The Interactive Effects of National Origin and Immigration Status on Latino Drug Traffickers in U.S. Federal Courts

Abstract

The present study was an intragroup examination of noncitizen Latino drug traffickers convicted in the federal courts from 2006 to 2014. Drawing upon focal concerns theory and Sayad’s state-centered perspective on the role of governments in framing the crime–immigration debate, this study assessed whether offenders’ national origin conditions the effects of immigration status on the odds of receiving a downward departure (lenient sentencing) and the magnitude of the sentence discount imposed. Particular attention was paid to whether any effects worked to the detriment of Mexicans relative to non-Mexicans. The findings revealed that the effect of immigration status was contingent on offenders’ national origin for the departure decision, but not the sentence discount. While being undocumented served to decrease the odds of a downward departure regardless of national origin, the effects were greater for Mexicans than for non-Mexicans. These findings indicate that contrary to the goals of the U.S. sentencing guidelines to reduce unwarranted disparities surrounding national origin, the current federal sentencing structure allows them to thrive.

Keywords

Latinos departures immigration national origin sentencing

The 21st century represents a time of demographic shifts in the U.S. population. With this transition, Latinos rather than Blacks represent the country’s largest racial/ethnic minority group (Fry, 2008). A significant portion of this change comes from increased immigration from Latin America (Fry, 2008, p. 1; Larsen, 2004). Many native-born Americans react to these immigration patterns and their effects with demands for tougher enforcement of immigration laws and the passage of new laws to reduce the number of undocumented immigrants entering the country. They demand enhanced security along the U.S.-Mexico border and a reduction in the number of legal immigrants admitted to the United States annually (Citrin & Sides, 2008; Espenshade, 1995; Freeman & Birrell, 2001; Huntington, 2004; Jaret, 1999; Light, Massoglia, & King, 2014; Logue, 2009). Additionally, many native-born Americans call for harsher punishment for undocumented immigrants when they commit crimes, especially those of Latino origin (Morin, 2008). Nonetheless, most sentencing research on Latinos focuses on their outcomes relative to Whites and Blacks, with no consideration of possible intraethnic variation fostered by the hostility targeted toward Mexicans relative to other groups of Latinos (Light et al., 2014; Wolfe, Pyrooz, & Spohn, 2011).

In light of the current climate, the present study seeks to address this gap by exploring whether or not noncitizen Latinos immigration status and national origin affect judges’ imposition of downward departures and the magnitude of the sentence discount they impose for those granted these departures. Downward departures are a mechanism that permits judges to sentence offenders below a range prescribed by the federal sentencing guidelines. The sentencing guidelines permit judges to utilize downward departures in cases where mitigating factors (e.g., family ties and responsibilities; community ties) exist that the federal sentencing guidelines fail to consider adequately, thus they provide a means to grant leniency (U.S. Sentencing Commission, 2014a). This study asks: (1) Is the effect of immigration status on the odds of receiving a downward departure contingent on offenders’ national origin? and (2) Is the effect of immigration status on the magnitude of the sentence discount contingent on offenders’ national origin? Provision 5H1.10 of the federal sentencing guidelines precludes the use of national origin in sentencing decisions, so one would expect the answers to both questions to be no (U.S. Sentencing Commission, 2014a).

However, it is important to examine whether any support exists for this expectation and the current study is timely. First, the number of Latino immigrants and Mexican immigrants in particular continues to grow. Second, there is increasing anti-immigration rhetoric and attitudes permeating the political arena and among the public, with much of the focus centering on Mexicans, particularly undocumented ones, even as they are not the only Latinos immigrating to the United States (Huntington, 2004; Jaret, 1999; Morin, 2008). A major reason for this sentiment is the ongoing drug policy in Mexico and in the United States that seeks to eradicate drug smuggling into the United States and the corresponding violence associated with it (Zarriello & Gray, 2014). For instance, efforts targeting Mexican drug smugglers go back to the Nixon administration with the initiation of Operation Intercept in 1969, as a means of stopping the flow of Mexican marijuana into the United States. However, the program failed to achieve its goals (Zarriello & Gray, 2014).

This failure led to a shift in strategy in the United States and Mexico during the 1970s that emphasized an eradication of marijuana fields in Mexico via training, funding, and military support from the United States that persists. In the 1980s, the U.S. focus shifted to Colombian drug cartels and eradication of cocaine smuggling into the United States. However, the intense interdiction efforts led the Colombian cartels to form alliances with Mexican smugglers for new routes into the United States, which exacerbated corruption among Mexican officials and allowed for the expansion and strengthening of Mexican drug cartels relative to Colombians (Zarriello & Gray, 2014). Consequently, the public image of contemporary drug cartel figures and the risk they and their organizations pose is frequently that of a Mexican national. This imagery contributes to significant racialization of Mexican drug smugglers, especially those involved in marijuana trafficking, which has implications for the treatment of Mexican nationals relative to nationals from other Latino countries.

Finally, the data used in the study come from the post–Booker era. U.S. v. Booker (2005) altered the status of the federal sentencing guidelines from a mandatory structure to an advisory structure. Under this new system, judges still must consult the federal sentencing guidelines and consider them when determining sentences, but they are not bound to impose sentences derived from their consultation of the guidelines. This increases judges’ discretion at sentencing, but their decisions remain subject to appellate review to ensure that they are reasonable (Scott, 2011). Nonetheless, they also pose the risk of undermining a fundamental goal of the federal sentencing guidelines by facilitating unwarranted disparities based on the intersection of offenders’ immigration status and national origin.

Impact of Perceptions of an Immigration–Crime Nexus

With perceptions of an “epidemic” of undocumented immigration by Latinos, many Americans also view immigrants as individuals who commit crimes at higher rates than the native-born population (Hagan & Palloni, 1999; Rumbaut & Ewing, 2007; Wolfe et al., 2011). Political rhetoric and media coverage that highlights criminal activity by a relatively small portion of immigrants, including undocumented immigrants, sustain these views. The result is a continued belief among the public in a crime–immigration nexus, even though research shows that immigrants commit crimes at equal or lower levels than native-born citizens (Hagan, Levi, & Dinovitzer, 2008; Hagan & Palloni, 1999; Rumbaut & Ewing, 2007). For instance, Rumbaut and Ewing (2007) highlight the lower incarceration rates of foreign-born offenders relative to native-born offenders both within and between racial and ethnic groups. Hagan et al. (2008) reference early research conducted by Edwin Sutherland that examined claims of a link between immigration and crime. In summarizing Sutherland’s findings, they note, “It is acculturation to the United States that, over time, exposes immigrants and their children to native-born levels of criminal involvement” (Hagan et al., 2008, p. 99). However, they caution against drawing conclusions that the children of immigrants are more crime-prone than native-born ones. They state, “This cross-generational pattern merely is a slow process of ‘catching up’ … to the baseline rate of the native-born population” (Hagan et al., 2008, p. 100).

