Chapter 9: Discussion

View the most recent version.

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.

Archived

This page has been archived on the Web.

• 9.1 Contribution of the current study
• 9.2 Other findings
• 9.3 Limitations and caveats

The importance of location of study on the post-migration labour market success of internationally-educated immigrant workers with a postsecondary education has been the focus of much recent Canadian research. But, most prior studies use databases containing no information that allows the location of study to be directly tracked. One solution often proposed in the literature is to resort to the country of birth. A salient assumption in prior research is that this measure can adequately capture location of study, irrespective of international migration dynamics. Interestingly, data not reported here show that among landed immigrants with a postsecondary education aged 25 to 64 who completed their studies outside Canada, the correlation between the country of birth and the location of highest degree completion is: 0.59 when countries are considered separately; 0.68 when countries are regrouped by per capita national income, and; 0.71 when countries are regrouped by geographical regions. These results imply that the country of birth is an imperfect indicator of the origin of skills acquired via education.

9.1 Contribution of the current study

Our report used descriptive and multivariate techniques to compare different cohorts of internationally-educated immigrants to the Canadian-born with respect to labour force status, earnings and the match between occupation and required schooling. Unlike most prior Canadian literature, we took advantage of the information on location of postsecondary education completion first collected by the Census in 2006 Census and simultaneously examined the labour market success of immigrants in terms of their labour force status, wage earnings and the match between jobs and required education.

We found that immigrants' labour market outcomes vary by location of study: those who completed their postsecondary education in the United Kingdom, France, the United States or, to some extent in Germany, were much more likely to do well on Canada's labour markets in terms of employment rates and earnings, regardless of immigration cohort compared to those who completed their postsecondary studies in any other foreign country, especially China, the Russian Federation, Pakistan and South Korea.

The immigration-based research conducted in other countries, such as the United States and Australia also provides evidence suggesting that post-migration labour market outcomes of immigrants are sensitive to their location of study. For instance, Chiswick, Cohen and Zach (1997) show that in the United States, while employment ratios for European or Canadian immigrants are similar to those of Mexican immigrants, they are higher than those of Asian immigrants. Bratsberg and Terrell (2002) show that in U.S. labour markets, education from northern Europe, Australia or Canada is more highly remunerated than education from Central America or the Caribbean.

In terms of Australian evidence, Liebig (2007) finds that immigrants with an Australian education have much better chances of securing paid employment than those with foreign degrees. He attributes this finding at least in part to the existence of imperfect transferability of skills between source countries and Australia. Thapa and Gorgens (2006) show that among female immigrants, those possessing an Australian educational qualification increase their likelihood of finding a first paid job at arrival in Australia. They also find that having a foreign degree assessed as being equivalent to Australian degree generated a similar positive impact on the employment perspectives of female immigrants to Australia.

Arguably, therefore, intra-group variations observed in the post-migration labour market success of Canada's immigrants with foreign postsecondary credentials are not a Canadian idiosyncrasy. Instead, as previously suggested in the international research (see, for instance, Chiswick, Cohen and Zach, 1997; Chiswick and Miller, 2008; 2009; Liebig, 2007) and recent Canadian studies (see, for instance, Ferrer and Riddell, 2008; Gilmore and Le Petit, 2008; Alboim, Finnie and Meng, 2005), they may reflect at least in part the fact that a typical Canadian employer who uses education to sort prospective labour market participants by potential productivity level may attach greater economic value to educational qualifications from countries that have linguistic, economic and socio-cultural similarities to Canada. Conversely, that employer may undervalue educational qualifications from other countries, especially those from China, Pakistan, South Korea, the Russian Federation and Iran. In other words, the average prospective employer may not understand very well, and so may not be in a position to evaluate, the properties of postsecondary credentials from Pakistan, the Russian Federation, China or South Korea for the productivity of prospective labour market participants.

9.2 Other findings

This study provided several other findings. Regarding labour market status, for instance, our multivariate results indicated that, on average, very-recent and recent immigrants who completed their postsecondary education in Canada were less likely to be overeducated and more likely to be out of the labour market, relative to the Canadian-born. When employed in the wage sector, they earned, on average, less than their Canadian-born counterparts, regardless of the level of education-job match. These results led us to argue that individual accumulation of skills through the Canadian postsecondary education is not itself the engine of the labour market success of immigrants to Canada.

Our estimates indicated that among correctly-matched employees, recent immigrants with a postsecondary education completed in the Philippines and or India earned, on average, more than their Canadian-born counterparts. Also, recent immigrants educated in the Philippines and classified as being undereducated enjoyed, on average, a significant wage advantage over their Canadian-born counterparts, even after controlling for other determinants of employment earnings in Canada. The earnings premium enjoyed by recent immigrants with a postsecondary education from the Philippines is surprising because prior research (for instance, Lindsay, 2001; England and Stiell, 1997) indicates that immigrants from the Philippines collectively experience a significant comparative disadvantage in the allocation process of well-paid job on Canada's labour markets. A disaggregation of our data by fields of postsecondary study led us to argue that, when given the opportunity to look for paid work that may be more suitable to their educational credentials in Canada, recent immigrants who completed their postsecondary education in the Philippines or in India were more likely to find a high-paying job, as compared to the Canadian-born and other internationally-educated immigrants.

9.3 Limitations and caveats

To start, we did not directly address the question of what influence location of study has on the speed of convergence of labour market outcomes of highly-educated prime-aged immigrants with those of their Canadian-born counterparts. Such an analysis requires longitudinal data since labour market behaviour is a function of permanent idiosyncrasies and states dependency. For example, states dependency may rise from the absence of temporal separability between the portions of time spent on non-market activities over a number of successive periods (Hotz, Kydland and Sedlacek, 1988). They may also result from the fact that the costs of searching for new paid employment vary with the quality of the current and previous jobs (Eckstein and Wolpin, 1990).

Also, employment history may send a signal to potential employers concerning the quality of productive attributes of prospective labour market participants, so that some individuals frequently end up at the tail of the paid employment distribution, despite their higher educational qualifications (Lee and Tae, 2005). Moreover, unobservable characteristics that drive some people to initially accept jobs that economically undervalue their labour force skills may be similar to those which keep them at the tail of the earnings distribution for a long time. It should therefore be kept in mind that the analysis reported here does not allow us to directly predict the impact of location of study on labour market "assimilation" of highly-educated immigrants after their arrival in Canada. This is an area that should be pursued in future research should longitudinal analysis become possible, especially if the fraction of immigrant workers who received their highest postsecondary education abroad grows substantially.

Likewise, it is worth stressing that for the sake of comparison, we follow prior Canadian research (see, for instance, Plante, 2010; Gilmore and Le Petit, 2008; Mata, 2008) by allocating immigrants into three broad and mutually-exclusive categories depending on the duration of permanent residence in Canada, namely: very-recent, recent and established immigrants. But, location of study could have differential effects on particular arrival cohorts either because of transitory or permanent fluctuations in the value allocated to foreign education in the host country's labour markets over time or because of the fact that unobserved factors determining labour productivity (such as motivation, talent, etc.) are fixed within an arrival cohort (Lalonde and Topel, 1992). Given that the data used in this report are cross-sectional, our main estimates may thus be capturing, to some extent, cohort effects rather than the dynamics of landed immigrants' integration into the Canada's labour markets.

Authors such as Edin, Lalonde and Aslund (2000), Borjas and Bratsberg (1996) and Lalonde and Topel (1992) argue that emigration of landed immigrants may not be a residual phenomenon in many destinations favoured by international migrants (such as the United States, Canada, Australia, New Zealand, etc.). Further, according to authors such as Edin, Lalonde and Aslund (2000), economic migrants, especially those who arrived within the previous five years, may have higher emigration rates than other immigrant categories because they tend to compare the present discounted value of earnings in the host country to that in the source country. Hence, standard measures of labour market success of immigrants in the host country (such as labour force status, employment earnings and the match between their jobs and required qualifications), especially shortly after arrival, may be biased to the extent that "there is non-random emigration of immigrants."

As shown in Chart 3, long-term immigrants to Canada are not uniquely selected on the basis of their skills and ability to contribute to Canada's economy or successfully enter its labour markets. However, there is a constant overrepresentation of economic immigrants (i.e., skilled workers, business immigrants, and live-in caregivers) among people accepted for permanent residence in Canada starting in 1987, with a historical peak in 2001, at 67%.

Chart 3 Percentage distribution of permanent migration to Canada, by category, 1984 to 2006

Selective emigration of landed immigrants, especially economic immigrants, could be non residual in Canada because the empirical migration research (see, for instance, Mayda, 2005) has taught us that international migration is determined by economic considerations (such as poverty, unemployment, prospects for higher wages, improved standard of living). Further, the recent inclusion of labour mobility provisions for many professional occupations in international agreements such as the North American Free Trade Agreement and the General Agreement on Trade in Services of the World Trade Organization has significantly eased international labour force mobility, especially between Canada and the United States (Blouin, 2005).

Thus, if a sizeable proportion of immigrants with a postsecondary education aged 25 to 64 who completed their postsecondary studies outside Canada, especially economic immigrants, emigrated back to their home country or to a third country, estimates of the impact of location of study on labour force status and earnings may suffer to some extent from selection bias. This is because constructed arrival cohorts only account for those landed immigrants who stayed in Canada, not for those who left. Indeed, if internationally-educated immigrants who remained in Canada have poorer (better) labour market outcomes than those who left, our results may suffer from an underestimation (overestimation) of variations of the gaps in earnings and employment status between immigrants from different locations of study and the Canadian-born.

Given that the testing of selection bias is beyond the scope of our analysis, we can only speculate that the global picture of immigrants' labour market success in Canada would have not changed dramatically if internationally-educated immigrants who possibly used Canada as a step for other North American labour markets have better outcomes would have they stayed. Nevertheless, the caveat that our findings could partly reflect the effect of exit among some high-earners educated abroad needs to be kept in mind. To cite Jasso, Rosenzweig and Smith (2002), it could be that some of our results "confound the skills transferability of an individual over time and changes in the skill composition of immigrants due to selective emigration."

As Chart 3 shows, Canada generally accepts international migrants under economic, humanitarian and family considerations. Understandably, therefore, the negative effect of pre-immigration schooling on the labour market outcomes of a sizeable number of immigrants who landed as refugees may be explained, for instance, by problems experienced in providing prospective employers with concrete proof of self-reported educational attainment. It would have been preferable to consider asylum-seekers and refugees as a separate category. Unfortunately, the data from the 2006 Census of population are not available in a form that allows us to identify the routes by which immigrants entered Canada.