The Role of Extended Family in Asian American Women’s Employment from 1990 to 2022

EXTENDED FAMILY STRUCTURE AND ASIAN AMERICAN WOMEN’S EMPLOYMENT

The prevalence of extended family structure among Asian Americans is often attributed to Asian culture, which emphasizes traditional family values and filial piety (Burr and Mutchler 1993; Speare and Avery 1993). While declining, the extended family structure is still common in Asia: for example, multigenerational households made up about 23 percent of all households according to the 2010 China Census (Wang 2013). Consistent with cultural explanations, Asian American families are more likely to form extended households even after controlling for demographic and economic characteristics (Kamo 2000). Some further argue that such a cultural effect on extended family living arrangements would last long after immigration, an effect not fully explained by the assimilation model (Kamo and Zhou 1994). Additionally, evidence that foreign-born Asian immigrants’ gains in household earnings from an extended family structure diminish or become negative as their duration in the US increases indicates that household extension can be considered a manifestation of cultural disposition beyond economic considerations (Kulkarni 2019).

Another explanation for household extension among Asian Americans is that it may be a strategy for immigrants to cope under structural constraints (Kulkarni 2019). Studies show that intergenerational coresidence serves as a survival strategy for low-income families that pool economic resources and receive help for childcare (Edin and Lein 1997). Extended households can also serve as a “private safety net” in the face of uncertainties and economic insecurities (Western et al. 2012). In fact, structural pressures including economic factors have been repeatedly identified as one of the major explanations for household structures across various groups and nativity statuses (for example, Glick et al. 1997). This strategy might be particularly relevant to Asian Americans as newcomers, who are often considered perpetual foreigners by US citizens, because shared households can help achieve economies of scale and invest in the human capital of household members (Ermisch 1988; Kaushal and Reimers 2007). Consistent with this perspective, Asian households benefit more from household extension than White households in terms of household-level earnings, and extended family structure is more prevalent among Asian families with lower education and limited English proficiency (Kulkarni 2019). Large family size is also theoretically linked to women’s greater labor force participation by increasing the economic needs of a family for a given level of consumption (Becker 1965; Tienda and Glass 1985).

However, cultural factors and structural strains as explanations for Asian Americans’ household extension (and their potential role in women’s employment) are not necessarily mutually exclusive and are often closely interconnected. For instance, recent research on women’s employment in China still documents gender asymmetries in multigenerational families: whether an extended family structure is conducive to women’s employment often depends on the gender of the live-in-parent (mother versus father) and the family relationship (own parents versus husband’s parents) (Yu and Cheng 2024; Yu and Xie 2018). The importance of parental help, particularly from one’s own mother, in women’s employment reflects broader patterns of traditional gender norms and workplace structures in Asia, in which women bear a disproportionate share of housework and childcare, while fathers’ involvement remains limited (Raymo et al. 2023). Similarly, a qualitative study on middle-class Korean women living in suburban areas of New York finds that full-time employment is often determined by the availability of transgenerational shared mothering, especially for those with young children (Moon 2003). At the same time, household extension can also be interpreted as a response to structural constraints, as evidence shows that Asian American women face limited access to parental leave benefits through work and formal childcare (Brown et al. 2020; Malik et al. 2018). In such contexts, forming a multigenerational household is a resource or a strategy for Asian American women to stay in the labor force, while the pattern of help is influenced by their cultural norms, such as living with one’s own parents being more beneficial (Greenman 2011; Kang and Cohen 2018; Moon 2003).

VARIATION BY ETHNICITY, IMMIGRANT GENERATION, AND TIME

One of the major limitations of the literature on Asian Americans is its failure to consider heterogeneity and differentials within this group. This limitation primarily reflects the lack of data sources that allow the disaggregation of Asian groups that have small population sizes. Nevertheless, the approach of considering Asian Americans as a whole group is problematic because it obscures the socioeconomic and cultural characteristics of various groups that may lead to different patterns of immigrant adaptation and assimilation strategies (Drouhot and Garip 2021). That said, while Asian Americans are considered to hold traditional family values, the organization of family life, including the prevalence and patterns of household extension and intergenerational support, differs across ethnicities. Family extension, for instance, is more common for first-generation Asian American women among Southeast Asians (Vietnamese and Filipina) than among East Asians (Korean and Japanese) (Kang and Cohen 2018; Kulkarni 2019). Evidence further suggests that the patterns of support from additional adults in the household and their impact on women’s work vary by national origin. For example, the burden of caring for disabled adults is negatively associated with women’s employment for Koreans only among seven Asian groups examined (Kang and Cohen 2018).

Additionally, whether the household extension of Asian immigrants is due to cultural factors or structural strains, the influence of these factors may not be constant across different immigrant generations. First-generation immigrants bring their cultural values and national traditions from their country of origin, both of which influence their family lives (Kamo and Zhou 1994; Kang and Cohen 2018). However, the process of acculturation leads to exposure to and adoption of the cultural values of the US society among the later generations (Gordon 1964). When it comes to living arrangements of the elderly in the US, independent residence is the most prevalent form. According to an analysis of the 2021 Current Population Survey, only 7.6 percent of parents aged sixty and over coreside with adult children aged forty or over (Eickmeyer and Brown 2019). Therefore, cultural preference for extended family is likely to attenuate for later immigrant generations in a way that it converges with that of the majority group (non-Hispanic Whites).

Considering differentials across immigrant generations in the patterns of household extension is also important, since structural constraints that may motivate extended living arrangements are not shared equally by different generations. The economic necessities of elderly parents, which are structurally conditioned in US society (for instance, eligibility for health and pension benefits), may be a strong driver of intergenerational coresidence among first-generation Asian Americans. A great majority of Asian American elders (aged sixty-five and over) are foreign-born and have migrated at a more advanced age due mostly to family reunification (Mui et al. 2007) and are likely economically dependent on their children (Mui and Shibusawa 2008). However, later generations of Asian Americans tend to achieve economic parity with—or even surpass—US-born Whites in outcomes such as hourly wages and annual earnings, largely due to their high level of educational attainment (Hirschman and Wong 1984; Wang et al. 2017; Kim and Sakamoto 2010; Zeng and Xie 2004).

Some evidence suggests a White-Asian disparity for the non-US-born immigrants (Kim and Zhao 2014; Shao 2023). One explanation is that they have completed their education outside of the US, where educational credentials are less likely to be similarly recognized (Zeng and Xie 2004). Another explanation points out the potential harm of the model minority myth since it penalizes Asian Americans who do not fit the stereotype of overachieving Asians, including men without a college degree (Kim and Sakamoto 2014). Andrew Taeho Kim and ChangHwan Kim’s study (2026, this issue), for example, documents earnings disadvantages of less-educated Asian American men, especially in local labor markets where the proportion of highly educated Asian Americans is high. Despite some discrepancies in their findings, prior studies consistently show that the nativity, which is associated with the place of education, is an important factor in determining Asian Americans’ labor market outcomes (Kim and Zhao 2014; Zeng and Xie 2004).

Lastly, the characteristics of Asian Americans and the US socioeconomic and cultural context surrounding them may differ across time. Compared to immigrants of earlier periods, recent Asian immigrants are increasingly better educated and more skilled, contributing to the image of a model minority. These changing compositional characteristics of Asian immigrants reflect multiple factors, including political and economic development in their origin countries and the US immigration policies and labor market context. Reflecting the recent trends of skill-biased labor market polarization and globalization (Autor et al. 2006), new immigrants from Asia are most likely to become legal permanent residents (LPRs) through employment-based preferences among all LPRs according to recent data (Hanna and Batalova 2021). Some argue that the increasing number of middle-class Asian immigrants in recent decades has contributed to the diversification and polarization of class among Asian Americans (Hu-DeHart 1999; Ong et al. 1994). To the extent that compositional characteristics of Asian immigrants as well as the US socioeconomic and cultural context differ across time, the role of extended households in Asian women’s employment can also be expected to vary over different periods.

