados. Con base en regresiones logísticas, se muestra que las personas más vulnerables son aquellas que migran por primera vez y contrataron los servicios de un coyote. Así mismo, aquellos migrantes previamente expuestos al riesgo de muerte o cuyas pertenencias fueron confiscadas y nunca devueltas tienen menor probabilidad de incurrir en subsecuentes intentos migratorios. Por último, ser víctima de abuso físico o verbal por parte de agentes de la Patrulla Fronteriza no está relacionado con las intenciones migratorias futuras.

Palabras clave: migración ilegal, muerte de migrantes, agresión, México, Estados Unidos.

 

Introduction

Unauthorized migration and the hardships encountered by individuals who illegally attempt to enter the U.S. across the Mexican border has been a topic of intensive debate and political discussion in recent years. As migrants journey to the U.S. to improve their quality of life and compensate for failed or non-existent markets in Mexico, U.S. policy-makers attempt to protect their boundaries and domestic labour market from unwanted visitors, by implementing policies that endanger the lives and well-being of illegal immigrants.

As a result of the rapid increase in the number of unauthorized migrants entering the U.S. in the 1980s, border control efforts were intensified following the enactment of the Immigration Reform and Control Act (IRCA) in 1986 (USCIS, 1986), increasing the difficulties associated with unlawful migration. The subsequent enactment of the Illegal Immigration Reform and Immigrant Responsibility Act (IIRIRA) in 1996 (USCIS, 1996) broadened the powers of the Immigration and Naturalization Service (INS), granting it wide-ranging authority to arrest, detain and remove immigrants, and generating unprecedented focus on border enforcement issues (Phillips, Hagan, and Rodriguez, 2006). The enactment of IRCA and IIRIRA, together with the implementation of preventive measures by the U.S. Border Patrol (USBP) designed to deter unauthorized migrants from crossing the border, made undocumented migration progressively more hazardous, forcing migrants to travel through isolated, hazardous terrains when attempting to enter the united States. As a result, the number of undocumented border-crosser deaths along the U.S.-Mexico border increased, where it is estimated by the USBP to have totalled 5 570 between 1998 and 2012 (USCBP, 2013).¹

The intensification of border enforcement efforts has been accompanied by an increase in the number of USBP agents, which stood at 4 028 in 1993 and had risen to 21 165 by 2012. The rapid surge in the number of USBP and Immigration and Customs Enforcement (ICE) agents has led to concerns about their training, since numerous cases of physical or verbal abuse against unauthorized migrants have been documented (OIG, 2006; Slack et al., 2013). Furthermore, studies have shown that migrants commonly report that while they were detained, their belongings, including money, mobile phones and identification documents, were confiscated and never returned (Danielson, 2013; Slack et al., 2013). This creates an additional set of problems for migrants, since after they are returned to Mexico they are unable to access certain types of jobs, receive money transfers, or buy bus tickets to their homes (Slack et al., 2013:24).

Due to the increasing perils unauthorized migrants encounter while illegally crossing the U.S.-Mexico border or on being detained by U.S. authorities, a study of the individuals that are exposed to these difficulties and their relationship with future migration intentions warrants attention. Specifically, the research questions we attempt to answer are: "What type of migrant is more likely to be exposed to the risk of death while illegally crossing the U.S.-Mexico border or be a victim of aggressions at the hands of USBP agents when apprehended?", and "what impact do previous exposure to the risk of death or aggressions encountered when detained have on the future migration intentions of previously repatriated or deported individuals?"

 

Border Enforcement, Unauthorized Migration and Exposure to Risk and Aggression

During the early 1990s, the U.S. Border Patrol adopted the "prevention through deterrence" strategy, altering its previous approach, which had involved apprehending migrants and subsequently returning them, to one that discouraged people from migrating altogether (Cornelius, 2001; Feldmann and Durand, 2008).² As part of this strategy, several preventive measures were carried out including Operation Hold-the-line (or Operation Blockade) in El Paso in 1993, Operation Gatekeeper in San Diego and Operation Safeguard in Nogales in 1994, and Operation Rio Grande in the Border Patrol's McAllen (currently Rio Grande Valley) sector in 1997. These similar measures strengthened patrol efforts in urban areas, thus channelling unauthorized migrants into more inconspicuous and less political sensitive regions (Cortés, 2003; Feldmann and Durand, 2008). Consequently, undocumented migrants were forced to alter their entry routes from the traditional Tijuana-San Diego and Ciudad Juarez-El Paso corridors, to more inherently treacherous crossing zones such as the Sasabe-Arizona Desert, the All-American Canal, or the Rio Grande (Eschbach et al., 1999; Cornelius, 2001; Cortés, 2003). This created a "funnel effect", which increased the flows of migrants travelling through hazardous and isolated areas and ultimately raised the number of migrant deaths (Cornelius, 2001; USGAO, 2006; Rubio-Goldsmith et al., 2006).

Efforts by the U.S. government to reduce the unauthorized entry of migrants have also been reflected in the continued growth of the enacted USBP program budget, which rose from 262.65 million dollars in 1990 to 3.53 billion dollars in 2012 (USCBP, 2013). By 2012, throughout the 1 969 mile U.S-Mexico border, close to 700 miles of vehicle and pedestrian fencing had been constructed, along with the installation of electronic sensing devices and aerial drones which significantly reduced the likelihood of evading capture (Alden, 2012:107).³

Altogether, these measures have been implemented with the purpose of deterring unauthorized migration by raising its physical and monetary costs, increasing the probability of detention, and lowering its expected benefits. In addition, the terrorist attacks of 9/11 heightened border enforcement efforts, prompting the creation of the Department of Homeland Security (DHS) and altering the view of the problem of clandestine border crossing to one of international security.

Despite increased border enforcement efforts, it is estimated that from 2005 to 2012, an average of approximately 570 000 Mexican-born migrants attempted to enter the U.S. each year. Graph 1 shows that among these individuals, 71.81 percent did not have the required documentation to lawfully enter the country. Nonetheless, there has been a significant drop in the number of migrants attempting to cross the U.S.-Mexico border in recent years, particularly from 2008 onwards.⁴ While the reduction in undocumented migration flows may suggest that the border enforcement strategy has successfully deterred clandestine immigration, previous research on the success of this policy suggests that its effect has been limited (Espenshade, 1994; Sorensen and Carrión-Flores, 2007), given that these flows are mainly driven by earnings differentials. Furthermore, it has been stated that the downturn of the U.S. economy and the high unemployment rates in sectors that traditionally employ a large proportion of legal and illegal migrants seem to be the main factors behind the reduction in the Mexico-U.S. migration flow (Alonso, 2012:229).

The number of deaths among migrants attempting to cross the border increased parallel to the decline in undocumented migration flows. According to the USBP, the number of migrant deaths rose steadily from 380 in 2000 to 463 in 2013, averaging 419 deaths per year between 2005 and 2012 (USCBP, 2013). The rise in the number of clandestine migrant deaths along the border has attracted the notice of researchers at the Centers for Disease Control and Prevention (CDC), who conclude that this phenomenon has developed into a major public health issue (Rubio-Goldsmith et al., 2006). Moreover, it is estimated that for every 100 000 detentions by the USBP, the number of bodies recovered along the deadliest parts of the border, such as the Sonoran Desert, was one and a half times higher in 2009 than in 2004, and 17 times greater than it had been in 1998 (Jimenez, 2009).

Due in large part to the substantial degree of attention that undocumented border-crosser deaths have received, the Mexican and U.S. governments have implemented a series of measures designed to reduce the risks and dangers migrants are exposed to when attempting to cross the border. These include posting warning signs at high-risk crossing points, distributing fliers in Mexican border towns, increasing the presence of search and rescue teams in hazardous areas on both sides of the border (Nevins, 2003:171), and the implementation of the Border Safety Initiative (BSI), the Arizona Border Control Initiative, and the Interior Repatriation Program, among others (USGAO, 2006).

In addition to the large number of migrant deaths, many of the individuals who manage to enter the U.S. encounter life-threatening conditions or instances of abuse, producing long-term physical and psychological scars (Marroni and Alonso, 2006). Graph 2 shows that it is not unlikely for migrants to be exposed to the risk of death when attempting to cross the U.S.-Mexico border. These risks are mainly induced by extreme temperatures, starvation, dehydration, exhaustion and fear of drowning. Moreover, a significant fraction of previously arrested migrants report that, when they were detained by USBP agents, they were subjected to different types of aggression including physical abuse, verbal abuse, or a permanent confiscation of their belongings. Given the significant number of migrants that are detained and subsequently repatriated or deported, this implies that tens of thousands of individuals encounter these harmful and potentially life-threatening experiences each year.

Infante et al. (2012:450) argue that illegal migrants are vulnerable to physical, sexual, and verbal abuse by members of the local population, authorities, and organized crime groups, and are commonly subjected to various displays of discrimination such as racism, xenophobia, abuse and violence.⁵ It has been documented that in multiple cases USBP agents and ICE personnel have committed human rights violations of detained migrants (OIG, 2006; nmd, 2008). These actions generally consist of ill-treatment and disregard towards detainees, insufficient or untimely provision of health care, physical aggression, verbal abuse, and safety and environmental health concerns.

Further examples include the 20 cases of lethal force by USBP agents from 2010 to 2012 reported by the Southern Border Communities Coalition (Soboco, 2013); the sexual assault of a female detainee in San Diego (OIG, 2006:28); and the hundreds of cases reported by the No More Deaths organization, including the case of a male migrant who was kicked in the stomach by an USBP agent, not given medical care, and on his return to Mexico was recommended for surgery given that he had swelling in his genitals and blood in his urine, indications of abdominal and testicular injury (nmd, 2008:17). Additionally, although there is an agreement between Mexico and U.S. officials to restrict the use of lethal fire-power on migrants, this remains an ever-present factor, the usual justification being that they are used in "self-defence" (Feldmann and Durand, 2008). Lastly, actions taken by local population members include manhunts for undocumented immigrants organized by xenophobic cattle ranchers in Arizona, in which they commonly shoot migrants inside their ranches, or allegedly "act in self-defence", since they feel "threatened" by the presence of intruders on their properties (Alonso, 2005:123).

Spener (2008) attributes the acts of violence committed by U.S. authorities on unauthorized migrants to the conflictive nature of their encounters, the nationalistic and racist attitudes of the agents, and the cultural differences between those involved. Nevertheless, not all detained immigrants report negative experiences with the USBP, with many of them expressing gratitude after being rescued by USBP agents (Slack et al., 2013).

 

Literature Review

Previous evidence suggests that the decision to illicitly migrate from Mexico to the U.S. is a function of several factors, including the wage differential between the two countries, the costs and fees incurred during the journey to the U.S., and access to migration networks, among others. Given that it has now been well-documented that heightened border enforcement as of 1993 led to a substantial increase in the total quantity of undocumented border-crosser deaths (Cornelius, 2001; Eschbach, Hagan, and Rodríguez, 2003), a growing body of literature has focused on migrant deaths along the U.S.-Mexico border.

Among these studies, Rubio-Goldsmith et al. (2006) examine autopsy reports on illegal migrants from the Medical Examiner's Office of Pima County. The authors show that during the period of study, there was an exponential rise in the number of migrant bodies held by the Medical Examiner's Office, from nine in 1990 to 201 in 2005. As a result, the Border Patrol's Tucson sector is now the place where the majority of migrant deaths are recorded. In addition, an escalation in the number of migrant deaths caused by heat exposure has been observed, together with a reduction in the number of deaths due to all other causes. In a comprehensive study, Sapkota et al. (2006) use medical examiners' data from 2002 and 2003 to investigate the characteristics of undocumented migrants who perished while attempting to enter the U.S. from Mexico. Focusing on 409 deaths registered in different counties in Arizona, New Mexico and Texas, their study shows that 72.8 percent of the casualties were men, 66.1 percent were between the ages of 20 and 39 years, and that the most common cause of death was heat exposure, accounting for 43.4 percent of victims. Furthermore, in an investigation undertaken by the U.S. Government Accountability Office (USGAO, 2006), based on information from BSI reports, the authors observe an upward trend in the number of migrant deaths, which rose from 266 in 1998 to 472 in 2005. This is in stark contrast to the figures registered from the late 1980s to the early 1990s, when there was a decline in the number of undocumented border-crosser deaths (USGAO, 2006:14).

With respect to the causes of death, Eschbach, Hagan, and Rodríguez (2003) observe that from 1985 to 2002, the leading causes of death among migrants were drowning and motor vehicle accidents, whereas after 1993, there was an upsurge in the proportion of deaths induced by environmental heat, cold exposure and dehydration. Jimenez (2009) reports similar results where, based on information from the USBP, 30.25 percent of the 390 deaths reported in the fiscal year of 2008 were the result of environmental heat exposure, while 13.84 percent were water-related and 11.28 percent were linked to motor-vehicle accidents. On the other hand, Santibáñez (2000), focusing on exposure to the risk of death and based on data on detained migrants for 1999 and 2000, states that approximately 70 percent of migrants report being exposed to some type of physical risk, in other words, dehydration, starvation or extreme temperatures, when trying to enter the U.S. illegally.⁶

Even though the main violent act migrants are exposed to is death (Alonso, 2005:122), they also face the risk of encountering abuses. Based on information from the Migrant Border Crossing Study, Slack et al. (2013) state that one in every ten migrants reports being the victim of some form of physical abuse during his or hers most recent detention while one in four reports being subjected to verbal abuse at the hands of U.S. authorities. Moreover, the authors observe that 12 percent of respondents report that they were robbed by bandits during their last crossing, while seven percent report that they were kidnapped, and 39 percent state that their possessions were confiscated and never returned. On the other hand, based on information collected at the Centro de Atención a Migrantes Deportados in Nogales, Sonora, Daniel-son (2013) observes that 24.8 percent of returned migrants report that they were abused in some capacity by USBP agents. Compared to the local police in Mexico and criminals, the USBP was the most common perpetrator of abuse against migrants (Danielson, 2013:8).

While it would be unwise to suggest that previous exposure to risk and aggression are the determining factors driving migrants' future migration intentions, as argued by Cornelius and Salehyan (2007:143), theoretically a significant proportion of individuals on the ''cusp'' of migrating are expected to be persuaded by the traumatic effects of the encounter with these adverse factors to return back home. Nevertheless, for some migrants, returning home is not an option, since they have to work in the U.S. to earn the money needed to repay the amount they borrowed in order to migrate.

Investigating the link between risk perception and migration intentions, Cornelius and Salehyan (2007) seek to determine whether the perception of the physical danger involved in illegally crossing the border is a sufficiently strong deterrent to influence the likelihood of migrating. Based on 603 interviews of return-migrants and potential first-time migrants in two rural communities in the states of Jalisco and Zacatecas, the scholars conclude that migrants' strategies were indeed affected by increases in border security. Changes in strategy include the selection of different border-crossing points, a higher propensity to hire smugglers, and an increase in the probability of trying enter the U.S. through legal ports of entry. Nevertheless, very few migrants appear to have been permanently persuaded to stay home. In a similar study, using data from the Mexican Migration Field Research Program, Cornelius et al. (2008) analyze the success and unintended consequences of the U.S. border enforcement strategy. Relying on information collected in Oaxaca and San Diego in 2007-2008, the authors analyze whether the perceptions of the risk involved in clandestine border-crossing or knowing someone who died while illegally entering the U.S. are useful predictors of whether an individual will migrate. The study shows that the respondent's perception of border-crossing difficulties and the dangers encountered when attempting to enter the U.S. do not have a significant effect on the intent to migrate.

Lastly, focusing on the experiences of previously detained migrants, DeLuca, McEwen, and Keim (2010) conduct a qualitative investigation in which they analyze how migrants receive information regarding the risks involved when crossing the border, how this information affects their strategy to gain entry into the U.S., and how the journey impacts their future migration intentions. The authors perform eight in-depth interviews on men who had attempted to enter the U.S. illegally and had recently been returned to Mexico. It is observed that migrants obtain information from the media, family members, and friends, and identify possible risks such as travelling with people who have health problems. Furthermore, among the eight men interviewed, five reported that they intended to cross the border at a later date. Those that did not intend to cross explained that this was because the risks associated with crossing the border were too great.

 

Data

The data used in this study is taken from the "Encuesta sobre migración en la frontera norte de México" (Emif Norte)—Survey on Migration in the Northern Border of Mexico—, a cross-sectional survey conducted by El Colegio de la Frontera Norte (El Colef et al.), similar in design to the UK "International Passenger Survey" (Office for National Statistics, 2014). The Emif Norte is conducted within Mexico, primarily in the border region, where the travellers interviewed are divided into visitors and migrants, and individuals are interviewed in bus stations, train stations, airports, international bridges, ports of entry and Mexican customs inspection points.⁷ The survey provides information on migrants who return to Mexico either voluntarily or because they were sent back to their native country by U.S. authorities. The Emif Norte collects detailed data on the migrant's socioeconomic characteristics and his/hers most recent trip to the United States. The survey also provides information on the respondent's previous migration experience and future migration intentions.

Nevertheless, due to the structure of the survey, it is only possible to capture the risks and aggressions encountered by unauthorized migrants who had previously been detained and subsequently returned back to Mexico.⁸ Insofar as returned individuals differ from all other illegal migrants, the interpretation of the findings must be limited to previously detained individuals, in which case the results cannot be generalized to all unauthorized migrants. It is therefore necessary to determine whether detained migrants constitute a representative sample of all the different cohorts of illegal migrants included in the Emif Norte.

