Socioeconomic Adaptation of Bangladeshi Immigrants in the United States

Abstract

Despite the tremendous growth of Bangladeshi immigrants in the United States in recent decades, systematic research on their socioeconomic adaptation is lacking, and existing portraits of their socioeconomic attainment are contradictory. Using the pooled samples from the 2001–2019 American Community Surveys with a substantial sample size and various regression models, this study analyzes the socioeconomic adaptation of foreign-born Bangladeshis by focusing on the most common indicators—educational attainment, occupational status, self-employment, and income. The results reveal that, in general, Bangladeshi immigrants are well educated and include a significant proportion of professionals, managers, and other white-collar workers, but their average income is very low. The regression analyses show that many demographic, familial, assimilation, and socioeconomic factors contribute significantly to the socioeconomic adaptation of Bangladeshi immigrants. Our findings provide a new portrait of Bangladeshi immigrants’ socioeconomic adaptation in the U.S. based on the latest generalizable data and new evidence on the determinants of their socioeconomic adaptation. The findings also have significant implications for scholarship and policy.

Introduction

Bangladeshi immigrants are one of the fastest-growing immigrant groups in the United States. In the period of 1971–1980 after the independence of Bangladesh in 1971, only around 4,154 Bangladeshis immigrated to the U.S., but the period of 2001–2020 witnessed an influx of 252,047 Bangladeshi immigrants (Akhter & Yang, 2023). Bangladeshis in the U.S. are estimated in the neighborhood of 500,000 (Akhter & Yang, 2023; Felts, 2020; Sultana, 2005). Bangladeshis in the U.S. are largely first-generation immigrants (Morshed, 2018; Paul, 2008; Rahman, 2011). Despite the sharp increase in Bangladeshi immigration over the past few decades, substantial scholarly research on Bangladeshi immigrants in general is lacking; research on the adaptation of Bangladeshi immigrants in the United States is grossly inadequate.

In particular, there is a dearth of systematic data about the socioeconomic attainment of Bangladeshi immigrants in the United States. Information on the socioeconomic adaptation of Bangladeshi immigrants is at best fragmentary and outdated. A few available descriptive analyses made some contradictory claims about the socioeconomic status of Bangladeshis in the U.S. based on earlier data and descriptive statistics. For example, in an encyclopedia entry, Zhao (2014) asserted that Bangladeshis in the U.S. had a relatively low socioeconomic status. In a study of South Asian Americans, Kibria (2006) included some statistics of foreign-born Bangladeshis based on the 5% Public Use Microdata Sample (PUMS) of the 2000 Census with a modest sample size of 153. The statistics showed that foreign-born Bangladeshis tended to be more college-educated than U.S.-born Whites but fared worse than foreign-born Indians and Pakistanis, to be more likely to do professional jobs than U.S.-born Whites but less likely to do so than foreign-born Indians and Pakistanis, to be less likely to have managerial and business/financial-related occupations than U.S.-born Whites and foreign-born Indians and Pakistanis, and to earn lower median income than U.S.-born Whites and foreign-born Indians and Pakistanis. Using the pooled data from the Current Population Surveys (CPS) of 2009–2013, a report of the Migration Policy Institute (MPI, 2014) concluded that Bangladeshis in the U.S. as a population was better educated than the general U.S. population and identical to the general U.S. population in the percentage of professional or managerial occupations, but fared slightly better in income than the U.S. population. Although Bangladeshis’ educational advantage seems to be consistent, the portraits of their occupational and earning status are conflicting. Furthermore, information on the determinants of Bangladeshi immigrants’ socioeconomic attainment is absent. Most recent data with a large sample size are called for in order to ascertain the socioeconomic attainment of Bangladeshi immigrants in the United States and its determinants. Do Bangladeshi immigrants really fare better than the general U.S. population? What is their relative socioeconomic position compared to other major Asian immigrant groups such as Chinese, Japanese, Filipinos, Koreans, Asian Indians, and Vietnamese? What factors may influence their socioeconomic performance? These questions beg answers and guide the current study.

To fill the void in the existing literature, the current study systematically investigates the socioeconomic adaptation of Bangladeshi immigrants using the pooled 2001–2019 American Community Survey data with a substantial sample size (N = 23,889). Socioeconomic adaptation is defined as the adjustment of immigrants to life in the host country through socioeconomic attainment. Socioeconomic attainment reflects immigrants’ core advancement in society and overall well-being (Sakamoto & Xie, 2006). Although socioeconomic attainment may be measured by many indicators, because of the space constraint in this study we center on the most common indicators including educational attainment, occupation status, self-employment, and income. The remainder of this paper briefly reviews the literature on immigrant adaptation in general and Bangladeshi immigrants’ socioeconomic adaptation relevant to their immigration in particular, outlines pertinent theoretical perspectives on immigrant socioeconomic adaptation, proposes hypotheses for testing, depicts the data and methods, presents our findings, and discusses their implications.

Literature Review

Immigrant adaptation to, or integration into, the host society has been a long-standing subject of scholarly inquiry. In the USA, research on immigrant adaptation can be traced back to the Chicago School of Sociology in the 1930s and has continued unabated until today. Classic works include, but are not limited to, those by Park (1937), Gordon (1964), Greeley (1974), Alba and Nee (1997, 2003), Portes and Zhou (1993), and Portes and Rumbaut (2014). Immigrant adaptation encompasses many dimensions such as cultural adaptation, structural adaptation, marital adaptation, identificational adaptation, attitudinal adaptation, behavioral adaptation, psychological adaptation, socioeconomic adaptation, political adaptation, etc. (see, Gordon, 1964; Yang, 2000, 2011). Socioeconomic adaptation is one of the most crucial and immediate challenges immigrants face.

Patterns of immigrant socioeconomic adaptation appear to vary in the U.S. over time. Extant research has shown that in the pre-1965 era, immigrants, who hailed primarily from Europe, tended to follow the traditional bottom-up route of socioeconomic mobility generation by generation (e.g., Lieberson, 1980; Warner & Srole, 1945). However, in the post-1965 era, the majority of immigrants came from the Global South such as Latin America and Asia, and the trajectory of socioeconomic adaptation and mobility for these new immigrants is not a straight line nor unidirectional but varies across different groups. For example, some new Asian immigrant groups such as Asian Indians, Japanese, Chinese, and Koreans joined the middle class upon entry because of the selective immigration of more educated and more skilled immigrants (Kochhar, 2024; Yang, 2011; Zhou, 1997). On the other hand, certain groups such as Mexicans tended to start at the bottom of the socioeconomic ladder because of the selective immigration of less educated and less skilled immigrants as well as undocumented immigrants (Portes & Rumbaut, 2014; Rendall & Parker, 2014). There are also arguments and evidence on downward socioeconomic mobility among poor immigrants, especially the dark-skinned ones, such as Haitians, Salvadorans, other Central and South Americans, some from the Caribbean, as well as some Indochenese (Gans, 1992; Han, 2020; Zhou, 1997). It is also interesting to note that Michalikova and Yang (2016) found mixed patterns of socioeconomic adaptation among new Eastern European immigrants with high levels of education and occupational status but very low income. Where do Bangladeshi immigrants fit in the trajectories of socioeconomic adaptation and mobility among post-1965 new immigrants? Are they similar to the more successful Asian immigrant groups, the dark-skinned immigrant groups, or somewhat in between? An investigation of their socioeconomic adaption will offer an excellent opportunity to answer these questions.