DATA AND METHODS

We use data from the US Census (1990, 2000) and the American Community Survey (2001–2022), provided by the Integrated Public Use Microdata Series (IPUMS) (Ruggles et al. 2024), to examine ethnic and generational variations as well as changes over time. Our analytic sample includes women of working age, between twenty-five and forty-nine years, who are either the head of household or the spouse of the head of household.¹ By doing so, we can approximate adult children “hosting” their parents in multigenerational households and evaluate whether parental coresidence is associated with women’s likelihood of working (rather than working while living as a dependent) (for example, Kang and Cohen 2018). Our focus on women between twenty-five and forty-nine years of age aligns with a stage in the life course when work and family responsibilities often conflict, particularly due to the demands of motherhood (Greenman 2011). As discussed above, the presence of parents or parents-in-law may have a different impact on women’s work depending on family conditions that vary by life course. For example, extended family living arrangements would be conducive to women’s employment if the elderly parent(s) can help with housework and childcare. However, living with an elderly parent who needs care may negatively affect women’s work (Kang and Cohen 2018). By restricting the analytic sample to women between twenty-five and forty-nine years of age and including a control for parental health status, we aim to minimize the cases in which the presence of an elderly parent is negatively associated with women’s work.²

Empirical Strategy

We estimate a series of logistic regression models that predict women’s employment as a function of the extended family structure while controlling for various compositional characteristics. Specifically, we examine how the association between coresidence with parents or parents-in-law and women’s employment varies by ethnicity across immigrant generations over different time points spanning from 1990 to 2022. We do so by assessing four-way interaction terms among these statuses: parental coresidence, detailed Asian ethnicity, immigrant generation, and time period. Appendix tables A.1 and A.2 demonstrate the importance of accounting for these characteristics by ethnicity, immigrant generation, and time, as the sociodemographic characteristics of Asian American and non-Hispanic White women have shifted substantially over the past several decades.

US-born, non-Hispanic White women who do not live with parents or parents-in-law are the reference group, as White women in extended families are very selective (see figure 1). Additionally, we do not account for differences by ethnicity and immigration status of White women as we want to examine the role of the extended family structure in Asian women’s employment, and any differentials across ethnicity and generations, in comparison with that of the majority group (non-Hispanic Whites) in the US, following the approach adopted by prior research (Kim and Sakamoto 2010; Shao 2023).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 1.

Prevalence of Extended Family by Ethnicity, Immigrant Generation, and Period

Source: 1990 Census; 2000 Census; ACS 2001–2022.

Note: Extended family refers to a household that includes at least one parent or parent-in-law. In this figure, “2000s” refers to the ACS data from 2001 to 2009, “2010s” refers to the ACS data from 2010 to 2019, and “2020s” refers to the ACS data from 2020 to 2022. Statistics are weighted using individual weights. For US-born, non-Hispanic White women, immigrant generation is not differentiated. NH = non-Hispanic.

RESULTS

We first describe patterns in the prevalence of extended family arrangements and employment rates in the analytic sample (figures 1–2). We then present the results from multivariate logistic regression analyses examining employment gaps between Asian and White women by family structure (figures 3–6).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 2.

Employment Rates by Ethnicity, Immigrant Generation, and Family Structure

Source: 1990 Census; 2000 Census; ACS 2001–2022.

Note: Extended family refers to a household that includes at least one parent or parent-in-law. In this figure, “2000s” refers to the ACS data from 2001 to 2009, “2010s” refers to the ACS data from 2010 to 2019, and “2020s” refers to the ACS data from 2020 to 2022. Statistics are weighted using individual weights. For US-born, non-Hispanic White women, immigrant generation is not differentiated. All group differences by family type are statistically significant at the 95 percent level unless noted as n.s., which stands for non-significant difference at the 95 percent confidence level. NH = non-Hispanic.

Prevalence of Extended Family by Ethnicity, Immigrant Generation, and Period

Figure 1 shows the prevalence of extended family households for five Asian groups—Asian Indian, Chinese, Filipino, Korean, and Vietnamese—and non-Hispanic Whites, from 1990 to the early 2020s, separately by immigrant generation. As noted, the values for extended family among US-born, non-Hispanic Whites are shown without differentiating generation.

Across all periods, the percentage of extended families is constant for non-Hispanic Whites at around 2–3 percent. Compared to US-born, non-Hispanic Whites, all Asian groups have a higher prevalence of extended households at all time points. This pattern holds for both the first- and 2+ generations, although the overall level of household extension is lower for the 2+ generation Asian Americans. As for trends over time, the percentage of the 2+ generation Asian women in extended households tends to increase after 2010, consistent with the growing trend of multigenerational households in the US in general (Cohn et al. 2022).

Additionally, figure 1 reveals some subgroup differentials. For instance, among the first generation, Korean women have the lowest level of coresidence with parents or parents-in-law. They even show a downward trend in coresidence while most groups in the first generation have more or less a similar prevalence of extended family across different periods. On the other end of the spectrum, Vietnamese women are more likely to be in extended households in recent decades, with about one in five first-generation women living in such circumstances around 2020. This upward trend is also clear for the 2+ generation of Vietnamese women. These ethnic differences in extended family living arrangements are consistent with findings documented in prior research (Kang and Cohen 2018; Kulkarni 2019).

Employment Rates by Ethnicity, Immigrant Generation, and Family Structure

Next, figure 2 shows women’s employment rates by ethnicity, immigrant generation, and family structure. Compared to figure 1, we observe relatively modest differences in employment rates between Asian groups and White women. Notably, Asian women in extended families are more likely to be employed compared to those in nuclear families for both the first- and 2+ generations. Specifically, first-generation Asian women living with parents or parents-in-law have much higher employment rates than those not in extended households. This pattern is found across all groups (p < 0.05). While the magnitude of the difference is smaller, the employment advantage from extended families persists for Koreans, Vietnamese, and Filipinas among the 2+ generation (p < 0.05). For Chinese and Asian Indian women, the group differences are not statistically significant. In contrast, White women’s employment rates are consistently lower when they coreside with parents or parents-in-law although group differences by household type are smaller than those of Asian groups. This, again, reflects that extended families are uncommon among White families (Lofquist et al. 2012) and that their patterns of intergenerational support in extended households (for example, an adult child caring for a sick parent compared to an elderly parent helping with childcare) might be different from those of Asian families.

Additionally, there is notable within-group heterogeneity among first-generation Asian women. Filipinas, for instance, have the highest employment rate of all Asian groups, surpassing Whites. At the same time, first generation Korean and Asian Indian women have the lowest employment rates, especially those in nuclear families (figure 2, left panel); however, the 2+ generation of these groups has higher employment rates than non-Hispanic White women, showing the most dramatic changes across immigration generations. When it comes to the 2+ generation, Asian American women in general have employment rates similar to or higher than those of White women. The parity or advantage of labor market activities of US-born Asian Americans is consistent with prior research (Kim and Zhao 2014).

Employment Gap Between Asian and White Women by Family Structure

Figures 3–6 show results from the logit models to predict women’s employment status as a function of family structure (extended family or not). For easier interpretation, we present the results in terms of the differences in the predicted probability of employment for a given Asian group relative to non-Hispanic White women who do not coreside with parents or parents-in-law (the reference group). That is, the dashed horizontal line, which is set at zero, refers to non-Hispanic White women, and each line represents the difference in employment probability of the group from the employment probability of non-Hispanic White women. Thus, any positive values (higher than the dashed horizontal line) indicate a group’s higher employment rates, and negative values indicate lower employment rates in comparison to non-Hispanic White women. Additionally, the circled dot represents statistical significance (between a specific Asian group and White women at a particular time point, p < 0.5), and the x-mark represents non-significance.⁷ All models control for demographic and socioeconomic characteristics. Predicted values were calculated while holding control variables (such as the level of education, age, and household income) constant at their observed distributions.

• Download figure
• Open in new tab
• Download PowerPoint

Figure 3.

Differences in the Predicted Probability of Employment Between Asian and White Women

Source: 1990 Census; 2000 Census; ACS 2001–2022.

Note: Extended family refers to a household that includes at least one parent or parent-in-law. In this figure, “2000s” refers to the ACS data from 2001 to 2009, “2010s” refers to the ACS data from 2010 to 2019, and “2020s” refers to the ACS data from 2020 to 2022. The figure displays differences in predicted probabilities of employment between US-born, non-Hispanic White women (represented by the dashed horizontal line) and various Asian groups. Circled dots indicate statistically significant differences between a specific Asian group and White women at the 95 percent confidence level, while x-marks represent non-significant differences. Predicted values are calculated while holding all control variables, including age, census region, education, marital status, parenthood status, and household income, constant at their observed distributions.

• Download figure
• Open in new tab
• Download PowerPoint

Figure 4.

Differences in the Predicted Probability of Employment Among Women with Young Children

Source: 1990 Census; 2000 Census; ACS 2001–2022.