Table 1 presents the means of selected variables of potential, repatriated or deported, and returning unauthorized migrants. Relative to potential and returning migrants, individuals that have been repatriated are slightly older, less likely to be male or a household head, and less prone to hire the services of human smugglers. Furthermore, they tend to have more migration experience and higher educational attainment than potential migrants, while the opposite is true for returning migrants. A t-test on the means of the continuous variables, in other words, age, years of schooling, and fee paid to the smuggler, shows that between potential migrants and individuals that have been repatriated, there is a statistically significant difference in age and number of years of schooling. Among returning migrants and repatriated or deported individuals, the only statistically significant difference is in smuggling fees. Given the mostly subtle differences between the three groups, this sample can be said to be representative of all unauthorized migrants surveyed in the Emif Norte.

Lastly, the sample comprises Mexican-born individuals between the ages of 18 and 65 that live in Mexico, do not have the required documentation to legally enter the U.S., and were previously detained by USBP agents. Furthermore, since the variables of interest were only included in the Emif Norte for the period from 2005 to 2010, the study is limited to pooled data from this period. The total number of respondents with no missing values rises to 31 213 observations, of which 26 486 are male and 4 727 female.

 

Analysis

The first part of the study examines the type of migrant who is most likely to be exposed to the risk of death while illegally crossing the U.S.-Mexico border, and is most likely to be a victim of aggressions at the hands of USBP agents when apprehended. The dependent variables are self-reported exposure to the risk of death while crossing the border during the most recent migration attempt and aggression encountered when detained by USBP agents during the most recent migration attempt. The second part of the study analyzes the impact of the exogenous covariates of exposure to risk and aggressions encountered on the future migration intentions of previously detained migrants. In this case, the dependent variable indicates whether the respondent intends to migrate to the U.S. at a later date.

It is difficult to know a priori exactly what type of migrants are most vulnerable. Nonetheless, if migrants have been negatively affected by previous life threatening conditions encountered while illicitly entering the U.S., or by aggressions from USBP agents when apprehended, then ceteris paribus migrants who have been exposed to risk or have been the victims of aggression will be less likely to attempt to enter the U.S. illegally at a later date, compared with those whose life was not in jeopardy or who did not suffer some type of abuse or aggression.

Due to the empirical nature of the study, a maximum likelihood estimation of a binomial logistic regression model with robust standard errors is undertaken, for which a series of different econometric specifications are defined. These include a general model, where risk and aggression are introduced in a broad form, and a detailed specification, where exogenous variables are disaggregated into different types of risks and aggressions.⁹ In keeping with the Mexico-U.S. migration literature, a series of controls are inserted and presented in table 2. Age and age squared are introduced in order to capture the non-linear relationship between the covariate and the likelihood of migrating; education is incorporated since numerous studies have shown that this is an important predictor of the migration decision, and a series of dummy variables are inserted in order to control for differences in the characteristics of migrants and external factors that may affect migration intentions. Quarter and year dummies are introduced, since illegal migrants are highly sensitive to business cycle conditions and relative wages in the U.S. and Mexico.

Table 3 presents a summary of the statistics, showing that a large proportion of respondents who crossed the border alone do not intend to migrate to the U.S. at a later date. Moreover, sampled individuals tend to be quite young, with an average age of 28.66 years. There is a difference of approximately one year between the mean ages of men and women, and between individuals who do or do not intend to enter the U.S. illegally at a later date. Additionally, education attainment is low, with 50.40 percent of men and 45.46 percent of women having fewer than nine years of schooling.

Regarding exposure to the risk of death, 23.64 percent of migrants state that they put their life in danger when attempting to enter the United States. Nonetheless, the raw data does not show any significant differences between the levels of exposure to risk encountered by men and women, or by those individuals who do or do not intend to cross at a later date. For all sampled migrants, the most common types of risks encountered are drowning, exposure to extreme temperatures, and dehydration. Moreover, 14.62 percent of respondents report having been subjected to some form of abuse by USBP agents, where a similar frequency of men and women state that they were victims of these harmfulacts. Among the most common forms of aggression, verbal abuse was experienced by 12.02 percent of the sample, while 4.76 percent of migrants report having suffered physical aggression, and 3.17 percent claim that their belongings were confiscated and never returned. Moreover, migrants that do not plan to illegally enter the U.S. at a later date are more likely to have been victims of aggressions than those that do intend to cross.¹⁰

A surprising characteristic is that the individuals included in the analysis are mostly first-time migrants, where only 27.80 percent of males and 13.64 percent of females report having previously migrated. This reflects the absence of migration specific human capital since respondents lack prior migration experience (Solís and Alonso, 2009:259). The high cost of international migration is reflected in the fact that 68.31 percent of respondents state that they were forced to borrow money in order to undertake their most recent trip to the United States. This is not unexpected given their low earnings in Mexico, the high transportation costs involved, and the rising prices charged by smugglers. Finally, a latent issue in the econometric exercise is that it is not possible to measure the migrant's expected earnings in the U.S., leading to a potential problem of omitted variable bias. In order to minimize this issue, additional controls are introduced such as the migrant's education attainment and age which, in addition to having an effect of their own, are highly correlated with expected income in the U.S. (Cornelius and Salehyan, 2007:146).

 

Results

Exposure to the Risk of Death and Aggression Encountered when Apprehended

Table 4 presents the marginal and impact effects of the logistic regression models that predict exposure to the risk of death while unlawfully crossing the U.S.-Mexico border. The results show that for both men and women there is a concave relationship between age and exposure to risk, although for men the quadratic term is not significant. Moreover, male migrants who previously entered the U.S. illegally are, on average and ceteris paribus, 3.51 percentage points less likely to encounter risk than respondents with no prior migration experience. This suggests that during their trips to the U.S., migrants acquire knowledge that helps them evade more dangerous border-crossing points in the future. Moreover, married migrants are 0.022 probability points less likely to encounter risk than unmarried respondents, which may reflect the former's desire to travel through less dangerous routes. Additionally, migrants who cross the border alone have a lower likelihood of being exposed to the risk of death, a relationship that is significant for both men and women.

