Publications
Culture, Tourism and the Centre for Education Statistics
Location of Study and the Labour Market Success of Immigrants to Canada
- Main page
- Acronyms
- Definition of concepts used in this report
- Introduction
- Literature review
- Where does this study stand?
- Data
- Descriptive analysis
- Empirical strategies
- Multivariate analysis
- Sensitivity analysis
- Discussion
- Concluding remarks
- Tables and charts
- Appendix
- References
- More information
- PDF version
Chapter 7: Multivariate analysis
Archived Content
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
To start, it is important to note that this chapter focuses on the effect of location of postsecondary education completion on immigrants' employment status and earnings, compared to the Canadian-born, after controlling for the effects of each of the independent variables. In doing so, we do not imply that the effects of background variables on labour market outcomes of internationally-educated immigrants compared to the Canadian-born is not policy relevant or intriguing for research. Much of our motivation for this analytical strategy comes from two factors. First, the payoff to other personal and non-personal characteristics is not central to our report. Second, it has been extensively covered in prior Canadian empirical research (see for instance, Ferrer and Riddell, 2006; Hansen, 2006; Alboim, Finnie and Meng, 2005; Sweetman and McBride, 2004). Therefore, unless otherwise stated, the discussion of the analytical results in this chapter is purposively limited to the outcomes related to location of postsecondary study. That being said, Appendices 11.1 and 11.3 present an analysis of the determinants of labour market behaviour and earnings in Canada, other than location of postsecondary study and immigration status, whereas Appendix 11.4 analyzes selected multivariate results by individual location of highest postsecondary study.
Prior research suggests that immigrants' labour market outcomes improve with duration of residence as they become increasingly familiar with Canadian employment opportunities and accumulate other characteristics relevant to Canadian jobs (see for instance, Ferrer and Riddell, 2006; Picot, 2008). So as to provide a comparative analysis, the findings are thus broken down by immigration cohort, distinguishing between very-recent, recent and established immigrants. Finally, as reported in Appendix Tables A.4 and A.6, we see that the size of the coefficients of location of study fall significantly when we take into account other observables such as gender, family structure, field of study, language spoken, etc. This reinforces the notion that by ignoring differences in personal and non-personal factors, descriptive analysis alone produces overestimated effects of postsecondary location of study on relative labour market outcomes of landed immigrants.
7.1 Results: Labour force status
Chart 1 shows the marginal effects of immigrants' location of postsecondary study on their relative labour force status, in comparison with the Canadian-born. We choose to analyse marginal effects instead of the coefficients because they are much easier to interpret: they reflect the magnitude of the effect a given location of postsecondary study has on an immigrant' labour force status as compared to the Canadian-born.1 Charts 1.A to 1.F present an overview of, in sequence, persons not in the labour force (Chart 1.A), self-employed workers (Chart 1.B), undereducated employees (Chart 1.C), correctly-matched employees (Chart 1.D), overeducated employees (Chart 1.E) and persons attending school full-time or part-time (Chart 1.F). In each graph, the 0 value on the X-axis represents our reference demographic group, i.e., the Canadian-born. A positive (negative) marginal effect of a given location of postsecondary study indicates an increasing (decreasing) effect of that location of study on the likelihood that an average immigrant from a given cohort will have a specific labour force status relative to the average Canadian-born, after controlling for other factors that influence employment status.
Chart 1.A Predicted relative impact of location of highest postsecondary study on relative labour force status of immigrants— Marginal effectsA. Very-recent immigrants
Compared with being unemployed, almost all very-recent immigrants were less likely to find paid employment and more likely to be out of the labour force than the Canadian-born. Particularly, very-recent immigrants were less likely than the Canadian-born to be undereducated or correctly-matched employees, everything else held equal. These results reinforce previous findings by Aydemir and Skuterud (2004) and Frenette and Morissette (2003) that Canadian newcomers may face a comparative disadvantage in the allocation process of paid jobs, suggesting that pre-migration skills are heavily discounted in Canada's labour markets. Only very-recent immigrants who completed their postsecondary education in India (+8%) or the Philippines (+18%) were more likely than the Canadian-born to be classified as overeducated employees. Likewise, self-employment was more likely to be important among very-recent immigrants who completed their postsecondary studies in South Korea, with a predicted average probability that was 9% higher than that of the Canadian-born.
The current acceleration of technological changes in most advanced economies of the world may lead most Canadian firms to constantly modify conditions under which they utilize capital and labour in the production process of commodities so as to remain competitive on national and international markets. Hence, prime age (25 to 64) individuals with a postsecondary education who have no attachment to the labour market might face a higher risk of downgrading their human capital. This in turn could result in a permanent confinement at the bottom of the earnings distribution and economic vulnerability in the long run. Chart 1.A shows that, although very-recent immigrants were more likely than the Canadian-born to be out of the labour force, those who completed their postsecondary education in Pakistan had the highest predicted likelihood of being out of the labour force (+27%), closely followed by those who graduated in South Korea (+22%).
Chiswick (1978) suggests that newly-arrived immigrants may be expected to have a weaker attachment to the host country's labour markets because the initial years of settlement in a new country are often marked by greater investment in skills relevant to local jobs. If differences in the predicted likelihood of being out of the labour force between very-recent immigrants and the Canadian-born are due to differences in initial investment in Canada-specific skills, then in accordance with this hypothesis, we would expect no huge gap within the same immigration cohort. Interestingly, we note that very-recent immigrants who obtained their highest postsecondary degree in Pakistan are: over ten times more likely to be out of the labour force than those who completed their postsecondary education in the Philippines; over four times more likely than those who graduated in India, Romania or Russia; and over twice as likely as those who finished their postsecondary studies in France or Poland.
It seems reasonable, therefore, to consider the following explanation proposed by Baker and Benjamin (1997): compared with the average Canadian-born, the typical newcomer has to simultaneously acquire skills specific to the host country and begin earning a living. To accomplish that, immigrant couples may adopt a strategy of labour specialization. One spouse, generally the wife, will work full-time to support the household or will produce home-made commodities. Meanwhile, the other spouse, generally the husband, will focus mainly on acquiring qualifications specific to the host country through education, apprenticeship or on-the-job training. Indeed, data not reported here show that as of 2006, more than half (52%) of the highly-educated population between 25 and 64 who were not in the labour force were women. This proportion is slightly higher for immigrants (56%) than for the Canadian-born (50%). Moreover, 30% of the Canadian-born who are not in the labour force are married women, compared to 40% of immigrants not in the labour force.
Compared to the unemployed, very-recent immigrants who completed their postsecondary education in Canada were, on average, 13% less likely to be overeducated and 8% more likely to be out of the labour market relative to the Canadian-born. This result is worth discussing because descriptive statistics not reported here indicate that over half (58%) of very-recent immigrants with a Canadian postsecondary degree aged 25 to 64 who were not school attendees in 2006 had graduated in fields of study identified by authors such as Sweetman and McBride (2004) as high-paying fields of study in the Canadian economy, namely: business / administration (26%); engineering (19%); and health / welfare (13%). Further, Chiswick (1978) argues that it is highly beneficial for almost all internationally-educated immigrants to invest in host-country educational capital because they will always face a comparative disadvantage in the allocation process of well-paying jobs as a result of "country-specific aspects of the knowledge acquired in school, lower quality of foreign credentials, or the poorer information they provide employers who use schooling as a screen." One implication of our findings therefore is that acquiring a postsecondary credential in Canada may assist very-recent immigrants in achieving positive labour market outcomes, but it does not necessarily guarantee positive outcomes for all who do so.
Our estimates indicate that very-recent immigrants who completed their highest postsecondary education in Canada, China, Romania, the Russian Federation or Iran were more likely than the Canadian-born to be attending school, either full-time or part-time, supporting the skills-investment hypothesis. However, very-recent immigrants who finished their postsecondary education in Pakistan and South Korea were, respectively, 4% and 3% less likely to be attending school than their Canadian-born counterparts. This last finding is especially surprising since immigrant members of those two groups were also more likely to be out of the labour force and had on average, the smallest chances of securing paid employment quickly. This result could be at least in part explained by differences in opportunity costs of noneconomic activities: the cost of spending disposable time in nonmarket activities including the accumulation of Canada-specific skills may be so high for some very-recent immigrants with a postsecondary education who completed their highest level of postsecondary education in South Korea or Pakistan that they chose to work for pay with foreign credentials despite having lower initial employment perspectives.