Note: Extended family refers to a household that includes at least one parent or parent-in-law. In this figure, “2000s” refers to the ACS data from 2001 to 2009, “2010s” refers to the ACS data from 2010 to 2019, and “2020s” refers to the ACS data from 2020 to 2022. The presence of young children is defined as having at least one child under the age of five. The figure displays differences in predicted probabilities of employment between US-born, non-Hispanic White women (represented by the dashed horizontal line) and various Asian groups. Circled dots indicate statistically significant difference between a specific Asian group and White women at the 95 percent confidence level, while x-marks represent non-significant differences. Predicted values are calculated while holding all control variables, including age, census region, education, marital status, parenthood status, and household income, constant at their observed distributions.

• Download figure
• Open in new tab
• Download PowerPoint

Figure 5.

Differences in the Predicted Probability of Employment by Type of Extended Family

Source: 1990 Census; 2000 Census; ACS 2001–2022.

Note: Extended family refers to a household that includes at least one parent or parent-in-law. In this figure, “2000s” refers to the ACS data from 2001 to 2009, “2010s” refers to the ACS data from 2010 to 2019, and “2020s” refers to the ACS data from 2020 to 2022. The figure displays differences in predicted probabilities of employment between US-born, non-Hispanic White women (represented by the dashed horizontal line) and various Asian groups. Circled dots indicate statistically significant difference between a specific Asian group and White women at the 95 percent confidence level, while x-marks represent non-significant differences. Predicted values are calculated while holding all control variables, including age, census region, education, marital status, parenthood status, and household income, constant at their observed distributions.

• Download figure
• Open in new tab
• Download PowerPoint

Figure 6.

Differences in the Predicted Probability of Employment by Household Income Level

Source: 1990 Census; 2000 Census; ACS 2001–2022.

Note: Extended family refers to a household that includes at least one parent or parent-in-law. In this figure, “2000s” refers to the ACS data from 2001 to 2009, “2010s” refers to the ACS data from 2010 to 2019, and “2020s” refers to the ACS data from 2020 to 2022. The figure displays differences in predicted probabilities of employment between US-born, non-Hispanic White women (represented by the dashed horizontal line) and various Asian groups. Circled dots indicate statistically significant difference between a specific Asian group and White women at the 95 percent confidence level, while x-marks represent non-significant differences. Predicted values are calculated while holding all control variables, including age, census region, education, marital status, parenthood status, and household income, constant at their observed distributions.

Results from figure 3 illustrate the employment-promoting role of extended family arrangements, showing that Asian women generally have a higher probability of employment (than White women) when coresiding with parents or parents-in-law. This pattern is consistent across generations and different time points, while it is more pronounced for the first generation. It is important to note that this positive association between household extension and Asian women’s employment is found net of various socioeconomic (such as education and household income) and demographic (including age, parenthood, and marital status) characteristics (Kang and Cohen 2018), as well as coresiding parents’ health conditions (Kang and Cohen 2018).

At the same time, the importance of extended family structure in women’s employment differs by ethnicity. Among the first generation, the presence of parents or parents-in-law appears to be decisive for Chinese women and Filipinas, as their employment rates in extended households are significantly higher than those for Whites, while employment rates in nuclear households are lower (and comparable for Filipinas). Intergenerational coresidence is also important for Korean and Asian Indian women—their employment rates are much lower than White women’s when they do not live in extended households, and the employment gap for those in nuclear families has widened in recent decades. Given the low prevalence of extended families among first-generation Korean and Asian Indian women (figure 1), it suggests that the employment rates would be higher for these groups if they had similar levels of extended households to other Asian groups. Additionally, Southeast Asian women (Filipinas and Vietnamese) in extended households have the highest employment rates across all groups, including Whites. In contrast, employment rates of those in nuclear households do not differ from those of White women for Filipinas and are lower for Vietnamese women in earlier periods.

Turning to the 2+ generations (figure 3, right panel), the most noticeable difference from the first generation is the disappearance or reduction of the White–Asian employment gap across family structures. Overall, differences in employment rates between Asian and White women become smaller or statistically nonsignificant in many cases. That is, the important role of extended families in keeping first-generation Asian women in the labor market (for example, Chinese and Filipina) is less pronounced for the US-born or US-educated Asian women. Still, an extended family structure seems to be conducive to Asian women’s employment: Asian Indian and Korean women not living with parents or parents-in-law, for example, have significantly lower employment rates than White women, especially in earlier periods. However, those in extended family households have comparable rates. Additionally, Filipinas are more likely than White women to work regardless of family structure, but the employment gap is larger for those in extended families.

As discussed earlier, evidence that Asian women’s labor supply is less disrupted by motherhood than White women’s (Greenman 2011; Greenman and Xie 2008) suggests the potential role of help from extended family members in Asian women’s employment (Kang and Cohen 2018). We thus examine the extent to which the employment status of women with young children is conditioned by the presence of parents or parents-in-law. According to figure 4, the employment-promoting role of an extended family member becomes even more pronounced when Asian women have young children under age five. In comparison with figure 3, which uses the analytic sample of all women, we observe much larger White–Asian employment gaps: the positive differences (that is, Asian groups having higher employment rates than Whites) for those in extended families increase in magnitude, and the negative differences (that is, Asian groups having lower employment rates than Whites) in nuclear families also increase. Specifically, first-generation Asian women in extended family households, except for Koreans and Asian Indian (non-significance difference), are much more likely than White women to be employed when they live with parents or parents-in-law. Conversely, first-generation Asian women with young children have lower employment rates than White women in the absence of coresiding parents or parents-in-law, except for Filipinas between the 1990s and the 2000s. This pattern is more pronounced in recent decades.

Importantly, the positive role of extended families in Asian women’s employment (in comparison with White women) persists for the 2+ generation. These results suggest that coresiding parents or parents-in-law in Asian American families are likely involved in household chores and childcare, especially in families with young children (Kang and Cohen 2018), thus facilitating women’s labor force participation (Greenman 2011). The finding that the positive role of extended families carries over to the 2+ generation further indicates that help from parents or parents-in-law is still an important resource or strategy for labor force attachment of the 2+ generation Asian women with young children. Additionally, our supplementary analyses show that Asian women living with parents with health limitations still have higher employment rates than those in nuclear families, especially the first generation. These results suggest that elderly parents, even with some health limitations, may provide help with childcare and housework, which leads to a greater labor supply of Asian women (Greenman 2011; Kang and Cohen 2018).

Next, figure 5 breaks down parental coresidence into two groups: women living with their own parents and women living with parents-in-law. The degree of familistic-collective value and traditional gender norms, which affect family structure and the patterns of intergenerational support, might not be uniform across subgroups of Asian Americans (Kang and Cohen 2018; Moon 2003). Therefore, whether and how extended households facilitate women’s employment might differ by the type of extended family—whether coresiding with their own parents or parents-in-law (Yu and Cheng 2024; Yu and Xie 2018).

As noted in figure 3, Asian American women in extended households in general have higher employment rates than Whites, especially among the first generation. Figure 5 further reveals nuanced differences in Asian women’s employment by extended family type. Among first-generation Asian women, for example, the positive difference relative to White women is larger when they live with their own parents. However, the White–Asian gap is smaller or nonsignificant for Asian women in extended families with their parents-in-law. Further, Korean women who coreside with parents-in-law stand out in that they show significantly lower employment rates than White women. The unique pattern found among first-generation Korean women is consistent with evidence of the negative association with employment when elderly parents with care needs are present in the household (Kang and Cohen 2018). Additionally, Chinese women living with their own parents show higher employment rates than White women, while there are no advantages from living with parents-in-law, a similar finding documented in mainland China (Yu and Cheng 2024; Yu and Xie 2018). It is also interesting that, for Asian Indian women, who often showed similar employment patterns to Korean women, differences in employment rates by the type of extended family appear to be smaller than those of Korean women, although living with their own parents is more conducive to their labor force attachment.

In contrast, the significance of the type of extended family in employment, particularly the negative role of coresiding parents-in-law found for some groups, is much less pronounced among 2+ generation Asian women (figure 5, right panel). Interestingly, until the 2000s, Filipina and Vietnamese women in extended households with parents-in-law have higher employment rates (relative to Whites) than those living with their own parents. With the caveat that the percentages of 2+ generation women in intergenerational households are small when broken into types of extended family structure, these findings suggest that 2+ generation Asian families may be less influenced by traditional family values, including the relationship between parent-in-law and daughter-in-law. Moreover, because of the positive role that coresiding parents and parents-in-law play in women’s employment, household extension can be a strategy for achieving social mobility through increased labor force participation among Asian American families (Ermisch 1988; Kang and Cohen 2018; Kaushal and Reimers 2007; Moon 2003).

