Are married immigrant women secondary workers?

Understanding conflicting evidence

Earlier studies on the assimilation of immigrants focused on men. Interest in the labor market outcomes of married immigrant women came later, framed by the notion that immigrant women are secondary workers. Thus, it was expected that recent immigrant women would join the labor market to support their husbands’ investment in local skills during the initial years of settlement and would withdraw from the labor market as their husbands assimilated. Studies in the early 1990s seemed to support this view for the US [2] and Canada [3], [4], [5]. However, this traditional model of labor decisions among immigrant women has been challenged recently by observations of increasing participation of immigrant women in the labor market and of their rising role as primary workers in some families.

Finding conflicting evidence of the role of married immigrant women in the labor market is not surprising. Immigration policies and institutions for immigrant assimilation vary, and these differences are bound to affect the composition of immigrants and the way immigrant families allocate their time. Further, changes in those institutional features could eventually alter the composition of immigrants and produce conflicting empirical results.

In the US, for example, evidence suggests that immigrant women from earlier arrival cohorts (1960s to early 1980s) followed an earnings path consistent with the theory of wives as secondary workers, but recent cohorts (arriving from 1984 through 1993) do not [6]. These results are consistent with a model in which the cost to the wife of lower investments in host-country-specific human capital (through high initial-wage but low wage-growth jobs) is weighed against the benefits of supporting the returns to investment in the husband’s host-country-specific human capital. In such a model, educated couples have fewer incentives to follow traditional patterns of household work as education raises the costs to a woman’s career of following such a strategy and lowers the benefits to the husband’s career.

These result seem to be supported by evidence of two very different patterns of work participation for immigrant women working “dead-end” jobs and those working career jobs. Further evidence of this possibility is provided by examining the work trajectories of these two groups of women [7]. Immigrant women working dead-end jobs behave according to traditional patterns of female participation and are more likely to leave the labor force over time than are native-born women. Immigrant women working career jobs, however, show work trajectories similar to those of comparable native-born women [8].

The most recent research on the labor force behavior of immigrant women in Canada [1] also indicates that later immigrant cohorts, arriving in the 1990s and early 2000s, do not behave as has been reported for cohorts who arrived in the 1970s and 1980s [5]. These differences are likely linked to the changing educational composition of Canadian immigrants. In the 1990s, Canada adjusted the point system used to select immigrants, reducing the weight of the “intended occupation” category and increasing the weight of education and other human capital. This model of selection focuses on characteristics believed to be important determinants of positive longer-term economic outcomes. The result was a rapid rise in the educational attainment of immigrants through the 1990s. Recent immigrant women in Canada are typically better educated than native-born Canadian women and might be driven by stronger attachment to the labor force. These changes could explain differences in labor force activities between recent and former cohorts of immigrant women.

Other studies also find that certain groups of immigrant women depart from traditional patterns. In Australia, for instance, married immigrant women who are listed as the principal applicant for immigration do not conform to old-style patterns of work, although other groups of immigrant women do. This divergence is typically linked to higher levels of human capital observed for married immigrant women who are listed as the principal applicant [9].

Labor force participation of married immigrant women

Analyzing immigration outcomes is not a simple matter. Measuring how labor market outcomes evolve requires observing the same individuals over time. However, longitudinal data of this kind (panel data) are typically unavailable. Instead, most research uses artificial longitudinal information that follows groups of individuals, rather than individuals themselves, over time, using several cross-section surveys such as the Canadian Census of Population to analyze the average characteristics of groups of entering immigrants.

A further problem in comparing immigrant women’s labor market outcomes internationally has to do with institutional differences across countries in how women’s labor force participation is encouraged and rewarded by the market (or discouraged and penalized). However, outcomes can be compared between immigrant and native-born women with similar personal characteristics, since they share the same institutional framework of employment incentives for women, gender wage differentials, family friendly benefits, and the like.

Three dimensions of women’s labor market performance are of interest here: labor force participation, wages, and occupation. The first dimension highlights differences in the resources devoted to the labor market, the second focuses on differences in how the market rewards these activities, and the third captures whether nonmonetary aspects of the type of jobs performed by immigrant and native-born women are similar.

In the US, labor supply assimilation patterns are similar for immigrant husbands and wives, and their labor force participation is lower than that of comparable native-born men and women [8]. For immigrants, work hours increase with time in the country, and for immigrants who arrived in the 1980s and 1990s, work hours eventually converge toward those of comparable native-born workers.

A similar pattern is reported for Germany, with immigrant husbands and wives both working less upon arrival than comparable native-born workers but then increasing their labor supply in successive years [10]. This pattern suggests a standard process of human capital accumulation.

For the UK, labor force participation of women in most of the immigrant groups examined also converges toward that of native-born women, but not all immigrant groups achieve full convergence [11].

Research for Canada also suggests that recent married immigrant women do not fit the profile of secondary workers, but rather, as observed in the US and Germany, labor market participation and wages rise as they stay longer in the country. Their behavior displays a path of assimilation similar to that of immigrant men. At arrival, they participate in the labor market at a much lower rate than comparable native women, but this rate increases steadily over time. The average immigrant woman arriving in Canada during the first half of the 1990s is only 34% as likely during her first five years in the country to enter the labor market as her native-born counterpart, but 72% as likely after being in Canada for 15–20 years [1]. This is in contrast with what was observed for earlier cohorts of immigrant women [5].

One plausible explanation for the difference between these findings and those of previous Canadian research is that the newest cohorts of immigrant women are more committed to the labor market than were previous cohorts. Appearing to corroborate this interpretation is the finding that immigrant women increase their participation in low-skill jobs over time relative to comparable native-born women. This rising pattern suggests that immigrant women either remain in or continue to enter low-skill jobs at a much higher rate than native-born women do [1]. If immigrant women were seeking only temporary attachment to the labor market to support family settlement in Canada, they would be more likely to enter (and leave) low-skill jobs without pursuing career advancement.