B. Recent immigrants
The effect of location of study is similar for very-recent and recent immigrant groups, though different in magnitude (see Charts 1.A to 1.F). Among recent immigrants with a postsecondary education, those who completed their education in South Korea were more likely to be self-employed (+14%) and to be out of the labour market (+17%) and less likely to be overeducated paid employees (-20%) compared to the Canadian-born. Likewise, among recent immigrants, immigrants from the Philippines and India tended to be overeducated on Canada's labour markets: relative to the Canadian-born, the incidence of over-education was 18% higher for recent immigrants with a postsecondary education completed in the Philippines and 8% higher for those who completed their studies in India. However, recent immigrants who completed their highest level of postsecondary education outside Canada were less likely than the Canadian-born to be attending school, everything else held equal.
Some interesting patterns emerge when we compare estimates for prime-aged very-recent and recent immigrants with a postsecondary education. For example, compared to the Canadian-born with comparable ages and education levels, very-recent immigrants who completed their postsecondary studies in France were 5% less likely to be overeducated, whereas their recent-immigrants counterparts were 3% more likely to be overeducated, everything else held constant. For immigrants who completed their postsecondary education in China, duration of residence made a slight difference in the likelihood of being self-employed: relative to the Canadian-born, very-recent and recent immigrants were 4% less likely and 2% more likely, respectively, to be self-employed. Assuming homogeneous human capital and time effects, the implication of these results is that after 5 to 10 years of permanent residence, the postsecondary education of many immigrants may still be less applicable to the Canadian economy.
Borjas (1985) argues for the U.S. that significant changes in national origin of the immigrant flow in the last decades have created cohort-based differences in the structure and the quality of American immigration over time. This in turn has produced intra-group variations in economic success of foreign-born individuals in the U.S, regardless of origin of individual human capital (i.e., education and/or experience). Interestingly, in Canada between 2001 and 2006, almost 60% percent of very-recent immigrants came from Asia and the Middle East, and the proportion of recent immigrants from the Caribbean, Central and South America and Africa increased over 10 percent (Corak, 2008). Following Borjas (1985), differences in employment patterns of very-recent and recent immigrants educated in the same country could be interpreted as reflecting cohort effects rather than immigrant adjustment effects. To cite Lalonde and Topel (1992), it may be that cohort-unobserved attributes (such as talent or immigrant quality) that "affect productivity are fixed within an arrival cohort."
C. Established immigrants
Charts 1.A to 1.F also present the marginal effects of location of postsecondary study for those individuals who have been landed immigrants to Canada for more than a decade. It is commonly argued that a longer duration of residence will have a positive effect on the employment prospects of most newcomers as they acquire skills and accumulate social capital over time. However, despite a lengthy period of permanent residence, the general pattern of established immigrants' marginal effects of location of study on labour force status remains quite similar to the marginal effects for very-recent and recent immigrants, though somewhat less pronounced.
Compared to being unemployed, most established immigrants were less likely to have paid employment and were more likely to be out of the labour force than the Canadian-born, everything else held equal. Similarly, our estimates indicate that established immigrants who completed their postsecondary education in Pakistan or South Korea were the most likely to be out of the labour force and the least likely to be overeducated employees, compared to the Canadian-born. As for other immigration cohorts, we also note that established immigrants who completed their postsecondary studies in the Philippines or India were more likely than the Canadian-born to be overqualified. Similarly, Nekby (2002) finds that after 20 years of permanent residence in Sweden, immigrants still have employment ratios that are significantly lower than those for the Swedish-born. Likewise, Chiswick, Cohen and Zach (1997) show that after ten years of residence in the United States, employment perspectives of the American-born are better than those for many immigrants, and far better than those for Asian immigrants. These findings suggest that the comparative disadvantage faced by established immigrants in the allocation of paid jobs is not unique to the Canadian economy.
According to the literature (see for instance, Ferrer and Riddell, 2008; Picot, 2008; Frenette and Morissette, 2003), established immigrants are expected to have employment ratios closer to those for the Canadian-born and higher than those for very-recent and recent immigrants because the latter two groups are generally neither familiar with employment opportunities in Canada nor do they have sufficient Canadian work history. This hypothesis is known in the empirical immigration literature as the new-labour-market entry effect. As we noted here, however, there is a gap in terms of employment in the wage sector between the Canadian-born and many established immigrant groups. Arguably, therefore, our findings are not exclusively explained by the so-called new-labour market entry effect. It may also be that many Canadian employers and regulatory bodies using education as a screening device may be less familiar with the implications of a foreign education for a worker's productivity (Chiswick, 1978). There is also the possibility that regardless of duration of residence, some foreign-educated immigrants may face other difficulties in Canada's labour markets, including a lack of mobility or good labour market information, economic discrimination and barriers to many professional occupations, especially in the regulated occupations.
Summing up, this chapter corroborates the descriptive evidence by showing that the Canadian-born had better employment outcomes than almost all internationally-educated immigrants, and especially very-recent immigrants in 2006. But, immigrants' relative disadvantage in terms of paid employment was sensitive to location of study: immigrants who completed their postsecondary education in Western Europe (France, Germany and the United Kingdom) or in North America (Canada, the United States) were more likely than other internationally-educated immigrants, especially those who received their highest degree in Pakistan or South Korea, to secure paid employment in Canada.
7.2 Results: Employment earnings
Before turning to the discussion of the regression results, it is worth clarifying the latter. First, separate regressions are performed for each of: undereducated, correctly-matched and overeducated paid employees, as well as for all paid employees. Then, we tested the equality of the regression coefficients for the three categories of paid employment, i.e., undereducated, correctly-matched and overeducated. The results of the Chow test (not reported here) suggest that there are differences in the earnings formation process between the various categories of employees. Second, as shown in Tables A.5 and A.6 in the Appendix, selection terms are all statistically significant at the 1% level, suggesting that education-job mismatch was not a random process in the Canadian wage sector.
Second, the selection term is negative for undereducated and correctly-matched employees. The implication of this result is that a number of unobserved characteristics (such as lack of credential recognition, discriminatory hiring/firing practices, tougher labour regulation, etc.) in those categories were negatively associated with the characteristics that affected individual economic rent, i.e., the salary for which prospective labour market participants were prepared to work. Second, the selection term is positive for overeducated employees and for all paid employees combined. For these groups, that likely points to a positive correlation between factors omitted from the model and factors that influence employment earnings.
Third, the magnitude of the estimated coefficients varies significantly between the two types of earnings model, particularly for the undereducated and correctly-matched groups. Take undereducated paid employees for instance. Estimates unadjusted for potential selection bias indicate that very-recent immigrants with a postsecondary education aged 25 to 64 who completed their postsecondary education in Poland or Germany were the only ones whose average earnings were lower than those for their Canadian-born counterparts, which is not the case when adjusted estimates are considered. Differences are also observed in terms of the significance of some effects of the selected explanatory variables (see Table A.5 in the Appendix). Fourth, the results of an examination of background characteristics affecting employment earnings are shown in Appendix 11.3. Last but not least, a country-level analysis of the effect of the origin of highest postsecondary education on prime age (25 to 64) immigrants' employment earnings is provided in Appendix 11.4.
Having said all this, unless otherwise stated, we will concentrate our analysis on estimates obtained after accounting for selectivity. For this purpose, Charts 2.A to 2.D presents, in sequence, the estimated relationships between relative employment earnings and location of study of landed immigrants who were undereducated (Chart 2.A), correctly matched (Chart 2.B) and overeducated (Chart 2.C) in 2006, as derived from Table A.6 in the Appendix. Finally, the relative earnings associated with location of study for immigrants who worked in the wage sector is presented in Chart 2.D.
A. Very-recent immigrants
Looking at the estimates for all wage-sector workers (Chart 2.D), we see that very-recent immigrants were likely to experience a significant earnings gap relative to the Canadian-born, ranging between -25% and -63%, even after taking into account several background factors. This result implies at least in part that newcomers' credentials (including education) may be valued differently than those for the Canadian-born.