Lastly, figure 6 presents the results for women’s employment by the level of household income, specifically the top 20 percent (panel A) and the bottom 20 percent (panel B). This analysis helps us examine whether the association between household extension and women’s employment varies by household economic conditions (Kulkarni 2019).⁸ Notably, almost all Asian groups in the top 20 percent of household income, regardless of immigrant generation, show higher or similar employment rates compared to White women in the presence of parents or parents-in-law. These findings may suggest the possibility of selection among Asian women in high-income households (such as those with higher education and professional occupations), reflecting compositional characteristics of recent Asian immigrants, which may increase their household income (Lee and Min 2015). At the same time, they may imply the importance of extended family support in women’s employment in high-income Asian households (Kulkarni 2019), consistent with evidence of a greater labor supply of Asian women (compared to White women) in the STEM fields, which thus contributes to the White–Asian earnings growth gap (Grenman 2011).

Interestingly, first-generation Korean women in the top 20 percent income bracket in nuclear families have the lowest employment rates, with a widening gap from other groups over time. First-generation Asian Indian women show similar patterns, but they are more likely to work in recent decades, reaching parity with White women.

Additionally, among low-income households, first-generation Asian women in extended households are more likely than White women to be working. Korean and Asian Indian women, who often show diverging patterns from other Asian groups in labor force attachment, have comparable rates to White women only when in extended families. These findings provide supporting evidence for the structural constraints argument that Asian American families may form extended households due to economic needs or as a mobility strategy (Kulkarni 2019). Supplementary analysis (not shown) reveals that Asian women often assume the role of a primary earner in low-income households, indicating the importance of extended family structure in providing financial support to family members through household extension, which may in turn encourage or require women’s employment (Becker 1965; Tienda and Glass 1985).

CONCLUSIONS AND DISCUSSION

In this study, we examine the role of vertically extended families in Asian American women’s employment with a focus on within-Asian and Asian–White variations by immigrant generations across multiple time periods from 1990 to 2022. Our findings indicate that all Asian groups considered in this study—Chinese, Korean, Filipino, Asian Indian, and Vietnamese—have a greater prevalence of extended families than US-born White women. This pattern still holds for the 2+ generation, although the gap in extended households narrows compared to first-generation Asian American women (figure 1). Further, Asian women living with parents or parents-in-law show higher employment rates than their counterparts in nuclear families (p < 0.05) across all groups among the first generation (figure 2). Differences in employment rates across household types are largest for Korean and Asian Indian women. While the gap in employment rates by family structure declines among the 2+ generation, the extended family structure appears to promote Asian women’s labor force attachment. It is, however, interesting that White women’s employment rates are lower when they live with parents or parents-in-law, probably reflecting differentials in motivations for household extension or the patterns of intergenerational help between White and Asian families (Lofquist et al. 2012).

Our results from a series of logistic regression models demonstrate that these patterns hold true after controlling for various demographic (such as parenthood and marital status) and socioeconomic (including household income) characteristics. The importance of the extended family in Asian women’s employment is particularly salient for the first generation: East Asian (Chinese and Korean) and South Asian (Asian Indian) women in nuclear families, for instance, have much lower employment rates than their White counterparts (figure 3). More importantly, results from the analyses that further consider demographic and economic conditions of the family consistently show the employment-promoting role of extended families for Asian women. For example, the presence of parents or parents-in-law helps Asian women with young child(ren) stay in the labor market at higher rates relative to White women (figure 4). First-generation Asian Indian and Korean women, who have the lowest levels of extended families and employment (figures 1 and 2), have employment rates close to those of Whites if they live in extended family households, even with the presence of young children. This finding suggests that their employment rates would be higher if they were similar to other Asian groups in terms of household extension. Additionally, first-generation Chinese women’s employment depends on whether parents or parents-in-law are in the household when they raise young children. We also find that extended families widen the White–Asian employment gap among the 2+ generation raising young children in a way that it boosts employment rates for most groups (Chinese, Filipina, Vietnamese) or stops some groups from withdrawing from the labor market (Korean and Asian Indian) (figure 4, right panel). As noted, even living with parents with health limitations appears to be conducive to Asian women’s employment. Additionally, this positive association between extended family structure and Asian women’s employment is found for both high- and low-income households. Taken together, these results indicate the important role that extended living arrangements play in supporting Asian women’s continuous employment, which contributes to their labor supply advantage over White women (Greenman 2011).