Judges are not immune from exposure to inaccurate information about immigration or from acting on their beliefs in the courtroom, which has significant implications for the criminal sanctioning of immigrants. Consider U.S. v. Borrero-Isaza (1989) where a Colombian defendant received harsher punishment than his Anglo codefendant did because the judge wanted to send a message to those from Colombia, which he highlighted as a source country for drugs. Additionally, in the U.S. v. Onwuemene (1991) case, an appellate court vacated the original sentence because the trial judge violated the offenders’ constitutional rights by improperly drawing upon his immigration status at sentencing. More recently, drug policy efforts have seen a significant focus on Mexico, which has become a primary source and transit country for illicit drugs, including marijuana and cocaine, into the United States. These efforts include training, data sharing, and surveillance cooperation efforts between U.S. and Mexican authorities (Executive Office of the President of the United States, 2014). Such efforts place an intense focus on Mexican offenders and the risk many officials and the public perceive they pose due to ties to drug cartels and other forms of organized crime (Morin, 2008). The advisory nature of the federal sentencing guidelines post-Booker presents an opportunity to examine whether judges exercise of greater discretion for sentencing decisions such as downward departures creates disparate treatment for offenders based on their national origin and immigration status.

Race/Ethnicity, Citizenship Status, and Sentencing

Given the evidence of national origin influencing sentencing decisions and Latinos’ status as the largest minority group, early studies that focus on Black–White disparities in sentencing outcomes produced a major gap in our understanding of race/ethnicity, sentencing, and immigration status (Demuth & Steffensmeier, 2004; Steffensmeier & Demuth, 2001). Part of that early focus rested on the availability of official data regarding Latino offenders in previous periods since officials often labeled them as Black or White. Additionally, early researchers often faced insufficient numbers of Latinos for analysis even when officials coded data on this group separately from that for Whites and Blacks (Demuth & Steffensmeier, 2004; Steffensmeier & Demuth, 2001). The result was a lack of information about the presence, origin, and dynamics of any disparities in the sentencing of these defendants relative to other groups.

More recent studies compare Latinos to Whites and Blacks on sentencing outcomes (Brennan & Spohn, 2008; Crow & Johnson, 2008; Demuth & Steffensmeier, 2004; Feldmeyer & Ulmer, 2011; Frenzel & Ball, 2007; Kautt & Spohn, 2002; Light, 2014; Light et al., 2014; Steffensmeier & Demuth, 2000, 2001; Sutton, 2013; Wolfe et al., 2011). Much of it indicates that Latinos fare worse than Whites do, and in some cases, worse than Blacks do (Albonetti, 1997; Demuth & Steffensmeier, 2004; Steffensmeier & Demuth, 2000, 2001). However, the research finds inconsistencies across sentencing decisions. Some research finds disparities at the incarceration decision, but not regarding the duration of incarceration (Bushway & Piehl, 2001).

Regardless of these findings, Demuth (2002) argues that many studies failed to consider citizenship status. Such neglect is surprising given the heated debate over immigration, especially Latino immigration, which many associate with drug trafficking and other crimes (Hagan & Palloni, 1999; Hagan et al., 2008; Jaret, 1999). More researchers now include citizenship status, but largely as controls (Feldmeyer & Ulmer, 2011; Kautt & Spohn, 2002). Nonetheless, some recent studies make citizenship status a fundamental component of their analyses.

Wolfe, Pyrooz, and Spohn (2011) examine interactions between citizenship status and race/ethnicity. They find that among undocumented immigrants, Latinos receive longer sentences than Whites do, but among U.S. citizens, the reverse occurs. They conclude that the focal concern of dangerousness operates differently when judges encounter U.S. citizens compared to undocumented immigrants. Their findings also indicate that while citizenship matters at sentencing, for noncitizens, immigration status also matters. Light (2014) and Light, Massoglia, and King’s (2014) research further highlight the importance of treating citizenship status as a key predictor rather than just as a control variable, especially with research on Latinos. Light (2014) estimates, “80 percent of Hispanics punished in U.S. federal courts over the past two decades lacked U.S. citizenship” (p. 448). He finds disparities arising from citizenship status to the detriment of noncitizens with this punitive effect being greater than minority–White disparities. More specifically, he found that much of the effect of being Latino on sentencing outcomes is a product of greater punitive treatment of noncitizens, which describes many Latino offenders in federal courts.

Hartley and Armendariz (2011) found that partitioning analyses by citizenship status revealed no effect of ethnicity on sentencing outcomes. However, they examined only sentence length and confirmed the lack of disparities between minorities and Whites found in other research using this measure (Bushway & Piehl, 2001). Additionally, this study only focused on five southwest districts along the U.S.-Mexico border rather than all districts in the federal courts (excluding territories). Moreover, it only focused on 1 year of data and thus failed to consider period effects like Light (2014).

Even with the continued improvements in research on ethnicity and citizenship status in relation to the federal courts, the bulk of current research assumes similar treatment of all Latinos by the courts, regardless of their national origin or ethnicity. Very few studies examine Latinos independent of their treatment relative to other groups (Frantzen & Worley, 2008; Logue, 2009). Frantzen and Worley (2008) examine variation among Latino drug offenders using data from two Texas counties along the Texas–Mexico border. The use of state-level data is unique relative to the significant amount of research on the federal courts. They also focus on an indeterminate sentencing structure rather than a guidelines system. A major weakness is that the authors fail to consider citizenship status and they only examine two counties in Texas, thus limiting generalizability (Frantzen & Worley, 2008).

In contrast, Logue (2009) also examines intraethnic variation in sentencing outcomes among Hispanics. She sought to compare Hispanics based on their national origin and immigration status and used multiple years of federal sentencing data from the pre–Booker period. A major limitation of the federal sentencing data is that she could only examine immigration status among noncitizens rather than more broadly considering the influence of citizenship status. This is because national origin/ethnicity is unavailable for Hispanic U.S. citizens. Even with this limitation, her work showed that national origin and immigration status mattered for sentencing outcomes, often to the detriment of Mexicans relative to non-Mexican Hispanics.