Furthermore, male migrants that have previously been repatriated or deported are 4.52 percentage points more likely to be exposed to risk compared to migrants undertaking their first trip to the United States. This result may indicate that previously detained migrants are more willing to travel through particularly inauspicious areas or use riskier methods in order to avoid being captured by USBP agents since repeat-offenders tend to face harsher sanctions, particularly since the implementation of Operation Streamline (Lydgate, 2010). Furthermore, it is observed that males who report borrowing money in order to travel to the U.S. have a higher likelihood of having been exposed to risk compared to male respondents that did not borrow money. This result suggests that financial constraints play an important role in the risks faced by migrants. Regarding the risks posed by relying on human smugglers, males and females who hired their services are 4.19 and 9.42 percentage points, respectively, more likely to have been exposed to the risk of death than those who did not hire a coyote. Finally, with respect to Tijuana, the most dangerous crossings zones were Matamoros and Nuevo Laredo for men, and Ciudad Juárez and Agua Prieta for women.

In regard to migrants that encounter abuse by USBP agents or whose belongings were confiscated and never returned, table 5 shows that for both men and women there is a positive, decreasing relationship between age and the likelihood of encountering aggression. For men, previous experience in entering the U.S. without authorization reduces the likelihood of having faced some type of aggression by 4.97 percentage points, while having previously been returned increases it by 5.48 percentage points. Furthermore, people that choose to cross the border alone or contract the services of a coyote are more likely to have encountered some type of aggression, suggesting that individuals that do not cross alone and do not hire a human smuggler are less likely to be victims of abuse by U.S. authorities or have their belongings permanently confiscated. In general, the probability of encountering aggression has been much higher in recent years relative to 2005, as yearly dummy variables are generally positive and statistically significant. On the other hand, few variables are significant predictors of whether a woman will encounter aggression at the hands of USBP agents when apprehended. This may be a product of the broad definition of the variable or of the smaller sample size for women. Among the exogenous variables, age and age squared are statistically different from zero. Furthermore, only one co-variate related to the border-crossing point is significant, where migrants that entered the U.S. from Tecate, into the region of east San Diego County, have a lower probability of having been victims of aggression relative to those who entered the U.S. from Tijuana. Moreover, migrants returned to Mexicali and Nogales are less likely to have been victims of some type of aggression compared with those returned to Tijuana.

 

Future Migration Intentions

Table 6 presents the results of the general model, where the relationship between aggregate measures of previous exposure to the risk of death, aggressions encountered when detained by USBP agents, and future migration intentions are explored. Men and women previously exposed to the risk of death are 3.72 and 9.44 percentage points respectively less likely to engage in future migration attempts compared to their counterparts who did not encounter these dangers. Moreover, while aggression encountered when detained is not a statistically significant predictor for women, men who were victims of some type of abuse or whose belongings were confiscated and never returned are 1.34 percentage points less likely to engage in future migration attempts. Additional controls such as previous migration experience or work experience in the U.S. are statistically significant predictors of future migration intentions, while variables related to the highest level of schooling do not seem to influence the migration decision. On the other hand, the lateral deportation variable has a negative, statistically significant effect in the case of men, since it reduces the intent to migrate at a later date by 2.77 percentage points. This result offers tangential evidence that this strategy, called the Alien Transfer Exit Program (ATEP) by the U.S. government, has proved to be a reasonably effective measure in deterring unauthorized migration.¹¹

Table 7 presents results in which the main exogenous variables of interest, in other words, previous exposure to the risk of death and aggression encountered when detained, are presented in a disaggregated manner. In regard to risk, the regressions confirm the expectation that migrants exposed to different kinds of dangers have a lower likelihood of engaging in future migrant attempts. Specifically, previous exposure to extreme temperatures lowers the desire to migrate by 3.37 percentage points for men and 6.59 percent for women, while dehydration lowers it by .025 probability points for men and .076 probability points for women, whereas exhaustion decreases it by 2.81 percentage points for men and 5.48 percent for women, among others. Amid the categories that appear to produce a positive effect, men and women who report being exposed to the risk of drowning are 2.35 and 6.06 percentage points, respectively, more likely to intend to migrate at a later date. Moreover, people who were exposed to "other" types of risks also present a heightened likelihood of intending to migrate in the future.¹² Since the results do not reflect a causal relationship, it is likely that for many migrants, being previously exposed to the risk of death is not a strong enough incentive to deter them from attempting to enter the U.S. illegally at a later date. As argued by Cornelius and Salehyan (2007:147), with successful crossings being made constantly, migrants perceive their probability of dying to be low, even though there is a general consensus that the likelihood of dying while attempting to cross the border has increased over time.

With respect to the connection between aggressions encountered when detained by USBP agents and future migration intentions, table 7 shows that the negative relationship between these two variables is mainly driven by the "confiscation of belongings" category, where this covariate decreases the probability of intending to migrate by 4.78 percentage points for men and 9.75 percentage points for women. This suggests that the reluctance of migrants to illicitly attempt to enter the U.S. at a later date may be more the result of financial constraints, particularly in the short run, rather than the psychological or physical damage caused by exposure to physical, verbal, or other types of abuse at the hands of USBP agents, since they are not statistically significant variables.

 

Conclusions

This study has shown that around one-sixth of Mexican undocumented immigrants claim to have been victims of some form of aggression at the hands of USBP agents while around one-fourth report having been exposed to the risk of death while attempting to enter the U.S. through its border with Mexico. While the present study does not claim to find a causal relationship, the results suggest that individuals who were previously exposed to the risk of death are less likely to migrate in the future compared to migrants whose life was never in jeopardy. On the other hand, being a victim of abuse when detained by USBP agents does not seem to be related to the future migration intentions of unauthorized migrants. Nevertheless, individuals whose belongings were confiscated and never returned are less likely to migrate to the U.S. at a later date compared to migrants that did not encounter such difficulties.

