ARCHIVED – Labour Market Outcomes for Migrant Professionals: Canada and Australia Compared

by Lesleyanne Hawthorne
University of Melbourne
Australia
December 2006

Copies of the full report are available upon request to Research-Recherche@cic.gc.ca.

Labour market outcomes for migrant professionals: Canada and Australia compared — Executive summary
[ PDF Version (PDF, 86.86KB) ]

Executive summary

A. Skilled migration to Canada and Australia

The Research Context

Economic Migration

Canada and Australia are global exemplars of nation-building through government planned and administered economic, family and humanitarian migration programs. By 2005 Australia included the world’s highest percentage of foreign-born (24.6% of the population, with over 240 nationalities) followed by Canada at 19.2% and the US at 11.7% (Miller 2005).

Within the past decade Canada and Australia have also placed extraordinary emphasis on the recruitment of migrants with skills. In 2004 Canada selected 133,746 people in the economic category, in particular substantial numbers of points-tested Principal Applicants qualified in the professions. Skilled migrants constituted 59.6% of Canada’s total planned intake at this time (224,346 people), far exceeding the targets set for family (51,500-56,800) and refugee/humanitarian (30,800-33,800) entrants. The proportion of economic migrants selected by Australia in 2004-05 was virtually identical to Canadian levels (58%), based on 77,800 applicants out of a permanent migrant/humanitarian intake of 133,000 people, with the 2005-06 target since set substantially higher (97,500) (Birrell, Hawthorne & Richardson 2006).

Despite numerical similarity in terms of skilled migration programs, primary sources for economic migrants now vary markedly between Canada and Australia. Canada has placed unprecedented reliance in the recent period on migration from developing countries, most notably China (18%), India (11%), Philippines (7%), Pakistan (4%) and Romania (4%). While China and India feature strongly in Australia’s skilled migration program, it has continued to maintain strong UK/Ireland and South African flows — the top 5 source countries for 2004-05 being the UK/Ireland (25%), India (13%), China (11%), South Africa (5%) and Malaysia (5%).

Further important differences exist. Most significantly, since 1999 Australia has developed substantial onshore as well as offshore economic migration flows, based on ‘two step migration’. By 2005 former international students with host-country degrees constituted 52% of all economic migrants — in particular Principal Applicants from India and China, who had self-funded in advance to address local employers’ language, training and credential needs (Hawthorne 2005). Such students had previously been ineligible to participate in the program.

In addition to economic migrants, it is important to note, large numbers of degree-qualified arrivals reach both Canada and Australia via family and humanitarian immigration categories. These applicants are unfiltered by selection criteria at point of entry, yet will be strongly motivated to work in their original fields, regardless of human capital requirements such as host-country language facility or foreign credential recognition.

Temporary Immigrants

In terms of degree-qualified migrants it is further important to factor in the growing number of skilled workers entering Australia and Canada on a temporary basis, with sponsorships providing incontrovertible evidence of local employer preference. In 2004, for example, Canada approved around 250,000 temporary visas, including 99,700 for Foreign Workers and 56,700 for international students. In nominating sources for Foreign Workers, Canadian employers’ countries of choice diverged markedly from contemporary Canadian economic migration sources.

Australia approved far greater numbers of skilled temporary arrivals at this time, reflecting its post-1996 highly deregulated temporary policy environment. ‘Short term’ flows included 339,424 business entrants, 104,353 Working Holidaymakers, and 174,787 international students. As noted above, substantial numbers of these arrivals would ultimately participate in ‘two-step migration’ — reaching Australia on a temporary basis, then exercising the potential to category-shift and stay. This strategy represents a fast developing global phenomenon, frequently associated with excellent labour market outcomes.

By the time of the 2001 Census, reflecting all such flows, Canada’s population included 3,374,057 degree-qualified arrivals and 3,801,118 migrants with post secondary diplomas or certificates, compared to 1,769,154 degree and 933,889 diploma qualified migrants in Australia. From 1996-2001 newly arriving migrants were more than twice as likely as the Canada-born to be degree-qualified (37% compared to 15%). Similar patterns were evident in Australia (26% compared to 14%). