Warriors Wanted: The Performance of Immigrants in the US Army

Abstract

The US Army offers English-language instruction and socio-cultural training to foreign-born personnel, and current US law allows some immigrants to apply for expedited citizenship through military service. The US Army, thus, offers a compelling context in which to explore how such institutional factors might facilitate immigrant incorporation, yet we know little about the experience of foreign-born soldiers because most surveys exclude active-duty personnel. Using novel data obtained from the US Department of Defense that are not available to the public, this research note describes the integrative nature of the US Army, and contrasts foreign-born and native-born soldiers in relation to what we know about selectivity and immigrant job outcomes elsewhere. We examine rank, promotion likelihood, and retention of newly enlisted citizen and noncitizen immigrant soldiers compared to their native-born counterparts who joined the US Army between 2002 and 2009. We show that immigrants perform equally well or better than native-born soldiers.

Keywords

military immigrant incorporation job outcomes

Introduction

Immigrants have served in the US military since the Revolutionary War (Jacobs and Hayes 1981). Special naturalization provisions for immigrant service members and their dependents date back to the US Civil War, and every subsequent war has brought about special enactments for naturalizing both legal and undocumented immigrants through military service (Goring 2000). Nonetheless, we know little about immigrant soldiers’ experiences because most surveys exclude active-duty personnel (National Research Council 2012). This gap in knowledge is problematic, as understanding the experiences of immigrant service members can also shed light on the process of immigrant incorporation elsewhere, especially with regard to institutional-level practices.

Evidence from Germany and Sweden show that workplace contexts shape immigrant incorporation (Avent-Holt, Hällsten and Cort 2019; Melzer et al. 2018; Tomaskovic-Devey, Hällsten, and Avent-Holt 2015), though obstacles beyond the workplace, such as increased immigration enforcement and weak social services, have been found to impede immigrant incorporation in the United States (Gandara and Contreras 2009; Gonzales and Ruiz 2014). In theory, workplace fixed-effects models account for the influence of residential segregation, but immigrants are more segregated in the United States than in Europe (Alba and Foner 2014). Non-workplace factors in the United States, thus, make it challenging to understand organizational variation in immigrant incorporation, but there is one institution that counteracts some of these factors hindering immigrant incorporation — the US military.

Unlike other employers, the US military possesses the ability to impose a set of laws and principles for institutional aims, requiring soldiers to work and live in racially and class-integrated environments (Lundquist 2004). Defense English Language Program (DELP) and Field Studies Program (FSP) expect foreign-born soldiers to achieve their target English comprehension level and acquire socio-cultural competence (Department of the Army [DA] 2015), and current US law allows some immigrants to apply for expedited citizenship through military service (McIntosh and Sayala 2011; Chishti, Rose and Yale-Loehr 2019). These institutional practices likely impact the processes of immigrant incorporation and immigrant job outcomes.

Using data from the Department of Defense (DoD), this research note describes the US Army’s unique context and documents immigrant job performance within it. We measure rank, promotion likelihood, and retention of newly enlisted citizen and noncitizen immigrant soldiers compared to their native-born counterparts who enlisted in the Army between 2002 and 2009. We find that citizen- and noncitizen-immigrant soldiers, on average, were promoted at equal or higher rates than native-born soldiers with otherwise-similar characteristics. Noncitizen-immigrant soldiers were more likely than native-born soldiers to become Sergeant, a mid-enlisted rank that requires rigorous reviews. Both citizen- and noncitizen-immigrant soldiers were less likely to be terminated for poor performance or misconduct than native-born soldiers. Our findings show that immigrant soldiers performed equally well, often better, than native-born soldiers in the US Army.

The US Army Context and Its Importance

The US military provides an important institutional context in which to study the process of immigrant integration because its strategic focus is to build and maintain unit cohesion. Cohesion was first identified as key to military stability and efficacy upon studying the tenacity of German prisoners of war during World War II (Shils and Janowitz 1948). By contrast, the US military’s failure in Vietnam was linked to its lack of troop cohesion (Savage and Gabriel 1976), leading to the conclusion that the success and survival of a force composed of diverse individuals depended upon its ability to form trusting bonds (Cox 1995). Unlike teamwork, unit cohesion stresses building a strong foundation of group identity and meaningful bonds in order for soldiers to accomplish goals in high-pressure scenarios and avoid deadly consequences (Van Epps 2008).

The US Army provides an appropriate starting point for understanding immigrant integration because it was one of the first US institutions to racially integrate and has over time crafted a “race-savvy, not race-blind” environment based on meritocracy (Moskos and Butler 1996, 71). The US Army works to reduce racial antipathy by acculturating white soldiers to become more attuned to nonwhite experiences, and status around military rank subsumes other social categories that typically matter in civilian society (Moskos 1983). Consequently, many inequalities found in the civilian world are significantly reduced or eliminated in the US military (Lundquist 2004; MacLean and Elder 2007), and minority veterans have higher socioeconomic status than their nonveteran counterparts (Xie 1992; Angrist 1998). Particularly relevant to immigrant integration, the US Army provides holistic programming to encourage acculturation among foreign-born soldiers. As part of training, foreign-born soldiers take classes, with subjects including the role of labor unions and minority programs, religious and social diversity, and government and political processes, and DELP offers English instruction and intensive technical language training in their field of specialization (DA 2015).

To be eligible to enlist in the US Army, applicants must be US citizens or permanent residents, be at least 17 years old, have a high school diploma or the equivalent, be in good health, and exhibit good moral character (10 USC §504; 32 CFR § 66.6). People with a felony record or a major misconduct offense are ineligible to enlist, but a waiver can be issued in meritorious cases (Lundquist, Pager, and Strader 2018). Non-citizens must provide proof of permanent residency, and foreign education credentials must be validated by a state Board of Education or a degree-granting college/university in the United States (McIntosh and Sayala 2011). Eligible candidates are screened using the Armed Forces Qualification Test (AFQT), which is designed to measure verbal, arithmetic, and analytical reasoning skills, and the chances of enlistment are very slim for those scoring below the 30th percentile (Armor and Gilroy 2010). Although disqualification is based on AFQT scores, the Army can administer the English Comprehension Level Test (ECLT) at any point during recruitment to gauge English proficiency or admission into DELP (Asch et al. 2011).

Once enlisted, permanent residents can apply for expedited naturalization (INA §328). Importantly for this study, the US Army recruits more noncitizens than the Navy, Marine Corps, and Air Force, and has the shortest average time-to-citizenship (McIntosh and Sayala 2011) — two characteristics that are pertinent to the discussion of immigrant integration. Civilian permanent residents must meet a five-year residency requirement to apply for citizenship, but a 2002 Executive Order authorized immediate application during hostilities, and the 2004 National Defense Authorization Act reduced the peacetime waiting period for citizenship application from three to one year and waved all filing fees (McIntosh and Sayala 2011). Expedited naturalization is vital for assigning immigrants to critical military positions, such as intelligence, nuclear, or Special Operations Forces, that require citizenship for security clearances and for attracting those with in-demand skills into the US military (McIntosh and Sayala 2011). Naturalized service members gain the ability to sponsor the migration of family members living abroad (Aptekar 2015), and surviving family members can apply for citizenship without residence or physical presence requirements in cases where posthumous citizenship is granted (INA §319(d)). These opportunities are extended to nonpermanent residents during wartime (Chishti et al. 2019).

These practices set the US military apart from other workplaces, but integration efforts across the military branches are still evolving and have a mixed record. Women’s integration into combat and leadership positions has been slow (Defense Advisory Committee on Women in the Services [DACOWITS] 2017), and combat casualties are higher among Hispanics (Fischer 2009). Black service members have a lower combat mortality rate than other enlistees in the post-Vietnam era (Gifford 2005; MacLean 2011), but the same cannot be said for earlier wars (Burk and Espinoza 2012). Moreover, the military naturalization process is bureaucratic, and citizenship is not guaranteed (Aptekar 2018). Denial rates are difficult to estimate, but the ratio of approval to denial is about 4.7 to 1 (U.S. Citizenship and Immigration Services 2019). Citizenship gained through military service can be revoked in cases of discharge under other than honorable conditions (INA §1439). We may ask, then, how do immigrants fare in the US Army, the military’s largest branch?

Predicting Immigrant Outcomes in the Us Army

Dominguez-Villegas (2019) finds an income premium for military service in the United States, with immigrant veterans earning higher wages than nonveteran immigrants, and Leal (2003) shows that Hispanic veterans speak more English at home and have a more diverse group of friends than do nonveteran Hispanics. These findings based on veterans and institutional-level military practices suggest that the US Army may facilitate immigrant incorporation and that active-duty immigrants will perform as well as their native-born counterparts. That said, we expect citizenship to mediate this process, giving naturalized immigrants an edge over noncitizen immigrants because naturalization has been found to accelerate wage growth among civilian men (Bratsberg et al. 2002), and because faster access to citizenship improves labor-market attachment and the language skills of immigrant women elsewhere (Gathmann and Keller 2018).

Another element to consider is that immigrants are positively selected from the population for US military service. As the pool of eligible native-born recruits dwindles due to poor education, widespread obesity, and criminal convictions (Mission Readiness 2009), immigrants constitute an increasingly important source of recruits for the US military (Chishti et al. 2019). Overall, immigrants tend to be younger and in better health (Guillermina and Massey 2004; Goldman et al. 2006), and they are less likely to commit violent crimes (Ousey and Kubrin 2009) than their native-born counterparts. Furthermore, immigrants in the United States are often better educated than their nonmigrating counterparts in their home countries (Feliciano 2005; Feliciano and Lanuza 2017). We lack data on the year of arrival to the United States for foreign-born service members, but the median age at which they joined was 20 years, based on our sample. Thus, many immigrant soldiers are 1.5 generation, whose earlier arrival to the United States correlates with greater English proficiency and higher education (Myers, Gao and Emeka 2009; Feliciano 2005; Feliciano and Lanuza 2017).

In the US civilian sector, immigrants earn less than natives initially, but their wages reach parity as immigrants adjust to the US labor market (Chiswick 1978). In the case of black immigrants from the Caribbean, studies find that they outperform native-born blacks in the labor market due to selective migration, suggesting that tangible and intangible characteristics that propelled immigrants to migrate explain this advantage (Model 2008; Hamilton 2014; Ifatunji 2017). Foreign-born blacks from the Caribbean possess greater human capital than native-born blacks (Model 2008; Hamilton 2014), but they do not differ much in terms of gumption and personal mastery (Ifatunji 2017). Selection into the US military is not random, since recruits are carefully screened, and evidence shows that the characteristics of those who have served differ from those who have not (Burk and Espinoza 2012; MacLean and Elder 2007). The most noticeable difference has been socioeconomic background, where those from disadvantaged backgrounds enlist at higher rates (Kleykamp 2006; Lutz 2008).

Based on the available literature, we expect that the initial immigrant–native earnings gap should be eliminated or reduced significantly due to the US Army’s integrative nature and the possibility should be considered that, with time, immigrant soldiers outperform their native-born counterparts on the basis of selectivity. This research note serves as a first step to documenting observable differences between the profiles of foreign-born and native-born soldiers and job performance, adding this population to the body of literature on immigrant incorporation and selectivity.

Data and Methods

We obtained our data by making a Freedom of Information Act (FOIA) request to the DoD. The FOIA establishes the public’s right to access US federal agency records (5 USC §552), and the DoD provided administrative records pertaining to all new active-duty enlistees who joined between 2002 and 2009. There are two paths to joining the US military: either enlisting or commissioning as an officer to be in a management position. The latter, however, requires citizenship and a college degree. Therefore, to estimate the performance of noncitizen and naturalized immigrants in the US Army, we limit our analysis to enlisted service members who lacked a college degree at the time of first enlistment. The unit of analysis is the individual soldier, and we analyzed 494,681 records. Immigrant performance is estimated by citizenship (citizen immigrants = 1, noncitizen immigrants = 2), using native-born citizens as the reference group (= 0). Because citizenship status was only updated when a promotion or separation record was entered, our data cannot estimate time-to-citizenship accurately; thus, we rely on citizenship status at the time of enlistment.

Job performance is measured in three ways to paint a comprehensive picture: (a) promotion rates, (b) likelihood of reaching Sergeant rank, and (c) early termination due to poor performance. Army ranks for enlisted personnel range from Private (E-1) through Sergeant Major (E-9). Without allowances, special incentives, and benefits, in 2009, Privates who had served at least four months earned $1,399.50 monthly, and Sergeant Majors who had served over 10 years earned at least $4,420.50 monthly (Defense Finance and Accounting Service n.d.). Military compensation is competitive compared to civilian pay because the total compensation paid to enlistees was around the 84th percentile of wages earned by civilians with comparable education in 2009 (Hosek et al. 2018).