Although undocumented migration flows from Mexico to the U.S. have decreased in recent years, due to its higher earnings potential, the U.S. will continue to offer powerful economic incentives for Mexican workers. Since the demand for cheap labour by U.S. employers easily surpasses legal immigrant and work visa quotas, individuals that are unable to gain legal entry into the U.S. will continue to resort to unlawful means in their quest to enter the country, risking their life and well-being in the process.

In order to diminish the adverse and potentially lethal effects encountered by unauthorized migrants when attempting to cross the border, a series of complementary measures could be implemented. As a means of reducing the exposure to the risk of death among migrants, the U.S. and Mexican governments should allocate additional resources to USBP search and rescue teams and the grupos Beta, respectively, and increase their support to private altruistic efforts along the border. On the other hand, in order to decrease the number of migrants who report being victims of some type of abuse or aggression when apprehended, further training should be provided for USBP and ICE agents, accompanied by appropriate monitoring of their treatment of detained migrants. Lastly, sustained growth by the Mexican economy and the implementation of a comprehensive legal temporal-worker program which adapts to U.S. labour demand surpluses and shortages would benefit both countries and reduce the flow of migrants who endanger their lives in their attempt to enter the United States.

 

References

ALDEN, Edward, 2012, "Immigration and Border Control", The Cato Journal, vol. 32, no. 1, Winter, pp. 107-124.         [ Links ]

ALONSO, Guillermo, 2005, "Violencias asociadas al cruce indocumentado de la frontera México-Estados Unidos", Nueva Antropología, vol. XX, no. 65, May-August, pp. 113-129.         [ Links ]

ALONSO, Guillermo, 2012, "Recesión económica, reflujos migratorios y violencia antiinmigrante entre México y Estados Unidos", Norteamérica, vol. 7, no. 2, July-December, pp. 221-252.         [ Links ]

ARIZONA STATE LEGISLATURE (ASL), 2007, Legal Arizona Workers Act (Arizona Revised Statutes 23-211 to 23-214). Available at <http://www.azleg.gov/ArizonaRevisedStatutes.asp?Title=23> (last accessed on February 20, 2014).         [ Links ]

CORNELIUS, Wayne, 2001, "Death at the Border: Efficacy and Unintended Consequences of US Immigration Control Policy", Population and Development Review, vol. 27, no. 4, December, pp. 661-685.         [ Links ]

CORNELIUS, Wayne and Idean SALEHYAN, 2007, "Does Border Enforcement Deter Unauthorized Immigration? The Case of Mexican Migration to the United States of America", Regulation & Governance, vol. 1, no. 2, June, pp. 139-153.         [ Links ]

CORNELIUS, Wayne et al., 2008, Controlling Unauthorized Immigration from Mexico: The Failure of "Prevention through Deterrence" and the Need for Comprehensive Reform, San Diego, University of California at San Diego.         [ Links ]

CORTÉS, Mario, 2003, "Política inmigratoria de México y Estados Unidos y algunas de sus consecuencias", Región y Sociedad, vol. 15, no. 27, May-August, pp. 3-33.         [ Links ]

DANIELSON, Michael, 2013, "Documented failures: The Consequences of Immigration Policy on the U.S.-Mexico Border", Nogales, Jesuit Refugee Service, USA/Jesuit Conference of the United States/Kino Border Initiative.         [ Links ]

DELUCA, Lawrence; Marylyn MCEWEN, and Samuel KEIM, 2010, "U.S.-Mexico Border Crossing: Experiences and Risk Perceptions of Undocumented Male Immigrants", Journal of Immigrant and Minority Health, vol. 12, no. 1, February, pp. 113-123.         [ Links ]

EMIF NORTE, 2005, "Encuesta sobre migración en la frontera norte de México 2005. Base de datos", México, El Colef/STPS/Segob/Conapo/INM/UPM/SRE. Available at <http://www.colef.net/emif/bases.php> (last accessed on May 13, 2013).         [ Links ]

EMIF NORTE, 2006, "Encuesta sobre migración en la frontera norte de México 2006. Base de datos", México, El Colef/STPS/Segob/Conapo/INM/UPM/SRE. Available at <http://www.colef.net/emif/bases.php> (last accessed on May 13, 2013).         [ Links ]

EMIF NORTE, 2007, "Encuesta sobre migración en la frontera norte de México 2007. Base de datos", México, El Colef/STPS/Segob/Conapo/INM/UPM/SRE. Available at <http://www.colef.net/emif/bases.php> (last accessed on May 13, 2013).         [ Links ]

EMIF NORTE, 2008, "Encuesta sobre migración en la frontera norte de México 2008. Base de datos", México, El Colef/STPS/Segob/Conapo/INM/UPM/SRE. Available at <http://www.colef.net/emif/bases.php> (last accessed on May 13, 2013).         [ Links ]

EMIF NORTE, 2009, "Encuesta sobre migración en la frontera norte de México 2009. Base de datos", México, El Colef/STPS/Segob/Conapo/INM/UPM/SRE. Available at <http://www.colef.net/emif/bases.php> (last accessed on May 13, 2013).         [ Links ]

EMIF NORTE, 2010, "Encuesta sobre migración en la frontera norte de México 2010. Base de datos", México, El Colef/STPS/Segob/Conapo/INM/UPM/SER. Available at <http://www.colef.net/emif/bases.php> (last accessed on May 13, 2013).         [ Links ]

EMIF NORTE, 2011, "Encuesta sobre migración en la frontera norte de México 2011. Base de datos", México, El Colef/STPS/Segob/Conapo/INM/UPM/SRE/SS. Available at <http://www.colef.net/emif/bases.php> (last accessed on May 13, 2013).         [ Links ]

EMIF NORTE, 2012, "Encuesta sobre migración en la frontera norte de México 2012. Base de datos", México, El Colef/STPS/Segob/Conapo/INM/UPM/SRE/SS. Available at <http://www.colef.net/emif/bases.php> (last accessed on May 13, 2013).         [ Links ]

