The “economic assimilation” of immigrants–or the capacity of the host economy to “absorb” new immigrants–has been the major focus of recent research on the economics of immigration. In Canada this has spawned extensive literature that compares the labour market behaviour and age-earnings profiles of immigrants relative to their native-born counterparts. While the empirical results vary depending upon the “cohort” group based on year of arrival, three clear trends emerge.

First, immigrants tend to have a higher rate of participation in the paid labour force and tend to work longer hours. This reflects a smaller accumulation of savings or wealth, or a relatively late entry into the labour market, often due to interruptions in their working lives caused by economic and political events. The relatively high rate of labour force participation is especially the case among women immigrants given the greater incidence of two-income families among immigrants than non-immigrants.[6]

Second, after adjusting for all measurable productivity characteristics (such as education, work experience, language skills), immigrant workers face an “entry penalty” in terms of a lower initial wage relative to their native-born counterparts. Over time however, the earnings of immigrants gradually “catch up” with those of native-born workers with equivalent observable skills. Interpretations of this “economic assimilation” vary. On the one hand, it may reflect the acquisition of “unobservable skills” (such as greater English or French language proficiency, or greater familiarity with Canadian cultural norms); on the other hand, it may reflect the gradual decline in discrimination faced by new immigrants, particularly visible minorities, as they acculturate.

Third, the rate of economic assimilation-indeed whether or not immigrants’ incomes ever “catch up” to those of native-born Canadians during their working lives-has varied significantly among different cohort groups depending upon their year of arrival. Evidence from the 1981 and 1991 Census indicated that more recent immigrant cohorts experienced a higher initial entry penalty and a much slower rate of assimilation such that the prospect of ever catching up with the earnings of their native-born counterparts was unlikely.[7] More recently, an exhaustive examination of the 1991 Census suggests a break in the pattern. While the 1981-85 cohort had lower initial earnings than the previous cohort, the following group arriving in 1986-90 had earnings that were approximately equal to those of the 1981-85 cohort, pointing to the possibility that the decline in entry earnings of successive cohorts has stopped. As well, the assimilation rate of the 1981-85 cohort was found to be 17.2 percent over the first five years in Canada, compared to a 0 percent assimilation rate over the first five years for those immigrants arriving between 1976-80.[8]

A separate question, unaddressed in the literature, concerns the post-retirement incomes of immigrants. The simple life-cycle model implies that individuals will save during their working lives in order to accumulate sufficient wealth for their retirement years; during retirement their assets gradually diminish as they dissave. In effect, individuals seek to “smooth out” their lifetime consumption when their annual income varies significantly over time. For the elderly, future income is known with reasonable certainty (notwithstanding variations in the rates of inflation and interest), but one=s longevity may not be. With perfect foresight regarding longevity, wealth should fall to zero at the time of death; under the more realistic assumption of uncertain longevity, wealth will gradually diminish as the retiree ages. Alternatively, if leaving a bequest is an important goal, the rate of dissaving may be less acute.

Empirical investigation of life-cycle models finds that the level of education, gender, marital status, ethnicity and cohort effects play a major role in explaining the heterogeneity in the saving behaviour of American households.[9] For the elderly with children, the “bequest motive” is also found to be a significant determinant of dissaving. The overriding result however, is that many individuals or households do not accumulate sufficient wealth for their retirement years, which results in a greater dependence on government transfers, a high incidence of low income, and a gradual decline in total income with age.[10]

As insightful as this literature is, it has not considered the importance of immigration status on the age-earnings and age-wealth profiles. Even if the rate of economic assimilation is sufficiently high such that immigrants= earnings indeed catch up to those of non-immigrants, it is unclear whether their life-long employment earnings provide sufficient savings for their retirement years. This argument is stylized in Figure 1. Assume that an immigrant enters the country in year “a” at the beginning of his or her normal working life. After adjusting for all “observable” differences in productivity (education level, work experience and language skills), the immigrant faces an entry penalty (e) in terms of a lower initial wage, but enjoys a faster increase in annual income with greater experience in the Canadian labour market. This more rapid rate of increase in income is attributed to the economic assimilation that occurs. The immigrant’s annual income then catches up to the non-immigrants (with the same observable skills) in year “c”. If however, year “c” occurs sufficiently late in the immigrant’s working life, his or her lifetime earnings may be lower and, assuming similar rates of savings, the immigrant enjoys a lower income during retirement.

It follows that if an immigrant’s entry into the Canadian labour market occurs at a relatively late age, such that the age-earnings profile is attenuated, the likelihood of his/her income catching up to that of the native-born counterpart is reduced. The implications for incomes beyond the normal age of retirement are apparent. Assuming a similar rate of savings out of current income, accumulated saving at the normal age of retirement will be less and, accordingly, income will be lower.

Figure 1: Hypothetical Income Paths

Immigrant and Native-Born Canadians