To understand the socioeconomic adaptation of Bangladeshi immigrants, it is useful to briefly review the history of Bangladeshi immigration to the U.S. pertinent to their socioeconomic backgrounds. There have been three major waves of Bangladeshi immigration to the United States. The first wave of Bangladeshi immigration (between 1885 and 1947) commenced in the late nineteenth century, when Bangladesh, Pakistan, and India constituted one big nation under the British rule, called India. People of those countries of the Indian subcontinent were perceived as illiterate and backward, and their communities were heavily discriminated against. A smaller number of Muslims from the Bengal region (Bangladesh) migrated to work as peddlers, lascars, and indentured servants in New York, New Jersey, Maryland, New Orleans, California, Oregon, Washington, Texas, Illinois, etc. (Bald, 2013).

The second wave of Bangladeshi immigration occurred between the 1960s and 1970s. Many East Bengalis arrived in the United States in the 1960s right before Bangladesh independence to flee political or religious persecution (Zhao, 2014). The Immigration and Nationality Act of 1965 boosted this wave of Bangladeshi immigration (Leonard, 1997). The well-educated, English-proficient, highly qualified professionals, and middle-class people constituted the majority of Bangladeshi immigrants during this period (Paul, 2008). There were also some Bangladeshi students in U.S. universities and colleges who gained permanent residence after graduation. After the independence of Bangladesh in 1971, about 4000 Bangladeshis immigrated to the U.S. in the 1970s (Akhter & Yang, 2023). They were largely professionals or white-collar workers.

During the 1980s, the third wave of immigrants brought a remarkable demographic shift, as more Bangladeshis began migrating in that period with diversified socioeconomic status. Since then, they came under the Diversity Visa (DV) Program and family re-unification criteria. The majority of the Bangladeshi immigrants who came under the DV Program were comparatively less educated and less skilled than their earlier counterparts. Many of them worked in low-paying jobs in grocery stores, hotels, restaurants, gas stations, and taxi companies in large cities such as New York, Chicago, Boston, Dallas, and Houston. Some ran small businesses such as gas stations, motels, restaurants, travel agencies, and small construction companies (Siddiqui, 2004). However, those who came as students were highly educated and professional (Paul, 2008). Despite crushing socioeconomic disadvantages and gender disparity, in the third wave of immigration, a large number of Bangladeshi immigrants became successful economically. They attained successful careers as engineers, professors, doctors, businessmen, and so on (Ahmed, 2006; Kibria, 2011). However, they have also experienced downward mobility and found difficulties in converting their education and work experiences from Bangladesh to highly-skilled jobs in the U.S. As a result, they may often have to seek employment at gas stations, restaurants, and other avenues for which they were overqualified. Their educational credentials and job experiences were devalued because of weak English-language skills (Kibria, 2013; Shams, 2020). Additionally, some Bangladeshi immigrants struggled to survive (Dutta & Jamil, 2013). They lived under poverty and could not even meet their fundamental needs such as health care. Language barriers prevented them from communicating effectively with the larger host society.

Foreign-born Bangladeshis have a lower level of educational attainment than U.S.-born Bangladeshis, according to the U.S. Bureau of the Census (2000). Approximately 53% of native-born Bangladeshis had a bachelor's degree, but only 39.9% of the foreign-born Bangladeshis possessed a bachelor's degree (Kibria, 2006). About 32.4% of foreign-born Bangladeshis held a high school diploma, and 19.3% of them had less than a high school degree (Kibria, 2013). In 2019, 36% of Bangladeshi immigrants aged 25 and older obtained a high school diploma or less, 16% attended college for an associate degree but did not earn a degree, 26% earned a bachelor’s degree, and 22% earned a postgraduate degree, according to a study by the Pew Research Center (2021). Bangladeshi immigrants included a significant proportion of professional and managerial workers. About one-third of foreign-born Bangladeshis worked in managerial and professional occupations, 29.6% held white-collar jobs in sales and office work, 18.1% held jobs in the service sector, and 15.5% were in production and transportation (Kibria, 2013). Foreign-born Bangladeshis have lower median earnings than their U.S-born counterparts. The median household income for foreign-born Bangladeshi Americans was $40,000 compared to $54,000 for native-born Bangladeshis (Kibria, 2006). About 16% of foreign-born Bangladeshis lived below the poverty level in contrast to 0% of their U.S.-born counterparts (Kibria, 2006).

Overall, the existing literature on the socioeconomic adaptation of Bangladeshi immigrants is very limited. There is a lack of extensive coverage on their educational attainment, occupational status, self-employment, and income. Moreover, the analysis of the determinants of socioeconomic adaptation is completely absent. Additionally, there is not much information about gender differences in socioeconomic attainment.

Theoretical Considerations and Hypotheses

Several theories are relevant to comprehending Bangladeshi immigrants’ socioeconomic adaptation to life in the United States. Among the theories of immigrant adaptation in the context of the United States, the most influential theory for a long time was classic assimilation theory, which emerged in the 1930s (Alba & Nee, 1997) and remained a key sociological paradigm until the 1960s and 1970s (Heisler, 2000). According to this model, immigrants will go through a straight-line pattern of upward socioeconomic mobility with time and generation by generation (Warner & Srole, 1945). The process of assimilation is a straight-line one, and the outcome is uniform upward mobility for all new groups (Warner & Srole, 1945). The straight-line assimilation model has been criticized and challenged by new theories in last few decades. A new assimilation theory formulated by Alba and Nee (1997, 2003) put forth a bumpy-line assimilation model, suggesting that the process of assimilation may be bumpy and jagged, but remains important and inevitable in the adaptation of contemporary immigrants. Different immigrant groups may experience assimilation differently.

Also challenging the classic assimilation theory, Portes and Zhou’s (1993) segmented assimilation theory suggested that different immigrant groups may undergo different paths and become assimilated into different “segments” of American society. Possible paths and outcomes include upward mobility into the white middle class, downward mobility into the underclass, and upward economic mobility together with delayed acculturation and conscious preservation of immigrant cultures and institutions. Segmented Assimilation focuses on identifying the contextual, structural, and cultural factors that separate successful assimilation from unsuccessful or even "negative" assimilation. According to this theory, immigrant groups adapt differently. Bangladeshi immigrants blend Islamic, Bangladeshi, and U.S. cultural traditions, leading to unique acculturation patterns. Segmented assimilation theory, which focuses on second-generation adaptation, may not fully explain how first-generation immigrants adapt.

What kinds of Bangladeshi immigrants are more likely to adapt well socioeconomically in the United States? Based on the theoretical perspectives, we propose a number of hypotheses for testing:

A longer U.S. residency helps Bangladeshi immigrants acquire English language skills, build up stronger social networks, gain more labor market experience, and therefore fare better socioeconomically. The longer the immigrants stay in the receiving country, the higher the level of their socioeconomic success (Jasinskaja-Lahti, 2008; Logan & Drew, 2011; Michalikova, 2018; Michalikova & Yang, 2016; Portes & Rumbaut, 2014). Hence, we hypothesize that there is a positive relationship between the length of U.S. residency and the degree of socioeconomic adaptation among Bangladeshi immigrants, controlling for other variables in the analysis.

Older immigrants tend to learn socio-cultural skills less quickly than their younger counterparts, because as older immigrants spend most of their time in a different cultural setting, it is difficult for them to forget their way of life (Berry, 1997; Schaafsma & Sweetman, 2001). They have deeper attachment to their country of origin than younger immigrants. Moreover, older immigrants are less adventurous than younger ones, who tend to have weaker attachment to their cultural orientation, to be more educated, and to have lower social capital than older immigrants (Ahmad, 2011; Khaleque et al., 2008; O’Neil & Tienda, 2014). Thus, we expect that older Bangladeshi immigrants adapt less well socioeconomically than younger Bangladeshi immigrants, all else being equal.

Existing research reveals that Asian immigrant women have disadvantages in socioeconomic success partly because of strict cultural norms (Kim & Zhao, 2014; Lueck, 2017). The cultural expectation of female Bangladeshis is to be shy and averse to the new culture. In general, Bangladeshi families are patriarchal. Men are supposed to be responsible for the family’s social and physical well-being, while women are expected to nurture the family. Women tend to experience a slower process of adaptation as they tend to stay home and to maintain their traditions. Since men have greater exposure to the outside world and the larger society, they tend to adapt better socioeconomically than women. Many Bangladeshi women are stay-at-home mothers, while Bangladeshi men work outside the home (Khaleque et al., 2008). Hence, we anticipate that female Bangladeshi immigrants tend to adapt less well socioeconomically than male Bangladeshi immigrants, ceteris paribus.

It is generally perceived that marriage is regarded as rewarding, or that it has a positive effect on an individual’s well-being (Lucas et al., 2003). When a married immigrant couple lives together in the same place, they can offer physical and mental support to each other. Marital status alone may not have direct influence on the immigrant’s adaptation, but it may depend on the presence of the spouse in the host country. This attachment acts as social support for married immigrants (Bratsberg & Ragan, Jr. 2002; Constant & Massey, 2002; Michalikova & Yang, 2016). Hence, we hypothesize that Bangladeshi immigrants who are married with their spouse in America adapt better socioeconomically than Bangladeshi immigrants who are married without their spouse in America or never married, holding other variables constant.

Children are enthusiastic and quick to adapt to American culture. They also can learn English and other subjects relatively easily. This helps them make more friends and become Americanized (Zhou, 1997). Besides, immigrant children have greater chances for better education, a secure future, and an overall wellbeing than their parents. Children also help their parents to adapt (Portes & Rumbaut, 2014). As Bangladesh is a poverty-stricken and politically unstable country, many Bangladeshi parents want to settle in America for their children to have a better life in the future. This desire for a brighter future of their children encourages Bangladeshi immigrants with children to embrace the host society and experience better socioeconomic adaptation. Therefore, we predict that Bangladeshi immigrants who have children tend to fare better socioeconomically than those who do not have children, holding other variables constant.

Most of the Bangladeshi immigrants live in New York, New Jersey, Washington, D.C., Detroit, Boston, etc. (Akhter & Yang, 2023). Although many Bangladeshis started to live in Southern states like Texas and Florida recently, New York/New Jersey still have mostly dense Bangladeshi concentration (Kibria, 2013). Generally, immigrants want to gravitate to areas where they can find the people of their own community for certain kinds of comforts (Ahmad, 2011). A large number of Bangladeshi communities in the Northeast may have a positive effect on the adaptation of Bangladeshi immigrants living in the Northeast. Conversely, Bangladeshis living in other regions may not fare better than the Bangladeshis living in the Northeast as they do not get enough support from their ethnic communities because of lesser density of co-ethnics. Thus, we expect that Bangladeshi immigrants who settle in non-Bangladeshi-concentrated regions of the U.S. tend to adapt less well socioeconomically than those who settle in the Northeast with a Bangladeshi-concentration, controlling for other variables.

Pooling multiple years of data creates a possibility to examine changes in adaptation across different survey years. This examination revealed variations in adaptation processes over time, resulting from possible differences in characteristics of subsequent survey years. Economic or political events in the United States may affect the adaptation processes. Developments in the country of origin may also play a role. Any kind of changes may affect the characteristics of later immigrants. Overall, it is expected that immigrants interviewed in later years adapt better, as later survey years are more likely to include immigrants who are more likely to be adjusted and are positively selected for immigration. Accordingly, Bangladeshi immigrants who were surveyed in later years tended to adapt better socioeconomically than those who were surveyed in earlier years.

Generally, a lack of proficiency in the prevailing language of the host nation will reduce the wage rates of foreign-born individuals, and the wage gap between natives and immigrants’ narrows as immigrants acquire greater language abilities. Immigrants with superior language skills are more likely to be able to obtain a higher level of education, get a better job and earn a higher salary (Bratsberg & Ragan, Jr. 2002; Chiswick, 1991; Chiswick & Miller, 2012; Lueck, 2017; Rivera-Batiz, 1990; Shields & Price, 2002). Hence, we predict that English proficiency is positively related to the degree of socioeconomic adaptation, controlling for other variables.

Education is linked with better immigrant adaptation. It is considered as a form of investment. Education facilitates occupational and income attainment in society (Berry, 1997; Bratsberg & Ragan, Jr. 2002; Chiswick, 1978; Tamborini et al., 2015). Thus, we anticipate that Bangladeshi immigrants with a higher level of education tend to have a higher level of occupational and income attainment than Bangladeshi immigrants with a lower level of education, holding other variables constant.

A better job leads to a higher remuneration. Hence, we expect a higher job status will lead to a higher income among Bangladeshi immigrants just as in the general population. Therefore, we hypothesize that Bangladeshi immigrants with a higher level of occupational status tend to have a higher level of income than Bangladeshi immigrants with a lower level of occupational status, all else being equal.

Finally, we hypothesize that self-employed Bangladeshi immigrants tend to have a higher level of income than non-self-employed Bangladeshi immigrants, all else being equal. Self-employment is a form of immigrant economic adaption to the host society. Self-employment is an indicator of immigrants’ economic self-reliance and mobility (Portes & Rumbaut, 2014; Sanders & Nee, 1996). The foreign-born are found to have higher rates of self-employment than natives (Borjas, 1986; Lofstrom & Wang, 2019). Evidence from some immigrant groups shows self-employed immigrants had a significantly higher income than their non-self-employed counterparts (Hamilton, 2000; Min, 1995; Waldinger et al., 1990). We expect the same for Bangladeshi immigrants.

Data ad Methods

Sample

The data for this study come from the Integrated Public Use Microdata Sample (IPUMS) of the 2001–2019 American Community Surveys (ACS), which are nationally representative samples of the U.S. population. The ACS is a national survey of the U.S. population collected annually by the U.S. Bureau of the Census since 2001 and replaced the U.S. Census long form used between 1970 and 2000. The IPUMS data were de-identified by the U.S. Bureau of the Census to protect the identities and confidentiality of the respondents. The ACS surveyed about 3.5 million U.S. residents in the 50 states, District of Columbia, and Puerto Rico on topics not on the 2000–2020 Censuses, such as education, employment, occupation, income, internet access, and transportation. Since the number of Bangladeshis included in the survey each year was relatively small, we pooled together 19 years of the ACS data to maximize the sample size. The data were restricted to foreign-born Bangladeshis residing in the United States. The valid sample size is 23,889. For socioeconomic adaptation, depending on the dependent variables, age restrictions varied. For the dependent variable educational attainment, we restricted the sample to respondents aged 25 years old or older because typically people have completed college education by the age of 25. For the dependent variable personal income, the sample was restricted to the respondents aged 16 years old or older. For occupational status and self-employment, the sample was restricted to respondents who were 16 to 64 years old, namely, those in the labor force.

To address the questions of how well Bangladeshi immigrants fare socioeconomically compared to the general U.S. population and other major Asian immigrant groups, we also analyzed ACS 2001–2019 by including the general U.S. population, and foreign-born Chinese, Japanese, Filipinos, Koreans, Asian Indians, and Vietnamese for comparison.

Variables and Measurements

Dependent variables

Because of space constraint, we only include four common measures of socioeconomic adaptation: educational attainment, occupational status, self-employment, and income. Education (EDUC) is measured by the highest year of school or degree earned. Education is an ordinal variable with eleven categories: 0 = No schooling; 1 = Nursery school to Grade 4; 2 = Grade 5, 6, 7, or 8; 3 = Grade 9; 4 = Grade 10; 5 = Grade 11; 6 = Grade 12; 7 = 1 year of college; 8 = 2 years of college; 9 = 4 years of college; 10 = 5 + years of college.

Income (INCTOT) is an interval/ratio variable measured by dollars. It is a 7-digit numeric code reporting each respondent's total pre-tax personal income or losses from all sources for the previous year, ranging from $ −9,999 to $1,376,000. Since a significant number of cases had negative or zero income, we recoded the negative values of income as 0 and copied other values. Because of the very high Kurtosis (54.73) and high skewness (5.34), we log-transformed income to correct for the non-normality. Before the log transformation, to preserve the cases with an income of 0, a small constant of 1 was added to the income variable.

The occupation variable (OCC2010) in the ACS has a broad range of categories as it followed the 2010 Occupational Code list from the U.S. Census Bureau. For this study, we created two dummy variables for occupational status: professional or managerial occupation (1 = managerial-professional, or related occupations; 0 = otherwise), and white-collar occupations including professional, managerial, sales and office occupations (1 = white-collar occupations; 0 = blue-collar occupation).

We constructed a dummy variable to measure self-employment coded 1 for self-employment and coded 0 otherwise.

Independent variables

The independent variables for all dependent variables include age, gender, marital status, having children, region, length of U.S. residency, English proficiency, and survey year. Age is a continuous variable measured by years. Gender is a dummy variable with 1 for female and 0 for male. Marital status is a dummy variable with 1 indicating currently married with a spouse in the United States and 0 indicating currently married without a spouse in the United States or not currently married. Having children is a dummy variable coded 1 if the respondent had children and 0 otherwise. We created three dummy variables for regions coded 1 for the designated category and coded 0 otherwise with Northeast as the reference category: Midwest, South, and West. Length of U.S. residence is measured by years. English proficiency is an ordinal variable with 5 categories ranging from not speaking English at all to speaking English only. Survey year was treated as linear without recoding. In addition, education was used as a predictor of occupational status, self-employment, and income. Occupation was a predictor of income.

Limitations of the Data

Limitations of the data should be acknowledged. First, although merging 19 years of ACS data into one file has the advantage of increasing the sample size and ensuring reliable parameter estimates, it may not best capture changes over the 19 years. Nonetheless, since our study is not a trend study, change over time is not our focus. Second, pertinent to the first limitation, there is a risk that the estimates over 19 years could be impacted by changes over time. Fortunately, the IPUMS of the ACS obtained from the Minnesota Population Center ironed out any inconsistencies of the ACS data over time. Furthermore, to ensure the accurate estimates in our regression analyses, we controlled for year of survey. As shown in the next section, the results show that survey year had no effect on educational attainment, self-employment, and income although it had significant, modest effects on occupational status. By and large, our findings are not much impacted by survey year. Finally, some potential predictors of economic adaptation are not available in the IPUMS data. For example, macroeconomic conditions at the time of immigration can facilitate or hinder immigrant economic adaptation, but the IPUMS of the ACS will not allow us to measure the macro-level economic conditions at the time immigration for individual immigrants. These limitations notwithstanding, our data remain the best data available to study the socioeconomic adaptation of Bangladeshi immigrants.

Methods of Analysis

To compare Bangladeshi immigrants with other immigrant groups in socioeconomic adaptation, we did one-way analysis of variance (ANOVA) for educational attainment and personal income and chi-square test for occupation and self-employment. One-way ANOVA is appropriate because we compared multiple groups in the mean of personal income, a continuous variable, and education attainment, which can be treated as a continuous variable. Chi-square test is most appropriate because group membership and occupation or self-employment are categorical variables.

The methods of regression analysis vary, depending on the measurements of the dependent variables. We used ordinary least squares (OLS) regression for dependent variables educational attainment and personal income since educational attainment is an ordinal variable with eleven categories and personal income is an interval-ratio variable. Additionally, to determine which predictor has the strongest effect on the dependent variable, standardized coefficients were used. For occupational status and self-employment, we performed logistic regression because these variables are dichotomous.

Results

This paper focuses on how well Bangladeshi immigrants fare socioeconomically to American life and what factors determine their socioeconomic adaptation. The results in this section present descriptive statistics in socioeconomic attainment by comparing Bangladeshi immigrants with major Asian immigrant groups as well as the average U.S. population and regression analyses of the determinants of socioeconomic adaptation among Bangladeshi immigrants.

Bangladeshi Immigrants’ Socioeconomic Status: A Glimpse and Comparison

What is the status of Bangladeshi immigrants’ socioeconomic adaptation to American life? How well do they fare compared to major Asian immigrant groups and the general U.S. population? We address these questions by providing evidence in terms of several common socioeconomic indicators, including educational attainment, occupational status, self-employment, and personal income.

Educational Attainment

Table 1 displays the means, medians, and standard deviations of educational attainment for Bangladeshi immigrants in comparison to major Asian immigrant groups and the general U.S. population. The results indicate that, on average, Bangladeshi immigrants had a mean education level of about 7.9, or almost two years of college education, with a standard deviation of about 3 years. The median (= 8) shows half of the foreign-born Bangladeshis had completed two years of college and half of them had not.

Table 1 Educational Level of Bangladeshi Immigrants in Comparison to Major Asian Immigrant Groups and U.S. General Population Aged 25 or Older, ACS 2001–2019

How well do Bangladeshi immigrants fare in educational attainment in comparison to major Asian immigrant groups? To answer this question, we conducted a one-way ANOVA. Since the test of homogeneity of variances indicates variances across multiple groups in educational attainment are not equal, we used the Welch test, instead of the F test. As can be seen in Table 1, the result indicates statistically significant differences in educational attainment across groups (Welch = 36,113.66, p ≤ 0.001). Table 1 also indicates that Bangladeshi immigrants had lower educational attainment than Japanese, Korean, Filipino, and Indian immigrants, consistent with our hypothesis. On the other hand, the average educational level of Bangladeshi immigrants was significantly higher than that of Vietnamese immigrants and somewhat higher than those of Chinese immigrants and the general U.S. population. These results are consistent with our hypothesis for the U.S. population but contrary to the hypothesis for Chinese and Vietnamese.

Occupation

As shown in Table 2, about 36% of Bangladeshi immigrants held professional/managerial occupations. As expected, Bangladeshis had a slightly higher percentage of professional/managerial occupations than the general U.S. population, but they were significantly behind most major Asian immigrant groups, especially Indians (70.2%), except for Vietnamese (30.5%). The χ² test indicates significant differences in the distribution of professional/managerial vs. non-professional/managerial occupations (χ² = 131,826, p ≤ 0.001). These results are largely consistent with our hypothesis.

Table 2 Percentage Distributions of Professional/Managerial vs. Non- Professional/Managerial Workers, Bangladeshi Immigrants in Comparison to Major Asian Immigrant Groups and General U.S. Population Aged 16–64, ACS 2001–2019

The distribution of white-collar versus blue-collar workers is similar to, albeit not exactly the same as, that of professional/managerial occupations. Table 3 shows that roughly two thirds (66.8%) of Bangladeshi immigrants were white-collar workers. Bangladeshi immigrants had a lower percentage of white-collar workers than Indians, Japanese, Koreans, and Chinese, but to varying degrees a higher percentage than the general U.S. population, Vietnamese, and Filipinos. The differences in the distribution of white-collar versus blue-collar workers among the groups were highly significant, according to the χ² test (χ² = 80,621.98, p ≤ 0.001). These results coincide with our hypothesis except for the Vietnamese and Filipinos.

