While perhaps somewhat more muted than elsewhere, research shows immigrants to Canada suffer hidden and not-so-hidden socio-economic barriers that are exacerbated during times of economic hardship. Unfortunately, as in many other areas, the research available on this has taken a hit in recent years.
Historically, most of the research examining the socio-economic outcomes of immigrants to Canada relied on data from the national long-form census. Several major studies using long-form data have found a growing gap in earnings between landed immigrants and the Canadian-born dating back to the 1970s (and the size of the gap has been significantly influenced by whether one arrived from a “traditional” or ‘non-traditional’ source country). But the long-form census was cancelled as of 2011, and the voluntary survey conducted in its place has proven both far more costly and less reliable.
The long-form census was not without its shortcomings. In addition to only providing a snapshot once every five years, the released data excluded first-year immigrants’ incomes, making it less useful for research on recent immigrants’ socio-economic outcomes. For this purpose, the Longitudinal Survey of Immigrants to Canada (LSIC) was undertaken between 2001 and 2005. One LSIC report noted that, despite two-thirds of landed immigrants being university educated – a much higher ratio than the Canadian-born workers – nearly half struggled to find employment four years after arrival. But the LSIC has been “inactive” since 2005.
The administrative data equivalent of the long-form census, the Longitudinal Administrative Database (LAD), suffers from a number of shortcomings associated with administrative data, such as scope, selection bias, etc. (Statscan also strictly limits access to the LAD.) One research paper based on the LAD and long-form census data found immigrants arriving to Canada in 1990 – in the midst of the last major recession – were 50 per cent more likely to leave than those arriving a few years prior (1986). Among immigrants who arrived under either the “business” or “skilled worker” visa classification, 40 per cent left within 10 years of arrival.
There are other voluntary national surveys that collect data on immigrants, such as the Canadian Community Health Survey (CCHS), Survey of Labour and Income Dynamics (SLID) and General Social Survey (GSS). Self-selection bias and small sample size limit their reliability and usefulness. These social surveys make use of the Labour Force Survey’s (LFS) sampling frame, which is updated using the census.
This has led to a question some have asked over the years: Why doesn’t the LFS, the country’s flagship survey for labour trends, simply include questions on immigration?
Well, it actually does, and has since January, 2006 – though Statscan’s monthly LFS reports have never mentioned it, limiting the demographic analysis to age and gender. The agency has, however, released occasional immigrant labour force analysis reports, with a view to including immigration in the regular monthly reports “some time this year.”
First, the good news. If/when the LFS begins providing information on immigrants’ labour market experience on a monthly basis, it will be the first Statscan social survey to do so. While the length of the series doesn’t allow for comparison with the last major (early 1990s) recession, it does begin a couple of years prior to the Great Recession, which allows for some interesting analysis.
For example, the available data show that in October, 2008 (when LFS began to capture major job losses related to the recession), the unemployment rate gap between landed immigrants and Canadian-born workers was less than one percentage point; it widened to nearly three points by mid-2009, remaining near that level until late 2010. The gap began to significantly narrow in 2011, the same year the (mostly Canadian-born) baby boomer cohort began to turn 65. The LFS data began to show a decline in overall employment for Canadian-born workers in 2012.
Now the not-so-good news: While the LFS appears to have captured some interesting immigrant labour market trends, whether it accurately captured magnitude is another matter.
Despite greater focus on immigrants in large urban centres, the LFS sample significantly under-represents landed immigrants (16.2 per cent in the sample, 23.5 per cent in the general population). Another shortcoming is that temporary foreign workers (TFWs) are not clearly distinguished in the survey; depending on how they respond to a couple of questions, TFWs may or may not be included in the LFS. If they are, they’re lumped in the immigration category “Other,” along with foreign students, refugee claimants and other residents not born in Canada. The immigrant “Other” labour market information is not publicly disclosed.
That said, it’s promising that regular, long-running national social survey such as the LFS is finally inquiring about immigrant socio-economic outcomes, even if specifically focused on the labour market. Barring a few issues that could be addressed over time, the LFS may yet prove to be a useful tool in analyzing barriers to immigrants’ labour market entry and advancement.
Sam Boshra is an independent Montreal-based economist, and editor for EconomicJustice.ca.