We measure promotion rates using the number of ranks increased, which is the difference between the earliest and the latest ranks on record; those who left without any promotion receive a zero for this measure. Next, we investigate if immigrants have equal access to opportunities and whether citizenship mediates this process by identifying soldiers who reached Sergeant rank (E-5). There are no quotas for promotion in the Army through Specialist/Corporal (E-4), but promotions become more merit-based at E-5 and above (Malone 2014). Lastly, we assessed underperformance that cannot be captured through the two promotion measures by distinguishing those who were terminated for poor performance or misconduct from those who were still serving or departed for other reasons. Poor performance and misconduct terminations are kept in administrative record and negatively impact eligibility for veterans’ benefits (DA 2017).

Table 1.

Descriptive Statistics of New Army Enlistees by Citizenship Status (N = 494,681.)

	Native-BornCitizens		CitizenImmigrants		Noncitizen Immigrants
	458,715 (92.7%)		18,466 (3.7%)		17,500 (3.5%)
Job performance indicators	Mean (SD)	Sig 1	Mean (SD)	Sig 2	Mean (SD)	Sig 3
Number of ranks promoted	1.59 (1.36)		1.59 (1.35)	***	1.90 (1.41)	***
Reached Sergeant rank	16.8%	***	18.7%	***	23.8%	***
Terminated for poor performance or misconduct	20.8%	***	16.4%	***	12.3%	***
Continuous independent variables
Months of active service	37.9 (25.88)	***	39.6 (25.79)	***	43.7 (26.7)	***
Age at entry to the Army	20.6 (3.4)	***	21.2 (4.1)	***	22.5 (4.9)	***
Starting pay rank	1.71 (0.82)	***	1.85 (0.87)	***	1.76 (0.86)	***
AFQT percentile	58.6 (19.1)	*	58.3 (19.9)	***	50.5 (17.9)	***
Categorical independent variables
Participated in delayed entry program (DEP)	27.7%	**	26.8%	***	25.6%	*
Education
Below HS	1.4%	***	1.0%	***	0.9%
GED equivalent	20.6%	***	16.3%	***	10.9%	***
HS graduate	72.9%	***	75.6%	***	79.3%	***
Some college	5.2%	***	7.1%	***	8.9%	***
Race/Ethnicity
NH (Non-Hispanic) White	67.6%	***	44.0%	***	15.6%	***
NH Black/African American	13.9%	*	14.5%	***	20.1%	***
NH Asian/Asian American	1.0%	***	11.5%	***	14.2%	***
NH American Indian/Alaska Native	1.1%	***	0.6%	***	0.4%	**
NH Native Hawaiian/Other Pacific Islander	0.6%	***	2.6%	***	2.9%
NH mixed race	0.5%	***	2.0%	***	2.5%	**
Hispanic	10.6%	***	17.6%	***	39.7%	***
Missing race/ethnicity	4.7%	***	7.1%		4.6%	***
Female	16.7%	***	18.7%	***	21.6%	***
Married	13.9%	***	16.0%	***	24.1%	***
Year of entry
2002	13.1%	***	12.0%	***	16.37%	***
2003	12.8%	*	12.2%	***	16.25%	***
2004	13.5%		13.8%	***	15.40%	***
2005	11.8%		12.0%	**	11.09%	**
2006	13.0%	**	12.2%	***	10.81%	***
2007	11.5%		11.2%	***	8.89%	***
2008	12.9%		13.2%	***	10.45%	***
2009	11.5%	***	13.3%	**	10.74%	***

Notes: * p<0.05 ** p<0.01 *** p<0.001.

Sig 1 = Mean native-born vs. Mean citizen immigrants.

Sig 2 = Mean native-born vs. Mean noncitizen immigrants.

Sig 3 = Mean citizen immigrants vs. Mean noncitizen immigrants.

Table 1 provides descriptive statistics by immigrant status and reports means alongside t-test results for all variables. Native-born citizens composed 92.7 percent of enlistees, with the rest composed of immigrants, about half of whom were noncitizens at the time of enlistment. Overall, native-born and foreign-born citizens were promoted at roughly the same rate, while noncitizen immigrants significantly outperformed native-born citizens. In terms of promotion to Sergeant, citizen and noncitizen immigrants outperformed native-born citizens, whereas native-born soldiers were most likely to be terminated. Below the three performance measures in Table 1, we describe human capital, demographic, and contextual factors that shape job outcomes in the US Army.

The length of active service is measured by month, being mindful that the average active-duty contract lasts around four years. We find that immigrants, on average, served longer than native-born soldiers. The age at enlistment and starting rank control for the trajectory of future rank and promotion because having relevant prior experience and special skills lead to a higher starting rank. Noncitizen immigrants were the oldest group and exhibited the largest variation in age at entry. Citizen immigrants started with the highest average rank, followed by noncitizen immigrants, and native-born soldiers had the lowest starting rank. Overall, roughly half the new recruits started from the lowest rank, and about one-third joined at the rank of E-2. A slightly higher share of immigrants started at E-3 or above than did native-born soldiers. Therefore, immigrants were more likely to join with relevant training and skills, indicating that they were positively selected into the US Army relative to native-born soldiers.

The AFQT and Delayed Entry Program (DEP) status are included to assess job readiness. We mentioned earlier that applicants scoring below the 30th percentile on the AFQT are typically disqualified, and our data show that only 3 percent of new enlistees scored below this threshold. Although the AFQT’s validity and neutrality have been questioned by some (Roberts and Skinner 1996), we note that noncitizen immigrants on average scored eight percentile lower than native-born recruits. The gap is smaller between native-born and foreign-born citizen soldiers. DEP delays the start date so that the recruit can effectively prepare for service by finishing school, meeting physical standards, and so forth. We identify those who participated in DEP against those who did not as the reference. A smaller portion of immigrants entering through DEP suggests their greater readiness to serve relative to native-born soldiers.

Educational attainment upon enlistment accounts for differences in human capital broadly, though it likely underestimates education’s impact on immigrant performance because foreign education credentials are often undervalued (Feliciano and Lanuza 2017). We treat education as a time-invariant factor influencing pre-military socioeconomic status at the time of entry. Education is measured as a categorical variable with high-school graduates as the reference compared to high-school dropouts, those with GED equivalent credentials, and those with some college education. Our data show that immigrants were less likely to be high-school dropouts or GED holders than native-born soldiers, even though immigrants scored lower on the AFQT. Therefore, it is likely that immigrants are positively selected into the US Army.

Race/ethnicity, gender, and marital status have been found to impact employment outcomes in the US Army (Burk and Espinoza 2012; DACOWITS 2017; Strader 2018). We combined race and ethnicity that were reported separately by the DoD, assuming that cases with race recorded without Hispanic ethnicity information were likely non-Hispanic. Non-Hispanic white is the reference, and we identified non-Hispanic black/African American, non-Hispanic Asian, non-Hispanic American Indian and Alaska Native (AIAN), non-Hispanic Native Hawaiian and/or other Pacific Islanders (NHOPI), non-Hispanic mixed race, Hispanic, and missing/other. Racial/ethnic minorities were overrepresented among immigrants, especially noncitizens. Gender and marital status are coded as binary variables, using male and those who were not married at the start of service as the references. Women were overrepresented among noncitizens, and immigrants were more likely to be married at the time of enlistment than native-born recruits. National contexts surrounding the supply and demand for soldiers are captured using categorical year of enlistment. More noncitizen immigrants enlisted in the early years of the Iraq War (2002 through 2004) than did citizen recruits. The number of noncitizens began to shrink around 2006, likely due to the Iraq War troop surge and a robust civilian sector, but their share increased again during the Great Recession (2008 and 2009).

We used Ordinary Least Squares (OLS) to estimate promotion rates and logistic regression to examine the likelihood of reaching Sergeant rank. For predicting promotion outcomes, we included a binary indicator of early separation, months of service and its squared term, to account for job tenure and its curvilinear effect on promotion. In each analysis, we tested for interactions between immigrant status and gender, and found that gender moderated immigrant performance. We expand on this finding in the next section. Using Kaplan-Meier estimates, we plotted continuous risk of termination and compared baseline survival trends between immigrant and native-born soldiers. We measured survival time in months from the start of service, which is right censored. Those who left within the same month of entry for any reason were excluded from the analysis because the survival time is zero. We employed Cox Proportional-Hazards models to estimate the time-dependent risk of termination within a multivariate framework.

As with any dataset, there are missing variables that could have provided more insight into immigrant job performance in the US Army. We lack data on prior occupations or earnings, and the data do not indicate location assignments or deployment; these factors may influence one’s decision to join the Army, as well as one’s overall quality of performance and subsequent promotions. Origin countries were not reported, but the most common origin countries, according to Barry (2013), are the Philippines (comprising 22.8% of active-duty immigrants), Mexico (9.5%), Jamaica (4.7%), Korea (3.1%), and the Dominican Republic (2.5%). Among veterans, Zong and Batalova (2016) find Mexico and the Philippines to be the top two origins, representing 16 percent and 13 percent of the 511,000 immigrant veterans in 2016. We expand on these data-related limitations in the conclusion.

Results

Table 2 provides estimates from models predicting promotion outcomes. In terms of promotion rates, we find that immigrants not only performed equally but also progressed farther than native-born soldiers with similar observable traits. Based on the available literature, we attribute equal performance of foreign- and native-born soldiers to the US Army’s integrative practices, and we deduce that positive selectivity propels immigrants to outperform native-born soldiers. Interestingly, this advantage was greater for noncitizens than for citizen-immigrants, with the former promoted roughly 0.08 ranks more on average than native-born soldiers. This finding contradicts our prediction based on what is known about civilian experiences, where citizenship improves, rather than worsens, foreign-born workers’ labor-market outcomes.

Table 2.

Predicting Promotion Outcomes (N = 494,681.)

	OLS Models PredictingNumber of Ranks Promoted				Logistic Models PredictingPromotion to Sergeant
Citizenship status(Omitted: Native-born citizens)	Without Female Interactions		With Female Interactions		Without Female Interactions		With Female Interactions
Citizenship status(Omitted: Native-born citizens)	Coef.	SE	Coef.	SE	OR	SE	OR	SE
Citizen immigrants	.0108*	(.0051)	.0044	(.0056)	1.0027	(.0263)	0.9718	(.0277)
Citizen immigrants × female	--	--	.0340**	(.0128)	--	--	1.2222**	(.0865)
noncitizen immigrants	.0833***	(.0053)	.0795***	(.0060)	1.2927***	(.0334)	1.2280***	(.0353)
Noncitizen immigrants × female	--	--	.0178	(.0125)	--	--	1.2854***	(.0782)
Continuous predictors
Months of active service	.0751***	(.0002)	.0751***	(.0002)	1.2576***	(.0031)	1.2576***	(.0031)
Months of active service²	-.0004***	(.0000)	-.0004***	(.0000)	0.9987***	(.0000)	0.9987***	(.0000)
Age at entry to the Army	.0074***	(.0003)	.0074***	(.0003)	1.0217***	(.0018)	1.0218***	(.0018)
Starting pay rank	-.6680***	(.0012)	-.6680***	(.0012)	1.4820***	(.0092)	1.4822***	(.0092)
AFQT percentile	.0014***	(.0001)	.0014***	(.0001)	1.0097***	(.0003)	1.0097***	(.0003)
Categorical predictors
DEP	.0104***	(.0022)	.0104***	(.0022)	1.0353**	(.0116)	1.0353**	(.0116)
Education (Omitted: HS graduate)
Below HS	-.0971***	(.0083)	-.0972***	(.0083)	0.7441***	(.0509)	0.7435***	(.0509)
GED equivalent	-.0755***	(.0025)	-.0756***	(.0025)	0.7558***	(.0112)	0.7555***	(.0112)
Some college	.0035	(.0044)	.0035	(.0044)	1.0340	(.0233)	1.0335	(.0233)
Race/Ethnicity (Omitted: Non-Hispanic White)
NH Black	-.0428***	(.0029)	-.0426***	(.0029)	0.8645***	(.0135)	0.8665***	(.0136)
NH Asian	.0268***	(.0072)	.0270***	(.0072)	1.0418	(.0373)	1.0457	(.0375)
NH AIAN	-.0595***	(.0093)	-.0593***	(.0093)	0.7317***	(.0383)	0.7333***	(.0384)
NH NHOPI	-.0039	(.0112)	-.0038	(.0112)	0.9728	(.0623)	0.9738	(.0624)
NH mixed race	-.0032	(.0118)	-.0032	(.0118)	1.0423	(.0596)	1.0431	(.0597)
Hispanic	.0193***	(.0031)	.0193***	(.0031)	1.0913***	(.0175)	1.0922***	(.0175)
Missing race/ethnicity	.0115*	(.0045)	.0116**	(.0045)	1.0789**	(.0259)	1.0795**	(.0259)
Female	-.0270***	(.0026)	-.0292***	(.0027)	0.8221***	(.0126)	0.8011***	(.0130)
Married	.0126***	(.0029)	.0125***	(.0029)	1.1528***	(.0187)	1.1528***	(.0187)
Separated early	-.1003***	(.0030)	-.1003***	(.0030)	0.6489***	(.0103)	0.6488***	(.0103)
Year of entry (Omitted: 2002)
2003	-.1079***	(.0037)	-.1079***	(.0037)	0.7740***	(.0131)	0.7740***	(.0131)
2004	-.1588***	(.0037)	-.1588***	(.0037)	0.7115***	(.0129)	0.7114***	(.0129)
2005	-.2103***	(.0040)	-.2103***	(.0040)	0.6741***	(.0131)	0.6741***	(.0131)
2006	-.2135***	(.0040)	-.2135***	(.0040)	0.5525***	(.0114)	0.5524***	(.0114)
2007	-.1080***	(.0042)	-.1080***	(.0042)	0.2279***	(.0081)	0.2278***	(.0081)
2008	-.3192***	(.0043)	-.3192***	(.0043)	0.0083***	(.0028)	0.0083***	(.0028)
2009	-.5220***	(.0048)	-.5221***	(.0048)	0.0066***	(.0066)	0.0066***	(.0066)
Constant	.7944***	(.0086)	.7947***	(.0086)	--	--	--	--