Theoretical Perspectives

In analyzing the findings, the current study draws upon two perspectives. One perspective on sentencing that researchers commonly use to explain minority–White disparities and citizenship status disparities is the focal concerns perspective (Steffensmeier, Ulmer, & Kramer, 1998). According to this perspective, three focal concerns drive judicial decision-making: offender blameworthiness, offender dangerousness, and the practical consequences of judges’ decisions. Offenders deemed more blameworthy and whom judges perceive as posing a significant risk to society are likely to face the harshest treatment. Additionally, from a practical standpoint, judges consider offenders’ “ability to do time” (Steffensmeier et al., 1998, p. 767) and the impact on organizational resources, such as overcrowding in correctional facilities and costs. This perspective is very relevant for studies on the main and interactive effects of citizenship status and race/ethnicity, or in the current case immigration status and national origin.

With the anti-immigration attitudes and corresponding rhetoric permeating society today, especially rhetoric linking immigrants to crime (Hagan & Palloni, 2008; Huntington, 2004), and the limited information that judges often have about offenders, especially noncitizens, they may rely on a “perceptual shorthand” (Steffensmeier et al., 1998, p. 767) in rendering their decisions. In the case of citizenship status, noncitizens may be perceived as more threatening and more blameworthy than U.S. citizens, which might explain research findings that noncitizens receive harsher treatment than U.S. citizens in most cases (Light, 2014; Light et al., 2014). From an immigration standpoint, judges may perceive undocumented immigrants as more dangerous and blameworthy than legal immigrants. From a practical standpoint, judges may grant what appears to be leniency with sentence length due to expectations of deportation proceedings that they wish to expedite to remove noncitizens from the country quickly. Doing so would remove a perceived threat and reduce burdens on the correctional system that has to supervise them (Light, 2014; Light et al., 2014).

Judges may also perceive Latino noncitizens as particularly dangerous and blameworthy because of their increased likelihood to be undocumented immigrants. The perceptions of immigrants’ inability or unwillingness to assimilate into the dominant Anglo culture lead to the assumption of criminal activity (Huntington, 2004). This might lead to particularly harsh treatment for this group. When one considers the disaggregation of noncitizens into legal and undocumented immigrants, undocumented Latinos are likely to face the harshest treatment. Moreover, given the focus on Mexicans in the immigration debate relative to other Latinos, this group may face the harshest treatment among Latinos.

Sayad (2004) offers some insight into how judges may develop this “perceptual shorthand” in his decision-making regarding noncitizens with his state-centered perspective about the role of governments in framing the crime–immigration debate. He argues that neither racism nor politics are the sole or primary origin of the perceived crime–immigration nexus. Instead, the state establishes a separate “immigrant” category that serves to place immigrants in the position of “the other” relative to native residents. Sayad (2004) argues that this process results in immigrant status becoming viewed as a form of situational delinquency in which immigrants alleged to be criminals face double punishment as immigration itself is put on trial for being “a form and source of delinquency” (p. 282). He asserts, “The fact of being an immigrant delinquent (or a delinquent immigrant) constitutes as a general rule, something of an aggravating circumstance” (Sayad, 2004, p. 282). However, he maintains that this bias against immigrants is frequently unconscious; thus no one, including judges, is immune to its influence in the face of a criminal offense.

In fact, research using implicit association testing to ascertain whether implicit bias affects attitudes regarding immigration policies shows the presence of a “negative implicit attitude toward Latino immigrants” (Perez, 2010, p. 529). Additionally, it shows that perspectives on immigration policies arose because of implicit biases against Latinos, even when the focus was on immigrants of other racial/ethnic backgrounds (Perez, 2010). While the present study precludes an analysis of judges’ attitudes toward immigrants, the presence of disparities in treatment that should not arise among similarly situated offenders suggests that implicit bias may be a factor. However, as Perez (2010) notes, additional research must occur to examine the link between implicit attitudes and behaviors.

At the point that the criminal offense occurs, immigration status becomes visible and can heighten punitive orientations among judges, other agents of the state, and the public. Even in the face of sentencing guidelines specifically geared toward reducing unwarranted disparities based on extralegal factors, such as national origin, Sayad’s (2004) concept of immigration as an “objective sin” suggests that the effects of being an immigrant cannot be “neutralized” (p. 283). Consequently, “the case against immigration is always inseparable from that made against the immigrant because of some offense, even a minor one that he has committed” (Sayad, 2004, p. 284).

However, Sayad (2004) offers no discussion of how judges might respond to immigrants based on their national origin. Additionally, Sayad focuses on the broad category of immigrant rather than distinguishing between documented and undocu- mented immigrants. Given Sayad’s view that being an “immigrant delinquent or a delinquent immigrant” (p. 282) serves as an aggravating factor, the expectation is that undocumented immigrants who commit serious crimes will be particularly vulnerable to harsh treatment by the criminal justice system relative to documented immigrants. Many immigration critics point out that they broke not just one law, but two.

First, they violated U.S. sovereignty by being in the country without authorization. Second, they compounded their illegal behavior by committing a more serious criminal offense (Hagan et al., 2008). If judges adopt these views, undocumented immigrants may face harsher punishment than legal immigrants do. As Sayad (2004) notes, their status as immigrants already makes them delinquents, but their undocumented status places them in the role of law breaker, even if their violation is restricted to illegal entry. When they commit more serious criminal offenses, such as drug trafficking, judges and the broader society may perceive them as compounding their criminality by further taking advantage of U.S. society. Since Latinos, and Mexicans in particular, arguably represent the most salient group in the debate over undocumented immigration (Keogan, 2002), examining the direct effect of immigration status on offenders’ sentencing outcomes and whether its effects are contingent on national origin to the detriment of Mexicans is crucial. Given prior research on the impact of citizenship status and race/ethnicity, I hypothesize that undocumented immigrants will be less likely to receive lenient treatment than legal immigrants regardless of their national origin. However, I expect the effects to be more pronounced for Mexicans than for non-Mexicans.

Methodology

Data

The current study used federal sentencing data from the U.S. Sentencing Commission’s Monitoring Federal Criminal Sentences series for the fiscal years 2006–2014. The data provided measures of defendant and court characteristics, drug-related measures, financial sanctions, current-offense type and characteristics, and sentencing information. Cases from the following district courts were excluded: Puerto Rico, the District of Columbia, the Virgin Islands, Guam, and the Northern Mariana Islands. The present study focuses solely on drug trafficking offenders of both sexes, who are Latino noncitizens (n = 27,698). The goal was to assess whether differential treatment exists for Mexicans relative to other Latino ethnic groups in regards to their odds of receiving downward departures, and the magnitude of the sentence discount departures provide. The focus on drug trafficking offenders arose because the rhetoric about a Latino immigration–crime nexus centers on drug trafficking, especially for undocumented Mexicans. The selection of noncitizens rather than both citizens and noncitizens occurred because data on national origin/ethnicity only exist for noncitizen offenders sentenced in the federal courts. While immigration status and national origin are the key variables of analysis, I examine several other extralegal and legal factors available in the data.