ESCHBACH, Karl; Jacqueline HAGAN, and Nestor RODRÍGUEZ, 2003, "Deaths During Undocumented Migration: Trends and Policy Implications in the New Era of Homeland Security", Defense of the Alien, vol. 26, pp. 37-52.         [ Links ]

ESCHBACH, Karl et al., 1999, "Death at the Border", International Migration Review, vol. 33, no. 2, Summer, pp. 430-454.         [ Links ]

ESPENSHADE, Thomas, 1994, "Does the Threat of Border Detention Deter Undocumented US Immigration?", Population and Development Review, vol. 20, no. 4, December, pp. 871-892.         [ Links ]

FELDMANN, Andreas and Jorge DURAND, 2008, "Mortandad en la frontera", Migración y Desarrollo, no. 10, 1^st semester, pp. 11-35.         [ Links ]

INFANTE, César et al., 2012, 'Violence Committed against Migrants in Transit: Experiences on the Northern Mexican Border", Journal of Immigrant and Minority Health, vol. 14, no. 3, June, pp. 449-459.         [ Links ]

JIMENEZ, Maria, 2009, Humanitarian Crisis: Migrant Deaths at the U.S.-Mexico Border, San Diego, United States, American Civil Liberties Union of San Diego and Imperial Counties Foundation/Comisión Nacional de los Derechos Humanos.         [ Links ]

LUHNOW, David, 2010, "Mexico Killings Show Migrants' Plight", The Wall Street Journal, New York, Americas News, August 27. Available at <http://online.wsj.com/article/SB10001424052748704913704575454033356912888.html> (last accessed on May 13, 2013).         [ Links ]

LYDGATE, Joanna, 2010, Assembly-line Justice: A Review of Operation Streamline, Berkeley, University of California at Berkeley Law School.         [ Links ]

MARRONI, María and Guillermo ALONSO, 2006, "El fin del sueño americano: Mujeres migrantes muertas en la frontera México-Estados Unidos", Migraciones Internacionales 10, vol. 3, no. 3, January-June, pp. 5-30.         [ Links ]

NEYINS, Joseph, 2003, "Thinking out of Bounds: A Critical Analysis of Academic and Human Rights Writings on Migrant Deaths in the U.S.-Mexico Border Region", Migraciones Internacionales 5, vol. 2, no. 2, July-December, pp. 171-190.         [ Links ]

NO MORE DEATHS (NMD), 2008, "Crossing the Line: Human Rights Abuses of Migrants in Short-Term Custody on the Arizona/Sonora Border", Tucson, United States, NMD.         [ Links ]

OFFICE FOR NATIONAL STATISTICS, 2014, "International Passenger Survey", London. Available at <http://www.ons.gov.uk/ons/guide-method/method-quality/specific/travel-and-transport-methodology/international-passenger-survey/index.html> (last accessed on February 20, 2014).         [ Links ]

OFFICE OF INSPECTOR GENERAL (OIG), 2006, "Treatment of Immigration Detainees Housed at Immigration and Customs Enforcement Facilities", Washington, DC, OIG-U.S. Department of Homeland Security, December (OIG-07-01).         [ Links ]

PHILLIPS, Scott; Jacqueline HAGAN, and Nestor RODRIGUEZ, 2006, "Brutal Borders? Examining the Treatment of Deportees During Arrest and Detention", Social Forces, vol. 85, no. 1, September, pp. 93-109.         [ Links ]

RUBIO-GOLDSMITH, Raquel et al., 2006, The "Funnel Effect" & Recovered Bodies of Unauthorized Migrants Processed by the Pima County Office of the Medical Examiner, 1990-2005, Tucson, United States, University of Arizona.         [ Links ]

SANTIBÁÑEZ, Jorge [paper], 2000, "Riesgos asociados al desplazamiento migratorio internacional", in "Diálogo Estados Unidos-México sobre Migración", México, Secretaría de Relaciones Exteriores.         [ Links ]

SAPKOTA, Sanjeeb et al., 2006, "Unauthorized Border Crossings and Migrant Deaths: Arizona, New Mexico, and El Paso, Texas, 2002-2003", American Journal of Public Health, vol. 96, no. 7, July, pp. 1282-1287.         [ Links ]

SLACK, Jeremy et al., 2013, In the Shadow of the Wall: Family Separation, Immigration Enforcement and Security, Tucson, United States, University of Arizona.         [ Links ]

SOLÍS, Marlene and Alonso, GUILLERMO, 2009, "Una caracterización de las mujeres en tránsito hacia Estados Unidos: 1993-2006", Papeles de Población, vol. 15, no. 62, October-December, pp. 253-283.         [ Links ]

SORENSEN, Todd and Carmen CARRIÓN-FLORES, 2007, "The Effects of Border Enforcement on Migrant's Border Crossing Choices: Diversion or Deterrence?", Tucson, University of Arizona, unpublished.         [ Links ]

SOUTHERN BORDERS COMMUNITIES COALITION (SOBOCO), 2013, "Border Patrol Abuse Since 2010". Available at <http://soboco.org/border-patrol-brutality-since-2010/> (last accessed on May 31, 2013).         [ Links ]

SPENER, David, 2008, "El apartheid global, el coyotaje y el discurso de la migración clandestina: Distinciones entre violencia personal, estructural y cultural", Migración y Desarrollo, vol. 10, 1^st semester, pp. 127-156.         [ Links ]

UNITED STATES CITIZENSHIP AND IMMIGRATION SERYICES (USCIS), 1986, "Immigration Reform and Control Act of 1986 (Pub. L.99-603)", Washington, DC, U.S. Department of Homeland Security. Available at <http://www.uscis.gov/sites/default/files/ilink/docYiew/PUBLAW/HTML/PUBLAW/0-0-0-15.html> (last accessed on January 27, 2014).         [ Links ]

UNITED STATES CITIZENSHIP AND IMMIGRATION SERYICES, 1996, "Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (Pub. L. 104-208)", Washington, DC, U.S. Department of Homeland Security. Available at <http://www.uscis.gov/iframe/ilink/docYiew/PUBLAW/HTML/PUBLAW/0-0-0-10948.html> (last accessed on January 27, 2014).         [ Links ]