Notes: * p<0.05 **p<0.01 *** p<0.001

Age at entry and AFQT percentile were positively associated with promotion rates, while a higher starting rank reduced promotion rates. Upon this finding, we conducted robustness checks by limiting our sample to those who began at E-1 or E-2 and then analyzed those who started at E-3 or higher separately, because starting at the lowest rank and having a higher ceiling for promotion could potentially bias promotion rates. Our earlier findings that noncitizen immigrants outperformed native-born soldiers remained consistent under both specifications, but the magnitude of the advantage was smaller when we examined those who started at E-3 or higher (results available upon request). Those who participated in DEP saw more promotions than those who did not, and non-Hispanic black and non-Hispanic AIAN soldiers experienced less promotions while Hispanic soldiers experienced slightly more promotions than non-Hispanic whites. Overall, women were promoted at a slower pace than men, but there was no gender difference among citizen immigrants, and Wald test results confirm that the difference is statistically insignificant. We also found that married soldiers saw more promotions.¹

Our analysis cannot explain why citizen immigrant women outperformed their male counterparts, but evidence from the civilian world suggests that immigrant women’s status relative to that of immigrant men tends to improve through better institutional protections provided by host countries (Itzigsohn and Giorguli-Saucedo 2005). Although sexual harassment and assault continue to plague the US military (DACOWITS 2017), Lundquist (2008) argues that black, Hispanic, and white female soldiers report higher job satisfaction compared to white men due to their perceived advantages of pay and promotional opportunity in the US military than in the civilian sector. Thus, it is possible that the US Army’s meritocratic environment facilitates the incorporation of citizen immigrant women better than that of their male counterparts. More research is needed to confirm this possibility.

Table 2 presents the results from the analyses predicting the likelihood of reaching Sergeant rank. We find no disparity between native-born and foreign-born citizens, but the odds of noncitizen immigrants being promoted to Sergeant are 29 percent higher than native-born soldiers with otherwise similar characteristics. This finding is notable because roughly one-half of US Army occupations require citizenship for security clearances, leaving noncitizens with fewer career options (McIntosh and Sayala 2011). Gender once again moderated the effect of immigrant status on promotion likelihood, and Wald test results suggest that this interaction is statistically significant. Native-born women faced 20 percent lower odds of reaching Sergeant rank compared to native-born men, but this gender disparity was felt much less among citizen immigrant women, and in the case of noncitizen immigrant women, they were significantly more likely to reach Sergeant rank than native-born men.

Using termination as the outcome variable, Figure 1 compares the baseline survival rates by immigrant status over months of service. The grey areas on either side of the risk curves indicate 95% confidence intervals. Without controlling for other covariates, the solid line in Figure 1 suggests that native-born citizens faced the highest risk of termination. Noncitizen immigrants (shown in dashes) faced the lowest risk of termination, and naturalized immigrants (shown in dash-dot line) fell between these two groups. However, these results could be an artifact of other factors, such as education level and job readiness, masking the possibility that immigrants and native-born soldiers with similar characteristics were terminated at similar rates.

Figure 1.

Kaplan-Meier Survival Estimates, Poor Performance and Misconduct Termination.

The estimates reported in Table 3 indicate that citizen and noncitizen immigrants with otherwise-similar characteristics faced lower odds of termination relative to native-born soldiers by 17 percent and 37 percent, respectively, and this difference is statistically significant (results available upon request). Increased risk is also marked by whiteness because racial/ethnic minorities, except for AIAN soldiers, were less likely to be terminated than non-Hispanic whites. High-school dropouts and those with GED equivalent credentials were predicted to face higher risks of termination than high-school graduates, which makes sense within the traditional human capital framework. However, increased risk observed among soldiers with some college education is rather baffling. Studies suggest that minorities and those with limited education enlist at higher rates and serve for longer tenures than whites and college graduates because the US military affords them more upward mobility than civilian jobs (Kleykamp 2006; Lutz 2008; Armor and Gilroy 2010). Our findings may, thus, reflect the idea that whites and those with some college education in the United States have less incentive to avoid termination in the Army, knowing they will have employment options elsewhere (Kleykamp 2006; Han 2018).

Table 3.

Cox Proportional-Hazards Models Predicting Termination (N = 493,753.)

	Without Female Interactions		With FemaleInteractions
Citizenship status (Omitted: Native-born itizens)	HR	SE	HR	SE
Citizen immigrant	0.8333***	(.0156)	0.8587***	(.0178)
Citizen immigrant × female	--	--	0.8623**	(.0407)
Noncitizen immigrant	0.6291***	(.0142)	0.6810***	(.0172)
Noncitizen immigrant × female	--	--	0.7109***	(.0387)
Continuous Predictors
Age at entry to the Army	0.9842***	(.0011)	0.9842***	(.0011)
AFQT percentile	0.9935***	(.0002)	0.9935***	(.0002)
Categorical predictors
DEP	0.9368***	(.0068)	0.9368***	(.0068)
Education (Omitted: HS graduate)
Below HS	1.8443***	(.0451)	1.8452***	(.0451)
GED equivalent	1.7744***	(.0130)	1.7749***	(.0130)
Some college	1.1126***	(.0173)	1.1130***	(.0173)
Race/Ethnicity (Omitted: Non-Hispanic white)
NH Black	0.9654**	(.0090)	0.9643**	(.0090)
NH Asian	0.7139***	(.0208)	0.7114***	(.0207)
NH AIAN	1.0690*	(.0309)	1.0675*	(.0300)
NH NHOPI	0.6624***	(.0311)	0.6620***	(.0311)
NH mixed race	0.8609***	(.0356)	0.8603***	(.0356)
Hispanic	0.7139***	(.0081)	0.7137***	(.0081)
Missing race/ethnicity	0.9550**	(.0147)	0.9547**	(.0147
Female	1.3325***	(.0111)	1.3507***	(.0116)
Married	0.9995	(.0100)	0.9996	(.0100)
Year of entry (Omitted: 2002)
2003	0.9738*	(.0114)	0.9739*	(.0114)
2004	1.0656***	(.0121)	1.0658***	(.0121)
2005	0.8245***	(.0102)	0.8244***	(.0102)
2006	0.7632***	(.0093)	0.7631***	(.0093)
2007	0.8226***	(.0105)	0.8225***	(.0105)
2008	0.8805***	(.0113)	0.8806***	(.0113)
2009	1.0915***	(.0153)	1.0914***	(.0153)

Notes: * p<0.05 **p<0.01 *** p<0.001

However, that does not mean that non-white soldiers and recruits with limited education are not competitive in the civilian sector. The demand for recruits is counter-cyclical to the economy, with enlistments rising when unemployment rates are high and declining when unemployment is low (Kleykamp 2006). This trend, coupled with downsizing of the military since the late 1980s, means that military service opportunity has declined and recruiters can be more selective about recruit quality in terms of education and AFQT scores, especially when unemployment is high (Armor and Gilroy 2010; Han 2018). Nonetheless, more research is needed to explain why soldiers with some college education faced higher risk of termination than did high-school graduates. Lastly, we assessed whether gender moderated the effect of immigrant status on risk of termination. Although native-born women faced 35 percent higher odds of termination relative to native-born men, citizen and noncitizen immigrant women were predicted to face lower odds of termination. Henceforth, immigrant women were much less likely to be terminated relative to native-born men. We discuss the implications of this and other findings for the understanding of immigrant incorporation within and beyond the US Army in the last section.

Discussion and Conclusions

In this research note, we have described the US Army’s unique institutional context and assessed how immigrants perform under such conditions. Our results show that in the 2000s, immigrants, especially those lacking citizenship, were promoted at equal or higher rates and experienced more successful advancement through the ranks compared to their native-born counterparts. We also found that immigrant service members were less likely to be terminated for poor performance and misconduct than native-born soldiers. What are the implications of these findings?

Our results are likely due to the unusually integrative nature of the US Army, which facilitates immigrant integration more quickly than do other contexts of reception. Although institutional practices may account for the absence of an immigrant-native performance gap, the US Army context alone may not explain immigrants’ superior performance. Based on what we know about some immigrants’ superior performance in the civilian sector (Model 2008; Hamilton 2014; Ifatunji 2017), positive selection may also have been important. However, contradicting the evidence from the civilian sector (Bratsberg et al. 2002; Gathmann and Keller 2018), noncitizen immigrants in the US Army outperformed naturalized immigrants. Perhaps our models did not adequately account for selectivity of noncitizen immigrants relative to naturalized immigrants as our data lack information on pre-migration health and socioeconomic status, pre-enlistment earnings, years since migration, and English fluency. There are also signs that noncitizen immigrants in the US Army may be highly selected on unobserved characteristics because they scored lower on the AFQT, yet were better educated and less likely to join through the DEP than their naturalized counterparts. Future research on immigrant incorporation and selectivity should pay more attention to institutional factors because they allow us to highlight the mechanisms influencing immigrant job outcomes.

An alternative explanation is that noncitizens are incentivized to overperform in the US Army because immigrants who naturalize through military service have no choice but to serve honorably for at least five years, lest their citizenship be revoked (INA §328(f), §329(c)). These stipulations raise potential ethical concerns about the expedited naturalization program, given that military service carries the risk of permanent injury, post-traumatic stress disorder, and even death (Aptekar 2015). Indoctrination into military culture can also be detrimental for veterans’ reintegration into civilian life, causing stress and mental health issues (Coll, Weiss, and Yarvis 2011). Because current US law permits expedited naturalization of immigrants through military service, it is important that future research investigate the potential negative consequences of military acculturation and assimilation. At a minimum, the superior performance of immigrants highlighted within this research note challenges any skepticism regarding the trustworthiness (Krikorian 2003) and perceived quality of immigrant soldiers (Stein 2018) and should encourage US policymakers who oppose the enlistment of undocumented immigrants to reconsider their position (161 Cong. Rec. H3163-H3164, 2015; H.R. 5710, 2016).

And how do we make sense of gender’s moderating effect on immigrants’ job performance in the US Army? Evidence from the European Social Survey suggests that migrant women adapt to egalitarian attitudes of the host country more straightforwardly than do migrant men (Röder and Muühlau 2014). If acculturation processes differ by gender, it is possible that immigrant women benefit more from the US Army’s integrative practices than immigrant men, and excel in the meritocratic environment. Citing selectivity as a possible explanation, studies in the United States have shown that West Indian women have higher labor force participation rates and earnings than native-born white women (Model 2008) and that first-generation English-speaking Afro-Caribbean women fare better in the labor market than native-born black women (Ifatunji 2017). Considering that the US Army is a masculine institution, yet the female share is higher among foreign-born enlistees, especially noncitizens, it is possible that immigrant women are highly selected on unobserved characteristics relative to native-born women who serve. Data limitations prevented us from being able to separate institutional effects from selection effects, but we can surmise that our results reflect some combination of both. More studies are needed to tease out gender’s effect on immigrant incorporation and selectivity across different institutional settings.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

ORCID iDs

Eiko Strader

Jennifer Lundquist

Notes

References

Alba

Foner

. 2014. “Comparing Immigrant Integration in North America and Western Europe: How Much Do the Grand Narratives Tell Us?” International Migration Review 48(1): 263–91.

Angrist

J. D

. 1998. “Estimating the Labor Market Impact of Voluntary Military Service using Social Security Data on Military Applicants.” Econometrica 66(2): 249–288

Aptekar

2015. The Road to Citizenship: What Naturalization Means for Immigrants and the United States. New Brunswick, NJ: Rutgers University Press

Aptekar

2018. “Citizenship Acquisition among Immigrant Members of the US Military.” Paper presented at the American Sociological Association meetings in Philadelphia, PA.

Armor

D. J.

Gilroy

C. L.

. 2010. “Changing Minority Representation in the U.S. Military.” Armed Forces & Society 36(2): 223–46.

Asch

B. J.

Buck

Klerman

J. A.

Kleykamp

Loughran

D. S.

. 2009. Military Enlistment of Hispanic Youth: Obstacles and Opportunities. Santa Monica, CA: RAND National Defense Research Institute.

Avent-Holt

Hällsten

Cort

. 2019. “Occupational Status and Organizations: Variation in Occupational Hierarchies across Swedish Workplaces”. Research in Social Stratification and Mobility, 100423.

Barry

C. N.

2013. New Americans in our Nation’s Military: A Proud Tradition and Hopeful Future. Washington DC: Center for American Progress.

Bratsberg

Ragan

J. F.

Jr Nasir

Z. M.

. 2002. “The Effect of Naturalization on Wage Growth: A Panel Study of Young Male Immigrants.” Journal of Labor Economics 20(3): 568–597.

10.

Burk

Espinoza

. 2012. “Race Relations Within the US Military.” Annual Review of Sociology 38(1): 401–22.

11.

Chishti

Rose

Yale-Loehr

. 2019. “Noncitizens in the U.S. Military.” Policy Brief. Washington, DC: Migration Policy Institute.

12.

Chiswick

B. R.

1978. “The Effect of Americanization on the Earnings of Foreign--Born Men.” Journal of Political Economy 86(5): 897–921.

13.

Coll

J. E.

Weiss

E. L.

Yarvis

J. S.

. 2011. “No One Leaves Unchanged: Insights for Civilian Mental Health Care Professionals Into the Military Experience and Culture.” Social Work in Health Care 50(7): 487–500.

14.

Cox

1995. Unit Cohesion and Morale in Combat: Survival in a Culturally and Racially Heterogeneous Environment. Fort Leavenworth, Kansas: United States Army Command and General Staff College.

15.

Defense Finance and Accounting Service. n.d. “Archived Military Pay Tables.” Accessed January 2, 2020, https://www.dfas.mil/militarymembers/payentitlements/Pay-Tables/PayTableArchives.html

16.

Defense Advisory Committee on Women in the Services (DACOWITS). 2017. “2017 Annual Report.” Alexandria, VA: Defense Advisory Committee on Women in the Services.

17.

Department of the Army. 2015, January 14. Army Regulation 621-8 Managing the Defense English Language Program. Washington, DC: Department of the Army.

18.

Department of the Army. 2017, November 7. Army Regulation 135-178 Enlisted Administrative Separations. November 7, 2017. Washington, DC: Department of the Army.

19.

Dominguez-Villegas

2019. “Does Serving in the Military Help Immigrants’ Socioeconomic Incorporation?”. Working Paper. University of Massachusetts, Amherst.

20.

Feliciano

2005. “Does Selective Migration Matter? Explaining Ethnic Disparities in Educational Attainment among Immigrants’ Children.” International Migration Review 39(4): 841–871.

21.

Feliciano

Lanuza

Y. R.

. 2017. “An Immigrant Paradox? Contextual Attainment and Intergenerational Educational Mobility.” American Sociological Review 82(1): 211–241.

22.

Fischer

2009. “United States Military Casualty Statistics: Operation Iraqi Freedom and Operation Enduring Freedom.” CRS Report for Congress No. RS22452. Washington, DC: Congressional Research Service.

23.

Gandara

Contreras

. 2009. The Latino Education Crisis: The Consequences of Failed Social Policies. Cambridge, MA: Harvard University Press.

24.

Gathmann

Keller

. 2018. “Access to Citizenship and the Economic Assimilation of Immigrants.” The Economic Journal 128(616): 3141–3181.

25.

Gifford

2005. “Combat Casualties and Race: What Can We Learn from the 2003–2004 Iraq Conflict?” Armed Forces & Society 31(2): 201–225.

26.

Goldman

Kimbro

R. T.

Turra

C. M.

Pebley

A. R.

. 2006. “Socioeconomic Gradients in Health for White and Mexican-Origin Populations.” American Journal of Public Health 96(12): 2186–2193.

27.

Gonzales

R. G.

Ruiz

A. G.

. 2014. “Dreaming beyond the Fields: Undocumented Youth, Rural Realities and a Constellation of Disadvantage.” Latino Studies 12(2): 194–216.

28.

Goring

. 2000. “In Service to America: Naturalization of Undocumented Alien Veterans,”

29.

Guillermina

Massey

D. S.

. 2004. “Immigrant Health: Selectivity and Acculturation.” IFS Working Papers. Institute for Fiscal Studies (IFS).

30.

Hamilton

T. G.

2014. “Selection, Language Heritage, and the Earnings Trajectories of Black Immigrants in the United States.” Demography 51(3): 975–1002.

31.

Han

2018. “Who Goes to College, Military, Prison, or Long-Term Unemployment? Racialized School-to-Labor Market Transitions Among American Men.” Population Research and Policy Review 37(4): 615–640.

32.

Hosek

J. R.

Asch

B. J.

Mattock

M. G.

Smith

. 2018. Military and Civilian Pay Levels, Trends, and Recruit Quality. Santa Monica, CA: RAND Corporation.

33.

Ifatunji

Mosi Adesina

. 2017. “Labor Market Disparitiesb Between African Americans and Afro Caribbeans: Reexamining the Role of Immigrant Selectivity.” Sociological Forum 32(3): 522–43.

34.

Itzigsohn

Giorguli-Saucedo

. 2005. “Incorporation, Transnationalism, and Gender: Immigrant Incorporation and Transnational Participation as Gendered Processes.” International Migration Review 39(4): 895–920.

35.

Jacobs

J. B.

Hayes

L. A.

. 1981. “Aliens in the US Armed Forces: A Historico-Legal Analysis.” Armed Forces & Society 7(2): 187–208.

36.

Kleykamp

M. A.

2006. “College, Jobs, Or the Military? Enlistment during a Time of War.” Social Science Quarterly 87(2): 272–290.

37.

Krikorian

2003. “Green-Card Soldiers: Should the US Military be Reserved for Americans?” National Review Online, 22.

38.

Leal

D. L.

2003. “The Multicultural Military: Military Service and the Acculturation of Latinos and Anglos.” Armed Forces & Society 29(2): 205–226.

39.

Lundquist

J. H.

2004. “When Race Makes no Difference: Marriage and the Military.” Social Forces 83(2): 731–757.

40.

Lundquist

J. H.

2008. “Ethnic and Gender Satisfaction in the Military: The Effect of a Meritocratic Institution.” American Sociological Review 73(3): 477–496.

41.

Lundquist

J. H.

Pager

Strader

. 2018. “Does a Criminal Past Predict Worker Performance? Evidence from One of America’s Largest Employers.” Social Forces 96(3): 1039–68.

42.

Lundquist

J. H.

. 2014. “Reinstitutionalizing Families: Life Course Policy and Marriage in the Military.” Journal of Marriage and Family 76(5): 1063–1081.

43.

Lutz

2008. “Who Joins the Military? A Look at Race, Class, and Immigration Status.” Journal of Political and Military Sociology 36(2): 167–88.

44.

MacLean

2011. “The Stratification of Military Service and Combat Exposure, 1934–1994.” Social Science Research 40(1): 336–348.

45.

MacLean

Elder

G. H.

Jr . 2007. “Military Service in the Life Course.” Annual Review of Sociology 33: 175–196.

46.

Malone

, 2014. “Hiring from High-Risk Populations: Lessons from the US Military.” Contemporary Economic Policy, 32(1): 133–143.

47.

McIntosh

M. F.

Sayala

. 2011. Non-Citizens in the Enlisted US Military. Alexandria, VA: CNA

48.

Melzer

S. M.

Tomaskovic-Devey

Schunck

Jacobebbinghaus

. 2018. “A Relational Inequality Approach to First- and Second-Generation Immigrant Earnings in German Workplaces.” Social Forces 97(1): 91–128.

49.

Mission Readiness. 2009, November 5. Ready, Willing, and Unable to Serve. Washington, DC: Council for a Strong America.

50.

Model

Suzanne

. 2008. “The Secret of West Indian Success.” Society 45(6): 544–48.

51.

Moskos

C. C.

1983. Soldiers and Sociology. Washington, DC: U.S. Army Research Institute.

52.

Moskos

C. C.

Butler

J. S.

. 1996. All That We Can Be: Black Leadership and Racial Integration the Army Way. New York: Perseus Book Group.

53.

Myers

Dowell

Gao

Xin

Emeka

Amon

. 2009. “The Gradient of Immigrant Age-at-Arrival Effects on Socioeconomic Outcomes in the U.S.” International Migration Review 43(1): 205–29.

54.

National Research Council. 2012. Small Populations, Large Effects: Improving the Measurement of the Group Quarters Population in the American Community Survey. Washington, D.C.: National Academies Press.

55.

Ousey

G. C.

Kubrin

C. E.

. 2009. “Exploring the Connection between Immigration and Violent Crime Rates in U.S. Cities, 1980-2000.” Social Problems 56(3): 447–473.

56.

Roberts

H. E.

Skinner

. 1996. “Gender and Racial Equity of the Air Force Officer Qualifying Test in Officer Training School Selection Decisions.” Military Psychology 8(2): 95–113.

57.

Roder

Muhlau

. 2014. “Are They Acculturating? Europe’s Immigrants and Gender Egalitarianism.” Social Forces 92(3): 899–928.

58.

Savage

P. L.

Gabriel

R. A.

. 1976. “Cohesion and Disintegration in the American Army: An Alternative Perspective.” Armed Forces & Society 2(3): 340–76.

59.

Shils

E. A.

Janowitz

. 1948. “Cohesion and Disintegration in the Wehrmacht in World War II.” Public Opinion Quarterly 12(2): 280–315.

60.

Stein

Dan

. 2018, May 22. “Our Military is Working Just Fine. It’s Immigration Policy That is Broken.” USA Today. https://www.usatoday.com/story/opinion/2018/05/22/military-should-not-be-backdoor-citizenship-editorials-debates/35228661/

61.

Strader

2018. “Do Race and Marriage Matter? The Motherhood Wage Penalty in the Military.” Paper presented at the Work and Family Researchers Network Conference, Washington, DC.

62.

Tomaskovic-Devey

Hällsten

Avent-Holt

. 2015. “Where Do Immigrants Fare Worse? Modeling Workplace Wage Gap Variation with Longitudinal Employer-employee Data.” American Journal of Sociology 120(4): 1095–1143.

63.

U.S. Citizenship and Immigration Services. 2019, June 26. “Number of Form N-400, Application for Naturalization by Category of Naturalization, Case Status, and USCIS Field Office Location, Fiscal Year 2019, Quarter 2.” Washington, DC: U.S. Citizenship and Immigration Services.

64.

Van Epps

2008. “Relooking Unit Cohesion: A Sensemaking Approach.” Military Review 88(6): 102–110.

65.

Xie

1992. “The Socioeconomic Status of Young Male Veterans, 1964–1984.” Social Science Quarterly 73: 379–96.

66.

Zong

Batalova

. 2016, October 13. “Immigrant Veterans in the United States”. Migration Policy Institute. Migration Information Source. https://www.migrationpolicy.org/article/immigrant-veterans-united-states