Table 3 Percentage Distributions of White-Collar and Blue-Collar Workers, Bangladeshi Immigrants in Comparison to Major Asian Immigrant Groups and General U.S. Population Aged 16–64, ACS 2001–2019

Self-employment

As shown in Table 4, nearly 12% of Bangladeshi immigrants were self-employed. Bangladeshi immigrants had a much lower rate of self-employment than Koreans immigrants (19.7%), who were well documented to have a high self-employment rate, and a somewhat lower rate than Vietnamese immigrants (13.8%). However, Bangladeshi immigrants had a slightly higher rate than the general U.S. population, Chinese, Japanese, and Indians, and a much higher rate than Filipinos (5.4%). These cross-group variations in self-employment revealed by a χ² test are statistically significant (χ² = 19,900.85, p ≤ 0.001). The evidence for our hypothesis is mixed.

Table 4 Percentage Distributions of Self-employment, Bangladeshi Immigrants in Comparison to Major Asian Immigrant Groups and General U.S. Population Aged 16–64, ACS 2001–2019

Income

The measurement of socioeconomic performance eventually comes down to income attainment. The average personal income of Bangladeshi immigrants was $29,484, with a standard deviation of $51,640 (see Table 5). Half of the Bangladeshi immigrants made less than $14,000, and half of them made more than that amount. Approximately about 25% of Bangladeshi immigrants made more than $36,000, but 75% of them made less than that amount, including about 25% with a personal income below zero.

Table 5 Income of Bangladeshi Immigrants in Comparison to Major Asian. Immigrant Groups and U.S. General Population Aged 16 or Older, ACS 2001–2019

In comparison to major Asian immigrant groups and the general U.S. population, we found that Bangladeshi immigrants fared the worst in terms of both mean income and median income. Consistent with our hypothesis, Bangladeshis had lower income than all major Asian immigrant groups. At odds with our hypothesis, they made a much lower income than the general U.S. population. The differences in income across the groups in the comparison are highly statistically significant based on the results of one-way ANOVA (Welch = 3861.85, p ≤ 0.001).

Determinants of Socioeconomic Adaptation

What factors contribute to the socioeconomic adaptation of Bangladeshi immigrants? We examine the determinants of four indicators of socioeconomic adaptation: educational attainment, occupational status, self-employment, and income.

Educational Attainment

We tested four nested regression models predicting education attainment (see Table 6). Model 1 includes demographic variables (i.e., age, sex, and year). Model 2 adds dummy variables for marital status and having children to Model 1. Model 3 adds English proficiency and the length of stay in the U.S. as adaptation/assimilation variables to Model 2. Model 4 adds region to Model 3. This sequential modeling technique enables the assessment of how variance explained in the dependent variable changes by adding additional predictors. The F values are significant at the 0.001 level for all models, demonstrating that all models fit the data well. Variables included in the models explain between 6% (Model 1) and 24.5% (Model 4) of the variation in educational attainment as shown by the values of R². Most regression coefficient estimates reach statistical significance at the 0.001 level, which can be partly attributed to the large sample size (N = 20,654).

Table 6 Estimates of OLS Regression Models Predicting Educational Attainment of Bangladeshi Immigrants Aged 25 or Older, 2001–2019 ACS (Standard Errors in Parentheses)

Model 1 examines the impact of basic demographic factors on educational achievement. As expected, the association between age and education is negative. This result is not surprising since older immigrants acquired less education than their younger counterparts. Consistent with our hypothesis, female Bangladeshi immigrants tended to have a lower level of educational attainment than their male counterparts. The survey year was not a significant predictor.

In Model 2, the two dummy family variables added are statistically significant predictors. Consistent with our hypothesis, Bangladeshi immigrants who were married with a spouse present in the U.S. had a higher level of education than those who were not married or who were married without a spouse in the U.S. Contrary to our expectation, Bangladeshi immigrants who had children had a lower level of educational attainment than those who did not.

In the third model, in addition to previous predictors, adaptation/assimilation variables length of stay and English proficiency were added. Surprisingly, at odds with our hypothesis the length of stay in the United States was inversely associated with educational level. As expected, a higher level of English proficiency was associated with an increase in educational attainment. This model explains 23.3% of the variance in the level of educational attainment.

Finally, region, a proxy for immigrant community concentration, was entered in Model 4. This model explains 24.5% of the variance in educational attainment. Since the R² of Model 4 is not much different from that of Model 3, we did a special F test. The result indicates that the increase in R² in Model 4 is statistically significant at the 0.05 level. Hence, Model 4 is the best fitting model with the most accurate estimates. In this model, all other predictors are significant, except for survey year and the dummy variable for the Midwest. This model reveals some regional differences in educational attainment among Bangladeshi immigrants. Contrary to our hypothesis, foreign-born Bangladeshis in the West and the South were 0.788 and 0.710, respectively, higher in the level of education than their counterparts living in the Northeast, but there was no significant difference between those living in the Midwest and those in the Northeast in educational attainment. The effects of other predictors in the model are very similar to those in Models 1–3 but are most accurate. Specifically, an older age was associated with 0.016 lower level of schooling completed. Females had 0.616 level less schooling than males. Those who were married with their spouse in the U.S. had a 0.949 level more education than those who were not married or whose spouses were not present with them. Those who had children were 0.504 levels lower in education than those who did not have children. Length of U.S. residence exhibited a negative association with educational attainment; each additional year of U.S. residence slightly decreased educational attainment by 0.009 level. Each level of English proficiency helped increase education by 1.250 levels. In Model 4, English proficiency is the strongest predictor (β = 0.392).

Occupation

The existing literature has seldom systematically examined the occupational patterns of Bangladeshi immigrants. For example, we do not know how likely they hold a white-collar occupation or to be professionals or managers. We examine both measures to assess the determinants of occupational attainment. Since both dependent variables are dichotomous, logistic regression was used.

Table 7 shows the effects of the predictors on the probability of holding a professional or managerial occupation. We tested five nested sequential logistic regression models by including or adding the following predictors in sequence: demographic variables (Model 1), marital status and having children (Model 2), English proficiency and the length of stay in the U.S. (Model 3), region (Model 4), and educational attainment (Model 5). This sequential modeling technique enables the assessment of how variance explained in the dependent variable changes by adding additional predictors. Significant values of model χ² indicate that all models fit the data well. The pseudo R² of Model 5 is 0.248, which is the highest among all five models and indicates the predictors explain 24.8% of the variation in the likelihood of holding a professional or managerial occupation. Model 5 also has the least log-likelihood and the highest model χ² value. Therefore, Model 5 appears to be the best-fitting model. Our interpretations focus on Model 5.

Table 7 Logistic Regression Estimates Predicting the Probability of Professional or Managerial Job, Bangladeshi Immigrants Aged 16–64, 2001–2019 ACS (Standard Errors in Parentheses; OR = Odds Ratio)

In Model 5, every later year surveyed was associated with an increase of 2.1% in the likelihood of holding a professional or managerial occupation. As expected, the relationship between age and the likelihood of professional or managerial occupation is negative; for each year increase, the odds of Bangladeshi immigrants holding professional or managerial occupation decreased by 1.7% (0.983 – 1 = −0 0.017). The gender variable is not significant, indicating no significant gender difference in the likelihood of holding a professional or managerial job. Unexpectedly, immigrants who had children were 20% ( 0.800 − 1 = −0.20) less likely to hold a professional or managerial job than those who did not have children, holding other variables constant. Marital status was significant in Models 1–4 but lost statistical significance in Model 5. Hence, there was no significant difference between those who were married with their spouses in the U.S. and those who were not married or did not have spouse present in the U.S. in the likelihood of holding a professional or managerial job. Consistent with our hypothesis, each year increase in the length of U.S. residence increased the likelihood of holding a professional or managerial occupation by 2.3% (1.023 – 1 = 0.023). Each level increase in English proficiency resulted in about 60% increase in the likelihood of holding a professional or managerial occupation. At odds with our hypothesis, Bangladeshi immigrants who resided in the Midwest, South, and West were 86.3%, 44.8%, and 79.9%, respectively, more likely to hold a professional or managerial occupation than those living in the Northeast. As anticipated, education had a significant positive effect on the likelihood of holding a professional or managerial position.

Table 8 presents the effects of the predictors on the probability of holding a white-collar job. The very same five nested logistic regression models in predicting professional or managerial job as shown in Table 7 except for the dependent variable were tested. Significant values of χ² indicate that all models fit the data well. The pseudo R² of Model 5 (0.173) is the highest among all five models and indicates the predictors in this model explain 17.3% of the variation in the likelihood of holding a white-collar occupation. Model 5 also has the least log-likelihood and the highest model χ² value. Thus, Model 5 appears to be the best-fitting model and is therefore the focus of interpretations.

Table 8 Logistic Regression Estimates Predicting the Probability of White-collar Occupational Attainment, Bangladeshi Immigrants Aged 16–64, 2001–2019 ACS (Standard Errors in Parentheses; OR = Odds Ratio)

As shown in Model 5 as well as in all other models, age is a significant predictor. As expected, the relationship between age and the likelihood of holding a white-collar occupation is negative; for each year increase, the odds of Bangladeshi immigrants holding a white-collar occupation decreased by 1.8% (0.982 – 1 = −0 0.018). Unexpectedly, female Bangladeshi immigrants were twice as likely as their male counterparts to hold a white-collar job. Also contrary to our hypothesis, Bangladeshi immigrants who had children were about 24% (0.763 – 1 = −0.237) less likely to hold a white-collar occupation than those who did not have children, holding other variables constant. Married immigrants with the U.S.-based spouse were more likely than those unmarried or married without a spouse present to hold a white-collar occupation. As anticipated, each level increase in English proficiency was associated with about 37% increase in the likelihood of holding a white-collar occupation. Bangladeshi immigrants residing in the South and West were 98% and 104%, respectively, more likely to hold a white-collar occupation than those living in the Northeast, but the difference between those in the Midwest and the Northeast was not significant at the 0.05 level. As hypothesized, each level increase in educational attainment resulted in 30% (1.303 – 1 = 0.30) increase in the likelihood of holding a white-collar occupation. More recent Bangladeshi immigrants were less likely than earlier immigrants to get a white-collar job.

Do determinants of professional/managerial jobs differ from those of white-collar jobs, or are they the same? To answer this question, we compare Tables 7 and 8. The examination of regression coefficients reveals, for the most part, no change in the direction and significance of the relationships between the dependent variable and independent variables. The variable gender is an exception. It indicates that women were more likely than men to do white-collar jobs. However, the gender difference in holding a professional/managerial job was not statistically significant. Regional differences in holding a professional/managerial job suggest that immigrants in the Midwest, South, and West had a higher rate of holding a professional/managerial job than immigrants in the Northeast. The direction for white-collar jobs is the same but the regional difference for the Midwest dummy variable does not reach the level of significance. These small variations notwithstanding, the results indicate that among Bangladeshi immigrants, determinants of white-collar status appear to be similar to determinants of holding a professional/managerial job.

Self-Employment

Table 9 shows the effects of the predictors on the probability of self-employment. We tested the same five nested logistic regression models shown in Tables 7 and 8, except for the replacement of the dependent variable. Significant values of model χ² indicate that all models fit the data well. The pseudo R² of Model 5 (= 0.051) is the highest among all five models and indicates the predictors explain 5.1% of the variation in the likelihood to self-employment. Model 5 also has the smallest log-likelihood and the highest model χ² value. Hence, Model 5 appears to be the best-fitting model and thus the focus of interpretations.

Table 9 Logistic Regression Estimates Predicting the Probability of Self-Employment, Bangladeshi Immigrants Aged 16–64, 2001–2019 ACS (Standard Errors in Parentheses; OR = Odds Ratio)

In Model 5, age is positively associated with the likelihood of self-employment; for each year increase, the odds of self-employment for Bangladeshi immigrants increased by 1.6% (1.016 – 1 = 0.016). Given their limited employment opportunities in the mainstream labor market, it is not surprising that older immigrants were more likely to be self-employed than their younger counterparts. As expected, females had a lower chance of self-employment than males by about 66% (0.342 – 1 = −0.658). Holding other variables constant, immigrants with children were 33% (1.331 – 1 = 0.331) more likely to be self-employed than those without children. Married immigrants with a U.S.-based spouse were about 30% more likely than those unmarried or married without a spouse present to be self-employed (1.296 – 1 = 0.296). Coinciding with our hypothesis, each year increase in staying in the U.S. increased the likelihood of self-employment among immigrants by 2.3% (1.023 – 1 = 0.023). Each level increase in English proficiency resulted in about 15% (0.852 – 1 = −0.148) decline in the likelihood of being self-employed. Bangladeshi immigrants residing in the Midwest and the South were 47% and 12%, respectively, less likely to be self-employed than those in the Northeast, but immigrants in the West were not significantly different from those in the Northeast in self-employment. Since the majority of Bangladeshi immigrants (more than 50%) live in the Northeast, this region has a long history of Bangladeshi immigration and thus might offer more employment opportunities to new immigrants than other regions where immigrant networks are smaller. The findings also demonstrate that self-employment declines with education. Each level of educational attainment was associated with a 5.1% drop in the chance of becoming self-employed (0.949 – 1 = −0.051). The reason behind this negative relationship is that probably better education and better English language skills may reduce the need for self-employment by increasing the chances of finding professional employment. Survey year is not significant.

Income

Income is the ultimate measurement of socioeconomic achievement as it reflects the outcome of educational and occupational attainment. As mentioned earlier, income was log-transformed to correct for skewness for the regression analysis. To test changes in the variance explained in the dependent variable, we tested five nested sequential regression models (see Table 10). Model 1 includes basic demographic variables (i.e., age, age², gender, and survey year). Age² is included in order to test the non-linear relationship between age and income. All predictors are significant. This model explains 23.5% of the variance in income. Model 2 adds dummy variables for marital status and having children to Model 1. Both married status and having children are statistically significant. The total variance explained by Model 2 increases to 24.6%. Model 3 adds length of U.S. residency and English proficiency to Model 2. Adding these two assimilation variables increases the variance explained to 30.5%. Both predictors are highly significant. Model 4 adds region to Model 3 and explains slightly more variance in personal income (30.7%) than Model 3. Finally, Model 5 adds educational attainment, managerial-professional occupation, and self-employment to Model 4. Model 5 is statistically significant (F = 611.406, p ≤ 0.001) and is the best-fitting model with the highest total variance explained of 33.2%. Hence, the results of this model are the focus of interpretations.

Table 10 Estimates of OLS Regression Models Predicting Logged Income of Bangladeshi Immigrants Aged 16 or Older, 2001–2019 ACS (Standard Errors in Parentheses)

As shown in Model 5, the significant positive coefficient for age and the significant negative coefficient for age² suggest a parabolic relationship between age and income, namely, the influence of age increased as people become older but declined after a certain point. This is in line with our hypothesis. Also consistent with our hypothesis, foreign-born Bangladeshi women tended to have a lower personal income (3.14 times less) than their male counterparts, controlling for other variables. The survey year is no longer statistically significant after other variables are held constant. Unlike the result in Models 2, 3, and 4, those who were married with their spouses in the U.S. were not significantly different from those who were not married or whose spouses were not present with them after other predictors are controlled. Unexpectedly, those who had children on average made 67.5% less income than those who did not have children. Consistent with our hypothesis, length of stay in the U.S. had a positive relationship with income; each year increase in U.S. residence was associated with 5.9% increase in income, holding other variables constant. As expected, English proficiency helped increase income; each level increase in English proficiency was associated with 55.6% increase in income. Region had a mixed effect on income. Bangladeshis in the South on average earned 24.8% more than their counterparts in the Northeast, but those in the Midwest and West were not significantly different from those in the Northeast. Coinciding with our hypothesis, each level increase in education was expected to increase personal income by 21.5%. As anticipated, managerial or professional workers on average made 60.4% more than non-managerial or nonprofessional workers. English language, education, and occupational status are all major drivers of income and work in the expected direction. However, self-employment tended to have a negative effect on income, as self-employed Bangladeshi immigrants were expected to earn 39% less than their non-self-employed counterparts.

Discussion

Despite the tremendous growth of Bangladeshi immigrants in the United States in recent years, systematic research on their socioeconomic adaptation is lacking, and existing portraits of their socioeconomic attainment are contradictory. Using the pooled samples from the 2001–2019 American Community Surveys and various regression models, this study analyzes the socioeconomic adaptation of foreign-born Bangladeshis by focusing on the most common indicators—educational attainment, occupational status, self-employment, and income. The results reveal that, in general, Bangladeshi immigrants are well educated and include a significant proportion of professionals, managers, and other white-collar workers, but their average income is very low. The regression analyses show that many demographic, familial, assimilation, and socioeconomic factors contribute significantly to the socioeconomic adaptation of Bangladeshi immigrants. Our findings provide a new portrait of Bangladeshi immigrants’ socioeconomic adaptation in the U.S. based on the latest generalizable data and new evidence on the determinants of their socioeconomic adaptation. The findings also have significant implications for scholarship and policy.

Our descriptive and comparative analysis indicates that on average foreign-born Bangladeshis were considerably more educated than the general U.S. population and foreign-born Chinese and Vietnamese, but less educated than foreign-born Japanese, Koreans, Filipinos, and Indians. The average education level of Bangladeshi immigrants was close to 14 years of schooling. Foreign-born Bangladeshis fared better in occupational attainment (with 36.3% managerial-professional workers and 66.8% white-collar workers) than the general U.S. population and foreign-born Vietnamese, but worse than foreign-born Chinese, Japanese, Koreans, Filipinos, and Indians. Foreign-born Bangladeshis’ average personal income was $29,484, which was the lowest compared to those of the general U.S. population and foreign-born Chinese, Japanese, Koreans, Filipinos, Vietnamese, and Indians. Relative to their education level and occupation status, Bangladeshi immigrants’ personal income was very low.

Our findings challenge Zhao’s (2014) claim that Bangladeshis in the U.S. had a relatively low socioeconomic status. Our results also contest the Migration Policy Institute’s (2014) conclusions based on the data from the CPS of 2009–2013 that foreign-born Bangladeshis were identical to the general U.S. population in the percentage of professional or managerial occupations and fared slightly better in income than the general U.S. population. More or less consistent with the statistics in Kibria’s (2006) study based on the 5% PUMS of the 2000 Census, our findings offer a mixed portrait of relatively high education and occupational attainment together with a very low income among foreign-born Bangladeshis in the United States. However, unlike Kibria’s (2006) study, our study relies on the latest pooled 2001–2019 ACS data with a very large sample size, so the results are up-to-date and trustworthy.

The inconsistency between Bangladeshi immigrants’ relatively high education and occupation status and their very low income begs the question of why. This disparity can be explained by a variety of factors. One important factor is the transferability of foreign credentials and foreign-trained skills. Immigrants’ education and skills acquired in their home country are less valued than what they acquire in their host country (Arbeit & Warren, 2013; Bratsberg & Ragan, Jr. 2002; Kaushal, 2011). According to Zeng and Xie (2004), Asian Americans' income is adversely affected by their place of education; Asian immigrants with an American education had a higher income than those who completed their school prior to immigration to the United States. Employers in the United States have depreciated academic credentials and professional experience from Bangladesh. This depreciation has prevented Bangladeshis from effectively converting their pre-migration abilities and expertise to the U.S. labor market (Kibria, 2013). Another factor is over-qualification, which is defined as being hired in a position that is below one’s education-based skills and experience. Over-qualification is associated with downward occupational mobility experienced by immigrants, including Bangladeshis (Cheng & Yang, 1996; Potochnick & Hall, 2021; Sirkeci et al., 2014). Bangladeshi immigrants were placed at the lowest tiers of the U.S. labor market despite their skills and professional positions in their home country. A third factor is a lack of U.S. job experience (Zeng & Xie, 2004). Since Bangladeshis are new immigrants, their limited experience in the U.S. job market may partly explain their lower pay. Finally, age at the time of immigration may impact the wages of immigrants with an older age at the time of arrival associated with a lower pay (O’Neil & Tienda, 2014; Schaafsma & Sweetman, 2001).

This study also assessed the determinants of Bangladeshi immigrants’ socioeconomic adaptation. The effect of age partially confirms our hypothesis that older Bangladeshi immigrants adapt less well socioeconomically than their younger counterparts. The results suggest that older immigrants tended to be less educated and less likely to hold white-collar and professional/managerial occupations. However, older immigrants tended to have a higher income up to certain point and be self-employed than younger immigrants. Women seemed to adapt less well socioeconomically than men. Female immigrants tended to be less educated, had a lower income, and had a lower propensity for self-employment, but they were more likely to hold white-collar jobs than men. Married immigrants with a spouse present tended to have a higher level of education and a higher likelihood of self-employment than unmarried immigrants or married women without a spouse present. Immigrants with children had a lower education, income, and occupational status, but were more likely to be self-employed.

Bangladeshi immigrants who had been in the United States longer were more likely to have a professional/managerial or white-collar position than those who had been in the country for a shorter period of time. In addition, the longer Bangladeshi immigrants lived in the United States, the higher their income. This is congruent with the findings of previous empirical research, which indicate that immigrants who lived longer in the United States adapted socioeconomically better than those who lived shorter in the country (Chiswick, 1978; Jasinskaja-Lahti, 2008; Logan & Drew, 2011; Michalikova & Yang, 2016; Portes & Rumbaut, 2014). It is worth mentioning that the variable years in the U.S. recoded from the variable year of immigration in the ACS also measures immigration cohort. Largely consistent with the literature (e.g., Borjas, 2015; Lubotsky, 2011), our regression analyses show significant effects of immigration cohort on the likelihood of obtaining a professional-managerial job, obtaining a white-collar job, and being self-employed and on income attainment (see Tables 7–10). Earlier cohorts appeared to fare better economically than more recent cohorts. Our result shows that unexpectedly a longer U.S. residency was associated with a lower level of education attainment; namely, earlier Bangladeshi immigrants tended to have less education than more recent Bangladeshi immigrants. This is probably due to the selective migration of more educated Bangladeshis in more recent years. This result is at odds with the findings of the economic literature, which lumps all immigrant groups together without differentiating between Asian immigrants and immigrants from Latin America, that recent immigrants have lower levels of education than earlier immigrants and native-born citizens (e.g., Borjas, 1990, 1992). This finding challenges classic assimilation theory and suggests the importance of selective migration rather than the assimilation effect on educational attainment for Bangladeshi immigrants. This result may also suggest that many Bangladeshi immigrants may have completed their education in their home country and may not have continued their education in the United States.

Our results show regional differences in socioeconomic adaptation. Bangladeshi immigrants in the Midwest and the South had a lower propensity for self-employment than their counterparts in the Northeast. Contradictory to our expectations, immigrants in regions other than the Northeast tended to have higher occupational attainment than those in the Northeast. Immigrants in the South tended to have a higher income than their counterparts in the Northeast. Immigrants in the South and West tended to have a higher level of educational attainment than the immigrants in the Northeast. Except for the effect of region on self-employment, the overall pattern is at odds with our expectation that immigrants in the Northeast tended to adapt better socioeconomically than immigrants in other regions. Perhaps, the concentration of Bangladeshi immigrants in the Northeast diversifies their socioeconomic quality, leading to the observed outcomes.

Existing studies found that human capital, such as English language skills and education, facilitates socioeconomic adaptation (e.g., Chiswick, 1991; Espenshade & Fu, 1997; Lueck, 2017; Rivera-Batiz, 1990; Shields & Price, 2002; Xie & Gough, 2011). Highly consistent with our hypothesis and with the findings of immigration studies, English proficiency largely influenced the socioeconomic adaptation of Bangladeshi immigrants. A higher level of English proficiency was associated with a higher education, a higher occupational status, and a higher income. Education is a strong predictor of economic adaptation. As expected, education increased Bangladeshi immigrants’ occupational attainment and income. However, better education and English proficiency were associated with a lower propensity for self-employment. Since education and English proficiency increase the likelihood of obtaining a professional/managerial job, both reduce the probability of self-employment (Raza & Sakamoto, 2024). Our results confirm the common sense that occupational status affects income. Managerial-professional workers were more likely to earn higher incomes. Nonetheless, self-employment decreased Bangladeshi immigrants’ income. Numerous studies in the entrepreneurship literature have established a two-pronged effect of entrepreneurship on individual income, revealing that entrepreneurship is a source of greater economic mobility for some, but results in lower-than-average incomes for the bulk of self-employed workers (Halvarsson et al., 2018; Hamilton, 2000; Lofstrom & Wang, 2019). Our result is in line with the latter for Bangladeshi immigrants.

The impacts of survey year on Bangladeshi immigrants’ socioeconomic adaptation deserve to be noted. Survey year had little impact on the educational attainment of foreign-born Bangladeshis. A later year was associated with a significantly higher likelihood of obtaining a professional or managerial job but a relatively lower probability of doing a white-collar job for foreign-born Bangladeshis. Survey year had no effect on the likelihood of self-employment for foreign-born Bangladeshis. For income, the effect of year was significant with basic demographic variables included; however, once familial variables, assimilation variables, region, education, and occupation were introduced into the regression models, year lost statistical significance, indicating it was not a significant predictor of income. Overall, survey year did not have much impact on most outcome variables.

Our study can help assess the utility of immigrant adaptation theories. Our evidence does not lend support to classic assimilation theory. This is so because as first generation immigrants foreign-born Bangladeshis do not start everything in socioeconomic attainment from the bottom and do not go through straight-line upward mobility. Their experience is unique. They fare better than the general U.S. population in educational and occupational attainment and better than some Asian immigrant groups such as Chinese and Vietnamese in educational attainment and Vietnamese and Filipinos in occupational attainment. However, in terms of income attainment they are at the bottom for every comparison despite their respectful educational and occupational status. Their socioeconomic adaptation experience appears to be different from those of some Asian immigrant groups such as Indochinese, Asian Indians, Japanese, and Chinese as demonstrated by Yang (2011). Because of the distinct adaptation patterns, new assimilation theory may better capture the experience of Bangladeshi immigrants. As Alba and Nee (2003) put, assimilation does not occur linearly from the time of immigrants’ arrival directly into the middle class; instead, the process may be bumpy with ups and downs, but eventually they move up. Assimilation may not be an inevitable consequence for all immigrant groups.

Our findings also refute the myth of “model minority,” as this characterization does not match the experience of Bangladeshi immigrants. In terms of socioeconomic attainment, Bangladeshi immigrants are neither exemplary nor subnormal. To a great extent, they are pretty normal. They may be under-recognized. They are not immune from racial prejudice, discrimination, and animosity. They still need help from the government and other organizations.

Our results about the incompatibility between educational and occupational status and income attainment among Bangladeshi immigrants raise some questions of policy concerns about Bangladeshi immigrants and immigrants in general. What can be done at the individual, community, and institutional levels to reduce such education/occupation-income mismatch or incompatibility? At the individual level, immigrants may do one of several things. One is to purposively opt for training that can maximize income return in the host country either before or after migration. Another option is to obtain additional training in the host country from accredited institutions to increase their marketability. The third is to gain market experience. The last but not least is to improve host language proficiency, which sometimes becomes an income depressor. At the level of immigrant community, training programs may be established to inform immigrants of their rights, to provide information about labor market conditions and income standards, and to teach immigrants job offer negotiation skills. At the institutional level, government agencies and organizations should establish policies that regulate the equivalency of educational credentials and occupational qualifications, and prevent or penalize discrimination against immigrants’ legitimate credentials and job qualifications.

Conclusions

This research makes a significant contribution to the existing literature by comprehensively investigating the socioeconomic adaptation of Bangladeshi immigrants in the United States using the latest, nationally representative data so the findings can be generalized to Bangladeshi immigrants in the United States. This study addresses not only their status of socioeconomic adaptation, but also the factors that influence their socioeconomic adaptation. Moreover, it compares Bangladeshi immigrants with Chinese, Japanese, Korean, Filipino, Vietnamese, and Indian immigrants as well as the general U.S. population in a number of adaptation measures. This study also evaluates the relevance of existing theories to the socioeconomic adaptation experience of Bangladeshi immigrants. The findings indicate that the adaptation experience of Bangladeshi immigrants is inconsistent with classic assimilation theory, but mostly consistent with the new assimilation theory. This study may help comprehend the socioeconomic adaptation process of other new immigrant groups.

This study provides some comparative descriptive statistics of Bangladeshi immigrants and large Asian immigrant groups as well as the general U.S. population in socioeconomic adaptation measures. Nonetheless, because of space constraint and small sample sizes of some Asian immigrant groups, we did not compare Bangladeshi immigrants with some other foreign-born Asian groups such as Cambodians, Laotians, Pakistanis, Thais, Indonesians, and so on. Future research should conduct more systematic cross-group comparisons of socioeconomic adaptation in order to better understand their differences in the socioeconomic adaptation process by including other smaller foreign-born Asian groups and perhaps non-Asian groups. Furthermore, the focus of this study is on the experiences of first-generation Bangladeshi immigrants, but, if possible, future research should examine the socioeconomic adaptation of the second generation, and the extent to which experiences of Bangladeshi immigrants’ children are similar to, or different from the experiences of their immigrant parents.