UNITED STATES CUSTOMS AND BORDER PROTECTION (USCBP), 2013, "U.S. Border Patrol Fiscal Year 2012 Statistics", U.S. Department of Homeland Security. Available at <http://www.cbp.gov/xp/cgov/border_security/border_patrol/usbp_statistics/usbp_fy12_stats/> (last accessed on May 31, 2013).         [ Links ]

UNITED STATES GOYERNMENT ACCOUNTABILITY OFFICE (USGAO) [report], 2006, "Illegal Immigration: Border-Crossing Deaths Have Doubled Since 1995; Border Patrol's Efforts to Prevent Deaths Have not Been Fully Evaluated", Washington, DC, USGAO.         [ Links ]

UNITED STATES GOYERNMENT ACCOUNTABILITY OFFICE (USGAO) [report], 2010, "Alien Smuggling: DHS Needs to Better Leverage Investigative Resources and Measure Program Performance along the Southwest Border", Washington, DC, USGAO.         [ Links ]

 

Notas

¹ USBP statistics on migrant deaths undercount the actual number of fatalities, since they employ a restrictive set of criteria when classifying mortalities as an unauthorized border-crosser death (Rubio-Goldsmith et al., 2006:10) and fail to include individuals that died in Mexico or whose bodies were never recovered. Taking this into account, it is estimated that from 1993 to 2012 the total number of migrant deaths surpassed 7 000 (Alonso, 2012:230).

² The "prevention through deterrence" strategy intensified enforcement activities in unauthorized points of entry throughout specific segments of the border, and augmented the number of inspection points along key transportation arteries (Eschbach et al., 1999).

³ Measures of internal border control have also been put into effect including the creation of the E-Verify system and the enactment of the Legal Arizona Workers Act (asl, 2007), among others.

⁴ This data is taken from the Emif Norte (2005-2012). The USBP reports that illegal detentions from Mexico dropped from 1 075 221 in 2005 to 262 341 in 2012 (USCBP, 2013).

⁵ Perhaps the most blatant display of the dangers encountered by undocumented migrants was the murder of 72 people from Central and South America in the municipality of San Fernando in the state of Tamaulipas in August 2010, massacred in an act related to organized crime groups on their journey to the U.S. (Luhnow, 2010).

⁶ Going beyond the statistics on migrant deaths, Marroni and Alonso (2006) try to give this phenomenon "a human face". The authors focus on a case study involving the disappearance of migrant women from the state of Puebla, the role played by the smuggler and the reaction by other members of the community of origin.

⁷ The survey therefore does not include migrants who travel directly from their place of origin in Mexico to their destination in the U.S. (Solís and Alonso, 2009:255).

⁸ Because of data constraints, the present study does not distinguish between the categories of repatriations and deportations, and instead focus jointly on these two groups of returned migrants.

⁹ In the general model, migrants are classified as encountering aggression when they state that they suffered physical, verbal, or some other type of abuse at the hands of USBP agents when apprehended, or that their belongings were confiscated and never returned.

¹⁰ The present study also analyzed whether there is a systematic difference among respondents who do or do not report suffering physical or verbal abuse at the hands of USBP agents. Compared to respondents that were not victims of abuse, migrants that were physically or verbally assaulted by USBP agents tend to be younger, are more likely to be male, and more educated, and less likely to have had previous migration experience.

¹¹ According to USGAO (2010), the objective of the ATEP is to disturb migration patterns and break up human smuggler rings. Deportees are flown hundreds of miles from where they entered the U.S. and returned to a different port of entry in Mexico. The program has been criticized since it leaves migrants in unfamiliar surroundings in the border region where, given the high levels of insecurity in certain Mexican states, they become highly vulnerable to crime.

¹² The category other type of risk includes respondents who, when asked about the type of risk they were exposed to, answered "other" from 2005 to 2010. It also includes migrants who stated in the 2010 survey that they were exposed to the risk of death caused by: wild animals, traffic accidents, asphyxiation, abandonment by a smuggler, or assault and robbery.

 

INFORMACIÓN SOBRE LOS AUTORES

PEDRO PAULO ORRACA ROMANO: Actualmente cursa el doctorado en economía en la University of Sussex. Es maestro en economía por el Centro de Investigación y Docencia económicas y licenciado en la misma disciplina por la universidad Autónoma de Baja California. sus líneas de investigación se centran en la economía laboral y la migración internacional. Es coautor (con Robert Duval Hernández) del artículo "Análisis por cohortes de la participación laboral en México (1987-2009)", publicado en El Trimestre Económico (vol. LXXVIII: 2, núm. 310, 2011). Durante sus estudios ha recibido diversas becas otorgadas por la Fundación Telmex, el Banco de México y el Consejo Nacional de Ciencia y Tecnología, entre otras. Anteriormente se desempeñó como subdirector de seguimiento del gasto en la Secretaría de Finanzas del gobierno del Distrito Federal.

FRANCISCO DE JESÚS CORONA: Villavicencio es estudiante del doctorado Economía de la Empresa y Métodos Cuantitativos en el Departamento de Estadística de la Universidad Carlos III de Madrid. Maestro en estadística aplicada por el Instituto Tecnológico y de Estudios Superiores de Monterrey y licenciado en economía por la Universidad Autónoma de Baja California. Sus principales líneas de investigación son el análisis de series de tiempo, principalmente desde una perspectiva econométrica y el análisis multivariante. Trabajó en el Centro de Investigación en Matemáticas, unidad Monterrey, colaborando tanto en el área de vinculación como en investigación aplicada. Es coautor del artículo "A Relationship between Acute Respiratory Illnesses and Weather", publicado en Epidemiology and Infection (Cambridge University Press, 2013). Actualmente desarrolla el tema de minimización de la incertidumbre asociada a la estimación de factores comunes dinámicos no observados. Obtuvo el reconocimiento al mérito académico en sus estudios de licenciatura y recibió la beca de excelencia para estudios de maestría.