However, the role of extended families is not uniformly positive in Asian women’s labor market activities, as within-group heterogeneity may exist with regard to family values and gender roles (Drouhot and Garip 2021; Kang and Cohen 2018). For example, certain cultural characteristics, such as collectivism in East Asian societies, may affect the level and process of integration, which in turn affects their family structure and labor market behaviors. As noted, first-generation Korean women are employed at comparable rates to White women when coresiding with their own parents; however, those living with parents-in-law show much lower employment rates (figure 5). At the same time, first-generation Korean women in high-income households have lower employment rates than other Asian groups, and their gap from other Asian groups is persistent in recent decades (figure 6). Prior research suggests that Korean immigrants’ economic segregation in ethnic enclaves and their affiliation with ethnic churches may perpetuate the patriarchal ideologies that stress husbands as primary family breadwinners and mothers as caregivers (Min 2001). Indeed, evidence from Korean women in New York, where there is a high concentration of Korean Americans, shows that full-time employment is influenced by the availability of transgenerational help from elderly female adults (Moon 2003). These diverging patterns found among the first generation largely disappear in the 2+ generation of Korean women, indicating the diminishing influence of traditional culture in family relations and division of labor for later immigration generations. As noted, ethnic variabilities in the patterns of household extension and employment rates, as well as the role of extended families in employment, become smaller in the 2+ generation of Asian women.

In conclusion, we would like to discuss the limitations of the present study and offer some suggestions for future research. First, given the structure of the data, we can only estimate the correlation of household extension with women’s employment measured at the same time. It would be ideal to have longitudinal data to examine the causal relationship between coresidence with parents or parents-in-law and women’s labor market activities by conducting analyses that account for unobserved individual- or family-level characteristics.

Second, we did not consider the potential role of a horizontally extended family structure in women’s employment. While horizontally extended families are less prevalent among Asians compared to other racial minorities (for example, Blacks and Hispanics) (Kamo 2000), there may exist ethnic variability in preference for and the patterns of support in horizontally extended households.

Third, due to data limitations, we cannot fully detect temporary parental help (such as visits from parents or parents-in-law from the country of origin), particularly for the first generation. Therefore, our estimates about the role of extended families in Asian women’s employment may be conservative if they receive help from nonresident parents more than White women (Greenman 2011).

Lastly, while the Census and ACS are among the best data sources that allow analyses for variations by ethnicity, immigrant generation, and time periods, we have to exclude some groups due to small cell size when broken into multiple categories (by household extension and generation). The lack of data that capture the heterogeneity of Asian American subpopulations is identified as the primary reason for a limited understanding of Asians in the US, thus calling for future data collection efforts targeting Asian Americans (Le et al. 2020). Better data will help further understand what factors lead to variations by ethnicity, generation, and time in the associations between household extension and Asian American women’s employment documented in this study.

Despite these limitations, our study is the first to examine the role of an extended family in Asian American women’s employment with a focus on differences across subgroups and immigrant generations spanning from 1990 to the early 2020s. We also document how extended family living arrangements are associated with Asian women’s employment in various family circumstances, including the presence of young children, the type of household extension (own parents versus parents-in-law), and household economic conditions. Combined with prior research (Greenman 2011; Kang and Cohen 2018; Moon 2003), the findings of this study document the employment-promoting role of extended family structure for Asian women, both for the first- and 2+ generations. While the motivations for forming extended households may reflect cultural factors as well as economic considerations (Glick et al. 1997; Kamo 2000), household extension would be a strategy or mechanism for Asian families to achieve mobility across generations (Greenman 2011; Kulkarni 2019). Building on the findings of this study, future research needs to examine detailed mechanisms linking extended family structures to the socioeconomic mobility of Asian families, such as educational attainment, household income, and asset accumulation.