Dependent Variables

Two measures served as the dependent variables in this study. The first measure was a dichotomous variable indicating whether offenders received a downward departure (0 = no; 1 = yes). The second measure was a continuous variable indicating the magnitude of the sentence discount earned by offenders who received downward departures. It was calculated by subtracting the sentence length imposed from the presumptive sentence length identified in the guidelines and capping the variable at 470 months (life). To address the skewed distribution, I conducted a logarithmic transformation on the measure prior to conducting my analyses.

Independent Variables

The key extralegal variables were immigration status and national origin. Immigration status was a dichotomous measure (undocumented immigrant = 1, legal immigrant = 0). National origin was a measure of whether offenders were Mexican or non-Mexican Latinos (non-Mexican = 0, Mexican = 1). Other extralegal measures served as control measures. Sex was coded 0 for females and 1 for males. Age was a continuous measure capped at 70 years. The measures for education involved four dummy variables: less than a high school diploma (reference category), high school graduate, some college, and college graduate. For each dummy variable offenders received a 1 if they met the criteria and a 0 otherwise.

Several legal factors also served as controls. The presumptive guideline minimum was an indication of the minimum sentence prescribed by the federal sentencing guidelines capped at 470 months. Drug type included six dummy variables indicating the primary drug involved in the case: crack cocaine, powder cocaine, heroin, marijuana (reference group), methamphetamine, and other drugs. For each drug, offenders received a 1 if the drug was involved in the case and a 0 otherwise. Application of a mandatory minimum sentence for guns was a composite measure recoded from the three gun mandatory minimum measures in the data set (i.e., gunmin1, gunmin2, and gunmin3). Offenders received a 1 if the mandatory minimum was applied and a 0 otherwise. Other measures included mode of conviction (0 = guilty plea, 1 = trial), presentence detention status (0 = not detained, 1 = detained), and weapon enhancement (0 = not applied, 1 = applied).

Two other sets of control measures were also included. The first set of dummy variables pertained to the 89 districts in which offenders were sentenced. Offenders received a 1 if they were sentenced in a district and a 0 otherwise. The reference category was district 42 (Texas West), as it had the highest frequency of cases at 24.2% of all cases. The last set of nine dummy variables pertained to the fiscal year in which offenders were sentenced as a way to control for any period effects. Offenders received a 1 for the year in which they were sentenced and a 0 otherwise. The year 2006 served as the reference group. These controls were excluded from the tables for brevity, but the full results are available from the author upon request.

Statistical Procedure

Logistic regression suited the analysis of the predictors of receiving a downward departure, as this measure was dichotomous. The Heckman two-step procedure was used regarding the length of sentence produced by the downward departure in order to account for sample selection bias. The pattern and direction of results regarding the key variables were similar to those obtained using ordinary least squares (OLS) regression, indicating that the findings are robust. However, the findings from the Heckman model were presented here. In order to obtain a more nuanced understanding of the interaction between immigration status and national origin, the study also utilized pooled analyses of the data followed by analyses in which the data were partitioned by national origin.

Results

Descriptive Statistics

Table 1 displays the results of the univariate analysis of the variables used in the study. The pooled analysis of the extralegal variables reveals that males constitute an overwhelming majority of the sample. Most offenders (75%) lack a high school diploma, are undocumented immigrants (75.4%), and are Mexican (82.1%). The analysis of the legal variables for the pooled sample reveals that 96.3% of offenders experienced presentence detention. Regarding the primary drug involved in their offenses, almost half of offenders’ cases (42.4%) involve marijuana. Another 27.5% involve powder cocaine and 19.8% involve methamphetamine. The number of cases involving crack cocaine and other drugs (e.g., prescription pills) is at a negligible level. The mean sentence prescribed by the guidelines is 68.6 months. With regard to the dependent variables, 24.1% receive a downward departure with an average sentence discount of 25.4 months.

Table 1.

Descriptive Statistics for the Variables Used in the Analysis.

Variables	Pooled	Mexican Legal Immigrants	Mexican Undocumented Immigrants	Non-Mexican Legal Immigrants	Non-Mexican Undocumented Immigrants
Males (%)	92.9	86.0	94.6	91.8	94.5
Education (%)
Not a high school graduate	75.0	69.4	81.5	50.8	61.6
High school graduate	15.5	18.2	11.8	30.8	22.9
Some college	7.1	9.7	5.0	14.2	11.4
College graduate	2.3	2.7	1.7	4.3	4.0
Undocumented immigrants (%)	75.4	—	—	—	—
Mexicans (%)	82.1	—	—	—	—
Mean age (years)	33.3	35.3	31.5	38.6	37.4
Trial (%)	3.9	5.1	3.2	6.7	4.4
Detained (%)	96.3	93.0	98.8	82.3	96.9
Drug type (%)
Powder cocaine	27.5	27.5	21.5	44.1	53.6
Crack cocaine	1.3	0.4	0.7	3.9	4.3
Heroin	8.4	3.7	4.5	28.1	25.9
Marijuana	42.4	48.9	49.4	15.4	7.5
Methamphetamine	19.8	19.2	23.8	4.5	6.6
Other drugs	0.6	0.2	0.1	4.0	2.1
Gun mandatory minimum (%)	0.9	0.4	0.9	1.1	1.2
Weapon enhancement (%)	6.5	4.4	7.5	4.1	5.5
Mean guideline minimum (months)	68.6	62.9	65.0	76.7	94.5
Downward departure^a	24.1	23.7	20.7	39.9	34.5
Mean sentence discount (months)^b	25.4	24.9	25.5	25.3	26.0
N	27,698	4,679	18,055	2,123	2,841

^aA significance test to compare the subgroups on their chances of receiving a downward departure found a χ² of 565.993, df = 3, p = .000. This statistically significant relationship between the subgroups and the receipt of a downward departure is borne out in the regression analyses regarding the likelihood of receiving a downward departure. ^bAn analysis of variance conducted to compare the subgroups on the sentence discount measure found F = .248, df = 3, p = .863. This indicates that the subgroups receive similar treatment once judges decide to grant the downward departure. The regression findings in Table 4 support this pattern.

In partitioning the sample into four subgroups, several differences emerge (see Table 1). Among undocumented Mexican immigrants, 81.5% of offenders lack a high school diploma, which is the highest proportion among all groups. Only 50.8% of non-Mexican legal immigrants lack a high school diploma, which is the lowest proportion for the four groups. The analysis of the legal variables shows that the overwhelming majority of offenders in all four subgroups were in custody prior to sentencing, but while it was over 90% for three groups, only 82.3% of non-Mexican legal immigrants experienced presentence detention. With regard to the type of drugs involved in the offenses, non-Mexicans are substantially more likely to have cases involving powder cocaine and heroin than Mexicans. The third most common drug for non-Mexicans is marijuana, regardless of immigration status. In contrast, for Mexicans, marijuana accounts for about half of their cases, regardless of immigration status, followed by powder cocaine and methamphetamine. The presumptive minimum sentence prescribed by the guidelines reveals that non-Mexicans have notably higher average minimum sentences prescribed than Mexicans do, with the highest average minimum sentence prescribed for non-Mexican undocumented immigrants (94.5 months). For non-Mexican legal immigrants, the average minimum sentence prescribed by the guidelines is 76.7 months.

In terms of the dependent variables, the proportion receiving a downward departure notably differs between Mexicans and non-Mexicans. Non-Mexicans have a higher likelihood of receiving downward departures than Mexicans do. For those who receive downward departures, the average sentence discount is similar across the subgroups.

Likelihood of Receiving a Downward Departure

Table 2 displays the results for the analysis of the main and interactive effects of immigration status and national origin on the odds of receiving a downward departure, net of other factors. Model 1 shows that immigration status but not national origin significantly influences the likelihood of a downward departure, with undocumented immigrants having lower odds of a departure than legal immigrants. The inclusion of the interaction term in Model 2 indicates that the effect of immigration status is contingent on national origin. The following analysis examines this interaction effect in more detail.

Table 2.

Logistic Regression of Immigration Status and National Origin on the Odds of a Downward Departure, FY 2006–2014.

Variables	Model 1			Model 2
Variables	B	SE	Exp(B)	B	SE	Exp(B)
Male	–.819	.060	0.441***	–.814	.061	0.443***
High school graduate	.087	.045	1.091	.088	.045	1.091
Some college	.280	.061	1.323***	.277	.061	1.319
College graduate	.260	.104	1.297*	.261	.104	1.298*
Undocumented immigrant	–.195	.040	0.823***	–.065	.070	0.937
Mexican	.103	.065	1.109	.231	.086	1.260**
Age	–.001	.002	0.999	–.002	.002	0.998
Trial	–.214	.085	0.807*	–.212	.085	0.809*
Detained	–.415	.084	0.660***	–.426	.084	0.653***
Powder cocaine	.269	.053	1.308***	.261	.053	1.299***
Crack cocaine	.485	.136	1.624***	.479	.136	1.615***
Heroin	.418	.070	1.519***	.411	.070	1.508***
Methamphetamine	.625	.060	1.868***	.622	.060	1.862***
Other drugs	.296	.183	1.345	.305	.183	1.357
Gun mandatory minimum	–.270	.164	0.763	–.265	.164	0.767
Weapon enhancement	–.219	.068	0.804**	–.216	.068	0.806**
Presumptive guideline minimum	.006	.000	1.006***	.006	.000	1.006***
Immigration Status × National Origin	—	—	—	–.192	.084	0.825*
Constant	–2.253	.146	.105***	–2.329	.150	0.097
Nagelkerke R ²		.297			.297
N	27,698			27,698

*p < .05. **p < .01. ***p < .001.

Table 3 displays the results of the partitioned models assessing how the effect of immigration status varies by national origin. As expected, undocumented immigrants of both groups have lower odds of receiving a downward departure than legal immigrants, but the effect of immigration status is only statistically significant for Mexicans. A test of the difference in regression coefficients advocated by Paternoster, Brame, Mazerolle, and Piquero (1998) finds a z-score of −2.66 (p = .004), which indicates that the effect of immigration status is contingent on national origin, with more pronounced negative effects for Mexicans.

Table 3.

Logistic Regression of the Immigration Status on the Odds of Receiving a Downward Departure Partitioned by National Origin.

Variables	Mexicans			Non-Mexicans
Variables	B	SE	Exp(B)	B	SE	Exp(B)
Male	–.826	.069	0.438***	–.797	.133	0.451***
High school graduate	.165	.055	1.179**	–.114	.081	0.893
Some college	.346	.076	1.414***	.109	.104	1.115
College graduate	.487	.128	1.627***	–.203	.177	0.816
Undocumented immigrant	–.238	.049	0.788***	–.017	.072	0.983
Age	.002	.002	1.002	–.010	.003	0.990**
Trial	–.284	.100	0.753**	–.093	.166	0.911
Detained	–.322	.118	0.725**	–.391	.121	0.676**
Powder cocaine	.313	.059	1.368***	–.159	.139	0.853
Crack cocaine	.147	.229	1.159	.394	.212	1.483
Heroin	.391	.091	1.478***	.066	.151	1.069
Methamphetamine	.590	.064	1.805***	.270	.233	1.310
Other drugs	.269	.464	1.309	–.038	.230	0.962
Gun mandatory minimum	–.280	.190	0.756	–.103	.339	0.902
Weapon enhancement	–.277	.076	0.758***	–.049	.169	0.952
Presumptive guideline minimum	.007	.000	1.007***	.004	.001	1.004***
Constant	–2.360	.165	0.094***	–1.483	.372	0.227***
Nagelkerke R ²		.290			.279
N	22,734			4,964

*p < .05. **p < .01. ***p < .001.

Beyond the difference regarding immigration status, other differences emerge in the partitioned analysis. First, more predictors exert statistically significant effects on offenders departure chances among Mexicans than among non-Mexicans. Among Mexicans, the factors that increase the odds of a downward departure include having at least a high school diploma; having offenses involving powder cocaine, heroin, or methamphetamine; and having a high presumptive sentence. Being male, opting to go to trial, being detained prior to sentencing, and receiving a weapon enhancement lowers the odds of a downward departure. Among non-Mexicans, only four predictors significantly affect departure chances. Offenders with higher presumptive sentences have higher odds of receiving a downward departure than those with lower presumptive sentences. Being male, older, and experiencing presentence detention decreases the odds of a departure.

Departure Sentence Discounts

In assessing the magnitude of the sentence discount that offenders experiencing a downward departure receive, Table 4 presents the results from models with and without an interaction term for immigration status and national origin. In the first model, neither national origin nor immigration status is significant, but the coefficients are negative as expected. In Model 2, which includes the interaction term for immigration status and national origin, the interaction term is not statistically significant. This indicates that national origin does not condition the effect of immigration status on the magnitude of the sentence discount offenders receive.¹

Table 4.

Linear Regression of Immigration Status and National Origin on the Logged Sentence Discount for Those With Downward Departures.