The wages of married immigrant women

Under the traditional framework discussed earlier, no significant wage advancement is expected for married immigrant women because they intend their positions to be temporary. Again, the results for recent cohorts of immigrant women do not support this view. Rather, their wages, like their labor force participation, increase steadily with years in the country. In the US, estimated wages of immigrant women reach parity with native-born women some 10–15 years after arrival [8].

Wage assimilation is weaker in Canada and Germany. In Canada, there is evidence of positive wage assimilation for immigrant women, but their wages never reach parity with native-born women [1]. Their wages rise more than 30% during the first ten years but then remain 15–30% below the wages of comparable native-born women. Similarly, in Germany, the wages of immigrant women fall short of those of comparable native-born women, even 20 years after migration [10].

In the UK, immigrant women of color remain substantially underpaid 20 years after arrival relative to native-born women, and there is little evidence of wage 
assimilation [11].

Skill mobility of married immigrant women

Unlike labor market participation and wages, other characteristics of immigrant women, such as occupation, have been much less studied. The reason is that occupation might not be the right dimension to look at in studying immigrant workers’ labor market outcomes, since definitions of occupational categories are either too broad to be useful or too narrow to be used effectively in an empirical model. Recently, however, new data sets such as the Dictionary of Occupational Titles (DOT) and the Occupational Information Network (O*NET), have made it possible to analyze the skill content of jobs, which is important for understanding women’s assimilation patterns. Little is known about the skill progression of immigrant women in countries other than the US and Canada.

US studies show that immigrant women are more likely to work dead-end jobs than are comparable native-born women. Immigrant women who start working in a dead-end job are also more likely to leave the labor force in the medium to long term [7].

In Canada, studies show that immigrant women generally work in jobs requiring more physical strength skills and less analytical skills, a sign of lower status jobs. For both strength and analytical skills, the patterns of skill mobility are surprisingly flat on average (Figure 1). Another surprising result is the relative absence of cohort effects in analytical skill progression, with cohorts starting at similar levels of skills over time. By contrast, estimates show strong cohort effects for the evolution of strength skills, which have been increasing for recent cohorts of immigrant women. This result suggests that the average immigrant woman experiences little career progression as measured by the skills embodied in the jobs she takes. This could be interpreted as a result of low attachment to the labor force, following the patterns of secondary workers. However, the results are puzzling when viewed together with the patterns 
of wage mobility and increasing participation in the labor market over time noted above [1].

The labor market role of highly educated married immigrant women

In most countries, the differences in labor market behavior of successive cohorts of immigrant women—some conforming to traditional models of participation and some not—seem linked to their human capital investment, which defines the relative rewards to their labor market activities.

The case of Canada is particularly interesting because of its large population of educated immigrants. There is strong evidence that the participation of highly educated immigrant women in low-skill jobs is significantly higher than that of comparable native-born women. Among university-educated women, the odds of working in jobs that require low levels of analytical skills are between two and four times higher for immigrant women than for native-born women, while the odds for jobs requiring high levels of strength skills are between 1.5 and two times higher.

Further, participation in jobs requiring low levels of analytical skills rises over time, suggesting that highly educated immigrant women have more difficulty entering medium- to high-skill positions than do native-born women. While it is also tempting to conclude that these immigrants experience difficulties with job mobility, this finding cannot simply be interpreted as portraying highly educated immigrant women as trapped in relatively low-skill positions.

It could instead be the case that some highly educated immigrant women start in low-skill jobs and move up to higher skill positions over time but are replaced (in the aggregate) by immigrant women of the same arrival cohort who had not been working before then. While cross-sectional data do not allow an examination of this scenario, the fact that the share of university-educated women in highly skilled occupations increases over time for most immigrant cohorts lends support to this suggestion.

In the first five years after migration, educated women enter low-skill positions at a much faster rate than they enter high-skill positions. After one year in Canada, the odds that a university-educated immigrant woman is in a low-skill job relative to a comparable native-born woman rise from around 1.3 at arrival to two, whereas the odds of holding a high-skill job remain stable at around 0.6 and only slowly increase to 0.8 after five years in Canada [12].

The wage gap at arrival between immigrant and native-born women is larger among university-educated women than among non-university educated women; around 40–50% for recent cohorts. However, the rate of wage progression is faster for university-educated women, so that wage differences narrow to 15% after 15 years in Canada (see the Illustration). In contrast, among the non-university educated women, the initial wage gap is smaller, at around 20%, and disappears within 15 years.

A similar dichotomy appears in US studies, with women of recent immigrant cohorts also exhibiting low initial earnings with high wage growth relative to earlier cohorts (who exhibit high initial earnings and low wage growth) [6]. One of the principal characteristics that distinguishes recent cohorts from earlier ones is the higher educational attainment of immigrant husbands and their spouses.

For skill progression, the picture for educated immigrants shows a more dramatic trend. The analytical skills required in jobs held by educated immigrant women rise over time, while the strength skills required in these jobs diminish considerably. Educated immigrant women from the 1996 cohort in Canada, for instance, start in jobs requiring levels of strength skills that are 0.3 standard deviations higher than those required for jobs of the average native-born woman (Figure 2). After five years in Canada, these women hold jobs that require strength levels that are only 0.15 standard deviations above the average. This is a remarkable change considering the flat profiles observed for the average immigrant woman (see Figure 1). It suggests that it is mainly immigrants with no post-secondary education who remain in low-status jobs, as measured by the levels of skills jobs require. This finding is mirrored by US studies [7].