Borjas (1992) argues that skills accumulated in economically developed countries have the highest transferability in most countries accepting international migrants for permanent residence because of similarities in industrial structures and personal characteristics used to sort prospective labour market participants by productivity level and to pay them accordingly. If variations in earnings gap between very-recent immigrants from different locations of postsecondary study and the Canadian-born are due to variations in international skills transferability, then, in accordance with the immigrant-skills less-than-perfect international transferability theory, we would expect the earnings gap to be higher for education completed in less-developed countries and smaller (or positive) for education completed in countries that are economically similar to Canada.
Chart 2.D indicates that this is the case: the relative earnings effect of location of study is not homogeneous across all very-recent immigrants. Very-recent immigrants with the largest earnings gap relative to the Canadian-born were those who completed their postsecondary studies in Pakistan, the Russian Federation, South Korea, China or Iran. In particular, those with a postsecondary credential obtained in Pakistan or the Russian Federation earned, on average, 62% less than the Canadian-born, holding everything else constant. Conversely, very-recent immigrants who graduated in the United Kingdom, the United States, France, Germany or Canada had the smallest wage gap compared to the Canadian-born. It is interesting to note that the smallest earnings gap was found for very-recent immigrants who finished their postsecondary education in the United Kingdom (-25%) or the United States (-30%).
Bratsberg and Terrell (2002) who show for the U.S. that the private rates of return to foreign education are significantly higher for immigrants from English-speaking countries than for immigrants from non-English-speaking countries conclude that advanced educational credentials from English-speaking countries are highly transferable in the U.S labour markets. We tend to follow Bratsberg and Terrell (2002) in arguing that everything else held equal, postsecondary certificates, diplomas or degrees obtained in developed English-speaking economies such as the United Kingdom and the United States were the most easily adaptable to Canada's labour markets in 2006.
The analysis of variations in the earnings gaps between very-recent immigrants from different locations of study and the Canadian-born by education-job match level corroborates most findings documented for all paid employees. Concerning overeducated employees, Chart 2.C shows for instance that the earnings gap experienced by very-recent immigrants with a postsecondary education who completed their highest postsecondary studies in Pakistan, South Korea or Russia was twice as high as for those who completed their highest postsecondary education in the United Kingdom, the United States or in the Philippines, compared to Canadian-born.
The case of very-recent immigrants with a postsecondary education completed in the Philippines is worth noting because differences in international skills transferability are commonly associated with socio-cultural and economic differences between source and host countries (see, for instance, Bratsberg and Terrell, 2002; Nekby, 2002; Borjas, 1992 ; Chiswick; 1978). As shown in Chart 2.C, very-recent immigrants with postsecondary credentials completed in the Philippines earned, on average, 33% less than the Canadian-born, whereas those who completed their postsecondary education in France or Germany earned, on average, 40% less, and those who completed their postsecondary studies in Canada earned 36% less, on average.
Further, correctly-matched very-recent immigrants who completed their highest postsecondary studies in the Philippines earned 12% more on average than their Canadian-born counterparts, everything else held equal (Chart B). This suggests two intuitive interpretations. First, there might be no earnings advantage for an average very-recent immigrant with a postsecondary education who completed his / her highest postsecondary education in the Philippines to acquire additional postsecondary education in Canada. Second, it might possible for prime-aged immigrants with a postsecondary education completed in the Philippines to improve their post-migration earnings simply through accumulation of Canadian skills other than education (such as Canada-specific experience / credentials, business / work culture, on-the-job training, etc.), or better selection of paid jobs.
Finally, Chart 2.A shows that there was no earnings gap compared to the Canadian-born for very-recent immigrants classified as being undereducated employees who were postsecondary graduates from Romania, the special administrative region of Hong Kong, the Russian Federation, Germany, Iran, the United Kingdom, Poland, Pakistan and the Philippines. The same is also true for very-recent immigrants educated in the United Kingdom, India, Pakistan, France, South Korea, Canada or Germany who worked in the Canadian wage sector as a correctly-matched employed.
As shown in Table 5, compared to the Canadian-born at 21%, relatively high proportions of very-recent immigrants who completed their postsecondary education in Romania, (42%), Russia (37%), Germany (31%), or Iran (34%) had graduated in engineering fields of study. Similarly, compared to the Canadian-born at 22%, relatively high proportions of very-recent immigrants who completed their postsecondary studies in the special administrative region of Hong Kong (43%), Poland (25%), Canada (27%) or France (26%) were graduates from business or administration fields of study. According to Sweetman and McBride (2004) engineering and business / administration fields of study belong to the subset of high-paying fields of study.
Everything else held equal, then, it may be that for many prime-aged very-recent immigrants with a postsecondary education completed in the Russian Federation, Germany, Iran, the United Kingdom, Poland, Pakistan or the Philippines and who worked as undereducated employees in Canada, the greater earnings derived from their professional occupation fully compensated for the wage loss associated with the lack of Canada-specific skills (such as Canadian education and labour market experience). The same explanation may also hold for very-recent immigrants with postsecondary credentials from India, France and the United Kingdom who worked as undereducated employees since they outnumbered the Canadian-born in the subgroup of graduates from business/administration fields of study (Table 5).
B. Recent immigrants
The impact of location of study on employment earnings in the wage sector taken as a whole is similar between very-recent and recent immigrants, though smaller in magnitude for the latter. As noted above, for instance, the predicted wage gap for recent immigrants (compared to the Canadian-born) is the largest among those who completed their postsecondary studies in Pakistan, South Korea or the Russian Federation. The smallest earnings gap compared to the Canadian-born was found for those who completed their postsecondary education in the United Kingdom or France. The findings are similar for employees classified as being overeducated (Chart 2.C). Everything else held equal, these results suggest that regardless of the duration of residence, the average prospective employer may not understand very well, and so may not be in a position to evaluate, the properties of postsecondary credentials obtained in Pakistan, the Russian Federation or South Korea for the productivity of prospective labour market participants.
Concerning correctly-matched employees (Chart 2.B), recent immigrants with postsecondary credentials earned in the Philippines, India, Germany or Iran enjoyed a significant earnings premium relative to the Canadian-born with similar observed characteristics. Data not reported here indicate that among people classified as correctly-matched employees, the proportion of recent immigrants with a postsecondary education who had studied "hard sciences" in the Philippines (58%), India (62%), Germany (80%) or Iran (77%) was remarkably higher than that for the Canadian-born, at 46%. Also, prior studies (see, for instance Sweetman and McBride, 2004) have linked "hard sciences" to higher earnings on Canada's labour markets. One implication of this result could be that recent immigrants with a postsecondary education who completed their studies in the Philippines, India, Germany or Iran outnumbered their Canadian-born counterparts in the group of people who were correctly matched in highly-paid occupations.
The earnings premium related to recent immigrants with a postsecondary education completed in Germany, the Philippines, and India was 30%, 37% and 18%, respectively, among correctly-matched employees compared to their Canadian-born counterparts. Regarding employees classified as being undereducated, our results indicate that postsecondary education from the Philippines was associated with positive employment earnings: relative to their Canadian-born counterparts, recent immigrants educated in the Philippines enjoyed on average, a 25% wage advantage, after controlling for other variables.2
The finding of higher wage premiums for recent immigrants who completed their highest level of postsecondary education in the Philippines is striking, even surprising, given that the existence of a significant earnings gap between the Canadian-born and most immigrants coming from less-developed countries is well-documented in Canada (see, for instance, Frenette and Morissette, 2003; Picot, 2008). Further, previous research suggests that immigrants from the Philippines collectively have lower incomes than the national average and experience one of the highest levels of occupational and sectoral segmentation, primarily into low-paying and insecure jobs in healthcare, childcare, clerical, retail and manufacturing (for instance, Lindsay, 2001; England and Stiell, 1997).3
Assuming that many Canadian employers who use education as a screening device assess skills acquired abroad in a manner that does not significantly change over time, this evidence at the very least suggests the possibility of good employment opportunities for recent immigrants who completed their postsecondary studies in the Philippines, as compared to the Canadian-born and other immigrants with a postsecondary education. It might be that when given the opportunity to look for paid work that may be more suitable to their level of human capital, recent immigrants who completed their postsecondary education in the Philippines were more likely to secure employment in occupations generating high wage equilibriums in the Canadian economy (such as natural and applied sciences occupations), as compared to the Canadian-born and other internationally-educated immigrants. Descriptive statistics reported in Table 5 provide some support to this explanation: 74% of recent immigrants with a postsecondary education completed in the Philippines graduated in generally high-paying fields of study such as engineering (24%), business and management (29%) and health (20%), whereas this was the case for instance for 57% of the Canadian-born, 53% of recent immigrants with a postsecondary education from the U.S. and 61% of those who completed their studies in the United Kingdom.