	Model 1			Model 2
Variables	B	SE	Exp(B)	B	SE	Exp(B)
Male	−.231	.031	0.794***	–.230	.031	0.795***
High school graduate	.014	.023	1.014	.013	.023	1.013
Some college	.103	.031	1.108**	.101	.031	1.106**
College graduate	.077	.053	1.080	.076	.053	1.079
Undocumented immigrant	–.017	.021	0.983	.021	.034	1.021
Mexican	–.016	.034	0.984	.022	.044	1.022
Age	.000	.001	1.000	.000	.001	1.000
Trial	–.275	.043	0.760***	–.273	.043	0.761***
Detained	–.254	.042	0.776***	–.258	.042	0.773***
Powder cocaine	.199	.030	1.220***	.197	.030	1.218***
Crack cocaine	.269	.070	1.309***	.269	.070	1.309***
Heroin	.274	.037	1.315***	.274	.037	1.315***
Methamphetamine	.491	.034	1.634***	.492	.034	1.636***
Other drugs	.263	.088	1.301**	.266	.088	1.305**
Gun mandatory minimum	–.597	.082	0.550***	–.598	.082	0.550***
Weapon enhancement	–.049	.036	0.952	–.048	.036	0.953
Presumptive guideline minimum	.008	.000	1.008***	.008	.000	1.008***
Immigration Status × National Origin	—	—	—	–.058	.042	0.944
Constant	1.987	.104	—***	–1.126	.083	—***
N	6,516			6,516

*p < .05. **p < .01. ***p < .001.

Discussion

Sanctioning offenders convicted in the federal courts relies on a uniform sentencing structure promulgated in the federal sentencing guidelines. These guidelines are now advisory via the decision in U.S. v. Booker (2005) and thus expand judges’ discretion. This transition to an advisory sentencing structure occurred as major demographic shifts occurred in the country that made Latinos, rather than Blacks, the largest minority group in the United States. Coinciding with these shifts and projections that they will continue as the United States moves toward a majority–minority country is increasing anti-immigration rhetoric and actions against Latinos, particularly undocumented Mexicans. While judges are to be impartial in their sentencing decisions in order to avoid unwarranted disparities that lead to unequal treatment based on ascribed characteristics, such as national origin/ethnicity, they are not immune from unconsciously internalizing biases regarding national origin (U.S. v. Borrero-Isaza, 1989) and immigration status (U.S. v. Onwuemene, 1991) that influence their sentencing decisions.

In fact, the U.S. v. Borrero-Isaza case occurred the same year that the U.S. Supreme Court upheld the mandatory sentencing guideline structure as constitutional in U.S. v. Mistretta (1989) and the U.S. v. Onwuemene case (1991) occurred 2 years after this decision. Additionally, the U.S. Sentencing Commission recognized the possibility of judges creating unwarranted disparities based on national origin by expressly prohibiting its consideration at sentencing (see Section 5H1.10 of the federal sentencing guidelines). While not expressly prohibited by the sentencing guidelines, appellate courts indicated that immigration status is also unacceptable for judges to consider (U.S. v. Onwuemene, 1991). With the advisory nature of the federal sentencing guidelines, however, the chance for biased attitudes toward immigrants to affect sentencing decisions increases, even unconsciously.

This study provides some support for this assertion regarding the likelihood of receiving downward departures. Based on the finding that national origin exerts no statistically significant effect on departure chances (see Model 1, Table 2), one might conclude that judges adhere to provision 5H1.10 of the sentencing guidelines. However, the model with the interaction term and the partitioned models (see Table 2, Model 2, and Table 3) show that the effect of immigration status depends on the offenders’ national origin, with a greater disadvantage for undocumented immigrants who are Mexican compared to non-Mexican. This finding reveals that additional efforts are necessary to reduce unwarranted disparities based on national origin.

These findings also support the focal concerns perspective (Steffensmeier et al., 1998). Judges may deem undocumented immigrants more blameworthy and dangerous than legal immigrants. An extensive focus exists on Mexicans who enter the country illegally and on the brutality of Mexican drug cartels that the media and politicians frequently warn of as a threat to U.S. citizens. Sayad’s (2004) state-centered perspective about how governments frame the crime–immigration debate and how this affects how immigrants are treated, offers some insights into the “perceptual shorthand” that guides judges’ assessment of offenders in light of the three focal concerns espoused by Steffensmeier, Ulmer, and Kramer (1998). Sayad (2004) argues that while racism and politics may seep into state actors’ decision-making via implicit and explicit bias, they are not the primary or sole mechanisms shaping their decisions. He noted that immigration status operates as a form of situational delinquency in which immigrants are deemed “the other” such that being an “immigrant delinquent or a delinquent immigrant” (Sayad, 2004, p. 282) acts to aggravate the effects of immigration status on state actors’ decision-making, albeit unconsciously. In fact, Sayad (2004) noted that immigration status effects “cannot be neutralized” (p. 283) because the status of immigrants as “the other” intertwines with their offending, even for minor offenses. Since drug trafficking is not a minor offense, this reality may become highly pronounced.

While Sayad (2004) failed to distinguish between immigrants based on their national origin or on their immigration status, the extensive anti-immigration climate surrounding undocumented Mexicans (Huntington, 2004) and the fact that they constitute a large number of offenders sanctioned in the federal courts during the study period, likely makes their immigration status more pronounced compared to undocumented non-Mexicans. Of particular interest is the finding that many of the cases involving Mexicans involve marijuana trafficking rather than cocaine trafficking, regardless of the immigration status. At first glance, the harsh punishment doled out to undocumented Mexicans is surprising in light of long-standing policy that emphasizes harsh treatment of cocaine traffickers, especially those who are part of organized crime groups such as drug cartels (Executive Office of the President of the United States, 2014). It is also surprising in light of the movement to decriminalize or legalize marijuana for medicinal and recreational purposes.

It becomes less surprising when one considers that this move toward liberal marijuana policies is at the state level. According to President Obama’s 2014 national drug control strategy, “marijuana availability appears to be growing because of sustained high levels of production in Mexico along with domestic cultivation” (Executive Office of the President of the United States, 2014, p. 35). Such a statement along with policy efforts geared toward improving estimates of the yield of marijuana globally and increasing crackdowns on those trafficking this substance (Executive Office of the President of the United States, 2014) indicate the importance the President places on aggressively combatting trafficking offenses involving marijuana (Executive Office of the President of the United States, 2014). This prioritization of combatting marijuana trafficking has direct implications for the way that federal prosecutors handle these cases, which also shapes the options available to judges and helps to guide their sentencing decisions.

Therefore, while judges may feel punitive toward all undocumented Latino offenders for their dual offenses of violating U.S. sovereignty and committing a serious crime associated with violence while here, their “perceptual shorthand” (Steffensmeier et al., 1998, p. 767) may make them act more punitively toward undocumented Mexicans than their non-Mexican counterparts. While judges may still want all undocumented offenders deported, they may also want to send a strong message to the group deemed to pose the biggest threat. They do so by ensuring that undocumented Mexican immigrants experience U.S. justice before any deportation proceedings occur. In fact, the desire to discourage further infractions with harsh punishment may override concerns about the practical costs (exacerbating prison overcrowding and financial expenditures to house offenders) of incarcerating undocumented immigrants. Moreover, judges may consider the fact that with the extensive places to enter the United States across the U.S.-Mexico border, simply expediting deportation of undocumented Mexicans without forcing them to serve harsh sentences is futile since they can so easily reenter the county relative to other Latino groups.

Another factor that leads to the lower odds of receiving a downward departure being more pronounced for undocumented Mexicans than for undocumented non-Mexicans is the greater involvement of undocumented non-Mexicans in cases involving drugs that carry the harshest penalties (i.e., powder cocaine and heroin). In 2014, powder cocaine and heroin carried average sentences of 76 months and 70 months, respectively. In contrast, about half of all undocumented Mexicans have convictions for marijuana, which carried an average sentence of 36 months (U.S. Sentencing Commission, 2014b). Therefore, while judges may still want undocumented non-Mexicans to experience U.S. justice before deportation proceedings occur, the much longer sentences they tend to face relative to undocumented Mexicans may lead judges to be more concerned with the practical costs of housing these offenders.

The findings regarding the magnitude of the sentence discount received for those granted a downward departure show that the more important decision for judges appears to be whether to grant a downward departure. Granting such a sanction already constitutes a form of leniency that lowers the sentence range that offenders now face. Judges might also want to send a strong message to offenders about committing drug trafficking offenses, especially since both groups have the status as “other” highlighted by Sayad (2004). Therefore, for those lucky enough to receive a downward departure, judges may treat them similarly.

Conclusion

While many studies compare the sentencing outcomes of Latinos relative to Whites and Blacks, the current study sought to examine intragroup differences among Latinos. This research examines judges’ decisions to grant downward departures and the magnitude of the sentence discount they provide. While undocumented immigrants had less chance of receiving downward departures than legal immigrants regardless of national origin, the results show that the magnitude of this disadvantage depends on national origin, to the detriment of Mexicans over non-Mexicans.

This is consistent with Logue’s (2009) finding that when aggravating effects occurred for sentence length they largely served to disadvantage Mexicans over non-Mexicans. While her study relied on data during the time when the guidelines were mandatory, this study draws upon data from the period after the guidelines became advisory. Although the sentencing outcome is not the same, it still raises questions about how the shift to an advisory structure affects the chances of unwarranted disparities at sentencing. Future research should examine whether the effects of immigration status on downward departures is contingent on national origin in the pre–Booker period and whether any differences are more pronounced in the face of advisory guidelines relative to a mandatory structure. If so, that would suggest the need for the U.S. Sentencing Commission and Congress to consider what changes need to be made to eliminate this unwarranted disparity, including whether any steps should be taken to try to return the guidelines to mandatory status.

The study also highlights that while immigration status is contingent on national origin for the decision to grant a downward departure, this is not true regarding the magnitude of the sentence discount offenders receive. Therefore, once judges decide to grant downward departures to offenders, they appear to be less concerned with how much of a discount they give. However, they may also feel constrained about how much of a discount they can give in light of the severity of drug trafficking offenses and the collateral consequences they have for community safety, given the high levels of violence associated with this crime.

While the findings highlight the importance of considering intragroup differences among Latinos on sentencing outcomes, especially in relation to immigration status and national origin, the study is not without limitations. First, there is no way to know if the inclusion of U.S. citizens would alter the results, as national origin/ethnicity information for that group is unavailable in the data. Being able to include U.S. citizens would bring the study more in line with Sayad’s (2004) focus on the situational delinquency attributed to immigrants relative to citizens and also permit the examination of immigration status among noncitizens used here. Second, the current study does not contain an interjurisdictional examination that allows for an assessment of whether national origin conditions immigration effects across all districts or in districts with large populations of Latinos, especially undocumented ones. Future research should examine this issue via multilevel modeling in order to see how extensive the problem of judges considering national origin at sentencing is in light of the provision in the sentencing guidelines saying it is irrelevant.

A third limitation is that the study lacks controls for other factors that might influence sentencing such as family and community ties and judge characteristics. While data on family ties are available in the federal sentencing data set, the U.S. Sentencing Commission urges caution against using the data (Reedt, Semisch, & Blackwell, 2013) and the information is limited to the number of dependents offenders have. While this information does highlight a potential social bond that may encourage offenders to conform, having dependents obviously failed to stop individuals from offending. Having more detailed information on the nature and quality of offenders’ family and community circumstances could shed some light on whether judges took these factors into account, and if so, how. The characteristics of the judges may also be relevant.

Future research should undertake interviews with federal judges to gain a deeper understanding of how immigration status and national origin factor into their sentencing decisions, as evidence suggests that these factors have an effect on sentencing outcomes (Johnson, 2006; Logue, 2011).

Finally, a comparison between Mexicans and other specific Latino groups, such as Puerto Ricans and Cubans, would enhance our understanding of ethnicity and sentencing. The three groups constitute the largest groups of Latinos in the United States. Combining these groups with others into a non-Mexican Latino category may mask some differences or similarities between them and Mexicans or between them and other Latino populations. Such a study would require a different data source, such as data obtained via primary data collection, as Puerto Ricans are U.S. citizens. As previously noted, their status as citizens means that it is impossible to determine who has this ethnicity among the Latino citizen offender population in the current data.

From a policy standpoint, the findings in this study suggest that achieving the goal of avoiding unwarranted disparities based on national origin and immigration status remains elusive. While appellate courts have deemed immigration status an inappropriate factor to consider at sentencing, national origin magnifies the disparities between legal immigrants and undocumented immigrants. This is problematic in light of the guidelines provision that says that national origin is not relevant to sentencing decisions. Downward departures are necessary tools for judges to have in order to mitigate the punishment imposed on offenders they deem to be less blameworthy and/or dangerous. However, the current advisory nature of the sentencing guidelines creates greater opportunity for judges to depart in ways that create or exacerbate unwarranted disparities.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Note

References

Albonetti

C. A.

(1997). Sentencing under the federal sentencing guidelines: Effects of defendant characteristics, guilty pleas, and departures on sentence outcomes for drug offenses, 1991-1992. Law and Society Review, 31, 789–822.

Brennan

P. K.

Spohn

(2008). Race/Ethnicity and sentencing outcomes among drug offenders in North Carolina. Journal of Contemporary Criminal Justice, 24, 371–398.

Bushway

S. D.

Piehl

A. M.

(2001). Judging judicial discretion: Legal factors and racial discrimination in sentencing. Law and Society Review, 35, 733–764.

Citrin

Sides

(2008). Immigration and the imagined community in Europe and the United States. Political Studies, 56, 33–56.

Crow

M. S.

Johnson

K. A.

(2008). Race, ethnicity, and habitual-offender sentencing. Criminal Justice Policy Review, 19, 63–83.

Demuth

(2002). The effect of citizenship status on sentencing outcomes in drug cases. Federal Sentencing Reporter, 14, 271–275.

Demuth

Steffensmeier

(2004). Ethnicity effects on sentence outcomes in large urban courts: Comparison among white, black, and Hispanic defendants. Social Science Quarterly, 85, 994–1011.

Espenshade

T. J.

(1995). Unauthorized immigration to the United States. Annual Review of Sociology, 21, 195–216.

Executive Office of the President of the United States. (2014). National drug control strategy, 2014. Washington, DC: Office of National Drug Control Policy.

10.

Feldmeyer

Ulmer

J. T.

(2011). Racial/Ethnic threat and federal sentencing. Journal of Research in Crime and Delinquency, 48, 238–270.

11.

Frantzen

Worley

(2008). Inter-ethnic variation in sentencing for Hispanic drug offenders. Justice Policy Journal, 5, 1–25.

12.

Freeman

G. P.

Birrell

(2001). Divergent paths of immigration politics in the United States and Australia. Population and Development Review, 27, 525–551.

13.

Frenzel

E. D.

Ball

J. D.

(2007). Effects of individual characteristics on plea negotiations under sentencing guidelines. Journal of Ethnicity in Criminal Justice, 5, 59–82.

14.

Fry

(2008). Latino settlement in the new century. Washington, DC: Pew Hispanic Center. Retrieved from http://pewhispanic.org/files/reports/96.pdf

15.

Hagan

Levi

Dinovitzer

(2008). The symbolic violence of the crime-immigration nexus: Migrant mythologies in the Americas. Criminology and Public Policy, 7, 95–112.

16.

Hagan

Palloni

(1999). Sociological criminology and the mythology of Hispanic immigration and crime. Social Problems, 46, 617–632.

17.

Hartley

R. D.

Armendariz

L. F.

(2011). Border justice? Sentencing federal narcotics offenders in southwest border districts: A focus on citizenship status. Journal of Contemporary Criminal Justice, 27, 43–62.

18.

Huntington

S. P.

(2004). The Hispanic challenge. Foreign Policy, pp. 30–45.

19.

Jaret

(1999). Troubled by newcomers: Anti-immigrant attitudes and action during two eras of mass immigration to the United States. Journal of American Ethnic History, 18, 9–39.

20.

Johnson

B. D.

(2006). The multilevel context of criminal sentencing: Integrating judge and county level influences in the study of courtroom decision making. Criminology, 44, 259–298.

21.

Kautt

Spohn

(2002). Crack-ing down on black drug offenders? Testing for interactions among offenders’ race, drug type, and sentencing strategy in federal drug sentences. Justice Quarterly, 19, 1–35.

22.

Keogan

(2002). A sense of place: The politics of immigration and the symbolic construction of identity in Southern California and the New York metropolitan area. Sociological Forum, 17, 223–253.

23.

Larsen

L. J.

(2004). The foreign-born population in the United States: 2003. Washington, DC: U.S. Census Bureau.

24.

Light

M. T.

(2014). The new face of legal inequality: Noncitizens and the long-term trends in sentencing disparities across U.S. district courts, 1992-2009. Law and Society Review, 48, 447–478.

25.

Light

M. T.

Massoglia

King

R. D.

(2014). Citizenship and punishment: The salience of national membership in U.S. criminal courts. American Sociological Review, 79, 825–847.

26.

Logue

M. A.

(2009). The price of being Mexican: Sentencing disparities between noncitizen Mexican and non-Mexican Latinos in the federal courts. Hispanic Journal of Behavioral Sciences, 31, 423–445.

27.

Logue

M. A.

(2011). Downward departures in U.S. federal courts: Do family ties, sex, and race/ethnicity matter? Ethnic and Racial Studies, 34, 683–706.

28.

Morin

J. L.

(2008). Latinas/os and US prisons: Trends and challenges. Latino Studies, 6, 11–34.

29.

Paternoster

Brame

Mazerolle

Piquero

(1998). Using the correct statistical test for equality of regression coefficients. Criminology, 36, 859–866.

30.

Perez

E. O.

(2010). Explicit evidence on the import of implicit attitudes: The IAT and immigration policy judgments. Political Behavior, 32, 517–545.

31.

Reedt

Semisch

Blackwell

(2013, November). Effective use of federal sentencing data. Paper presented at the meeting of the American Society of Criminology, Atlanta, GA.

32.

Rumbaut

R. G.

Ewing

W. A.

(2007). The myth of immigrant criminality. Retrieved from http://borderbattles.ssrc.org/Rumbault_Ewing/

33.

Sayad

(2004). The suffering of the immigrant. Cambridge, England: Polity.

34.

Scott

(2011). U.S. v. Booker: System failure or system fix? University of Pennsylvania Law Review, 160, 195–208.

35.

Steffensmeier

Demuth

(2000). Ethnicity and sentencing outcomes in U.S. federal courts: Who is punished more harshly? American Sociological Review, 65, 705–729.

36.

Steffensmeier

Demuth

(2001). Ethnicity and judges’ sentencing decision: Hispanic-Black-White comparisons. Criminology, 39, 145–178.

37.

Steffensmeier

Ulmer

Kramer

(1998). The interaction of race, gender, and age in criminal sentencing: The punishment cost of being young, black, and male. Criminology, 36, 763–798.

38.

Sutton

J. R.

(2013). Structural bias in the sentencing of felony defendants. Social Science Research, 42, 1207–1221.

39.

U.S. Sentencing Commission. (2014a). Guidelines manual. Washington, DC: Author.

40.

U.S. Sentencing Commission. (2014b). Sourcebook of federal sentencing statistics. Washington, DC: Author.

41.

U.S. v. Booker, 543 U.S. 220 (2005).

42.

U.S. v. Borrero-Isaza, 887 F.2d 1349, 1352 (8th Cir. 1989).

43.

U.S. v. Mistretta, 488 U.S. 361 (1989).

44.

U.S. v. Onwuemene, 933 F.2d 650, 651 (8th Cir. 1991).

45.

Wolfe

S. E.

Pyrooz

D. C.

Spohn

C. C.

(2011). Unraveling the effect of offender citizenship status on federal sentencing outcomes. Social Science Research, 40, 349–362.

46.

Zarriello

Gray

D. H.

(2014). The war against public enemy number one: U.S. miscalculations in Mexico’s War on Drugs. Global Security Studies, 5, 1–11.