C. Established immigrants
Similar to results reported earlier, there is an earnings gap between established immigrants with a postsecondary education and their Canadian-born counterparts (Charts 2.A to 2.D). In fact, only established immigrants who completed their postsecondary education in the United Kingdom enjoyed a 5% earnings premium relative to the Canadian-born. Other studies have found similar results (see, for instance, Frenette and Morissette, 2003; Baker and Benjamin, 1994). In particular, Frenette and Morissette (2003) report that 20 years after arriving in Canada, the employment earnings of male and female immigrants were 14% and 16% lower, respectively, than the employment earnings of Canadian-born men and women. This wage gap shows that there is still a possibility that immigrants face various barriers in Canadian labour markets, which do not disappear with duration of permanent residence in Canada. As Chiswick (1978) suggests, occupational segregation and wage discrimination may foster low earnings among highly-skilled and highly-educated immigrants in comparison with non-immigrants of comparable measured attributes.
There are striking differences across the three categories of education-job matches. Employees classified as being overeducated display an earnings portrait that mirrors that of very-recent and recent immigrants, that is, established immigrants with a postsecondary education completed in the United Kingdom earned on average, 7% more than their Canadian-born counterparts, whereas established immigrants from other countries showed a negative earnings gap, everything else held equal. Regarding correctly-matched workers, established immigrants with postsecondary credentials earned abroad had an earnings premium compared to the Canadian-born. But, established immigrants with a postsecondary education completed in Pakistan (+41%), India (+64%) and Germany (+40%) had higher earnings advantages relative to the their Canadian-born counterparts. Those with a Canadian postsecondary education who were correctly matched had employment earnings comparable to those of their Canadian-born counterparts, while those who completed their postsecondary education in the United States earned slightly less, on average.
Concerning employees classified as being undereducated, established immigrants with a postsecondary education completed in South Korea earned on average, about 34% less than their Canadian-born counterparts, whereas those who completed their studies in the United Kingdom earned on average, about 17% more than their Canadian-born counterparts. These differences are far from being trivial. On the one hand, they indicate that some highly-educated immigrants who completed their studies abroad may face a persistent comparative earnings disadvantage during their working lives. On the other hand, these results imply the existence of some convergence between the employment earnings of most landed immigrants and the Canadian-born over time.
Summing up, Chapter 7 evaluated from a multivariate perspective, the less-than-perfect international skills transferability hypothesis using direct information on location of study. Overall, we found evidence supporting this hypothesis; that is, most foreign credentials have on average, a depressing effect on earnings and employment prospects of immigrants, especially for very-recent cohorts. With these multivariate results in hands, we conclude that in Canada, many prospective employers and several other stakeholders (such as regulatory bodies, assessment agencies, etc.) may not value postsecondary educational qualifications from all source regions or countries on a unique standard. In other words, different economic values may be assigned to "outcomes" of different foreign postsecondary educational systems in Canada. Before concluding, however, the next chapter addresses two potential critiques of these findings.
Notes
- Marginal effects are all calculated at means of explanatory variables.
- This result highlights the importance of performing a multivariate analysis. Indeed, descriptive statistics reported in Table 7 indicate that recent immigrants with postsecondary education and aged 25 to 64, who completed their highest degree in the Philippines, had a comparative wage disadvantage regardless of the level of education-job match and the latter is particularly higher in the undereducated group, at, -42.4%. As shown in regression results, this is not true at all.
- We are grateful to an anonymous referee for pointing this out.
- Date modified: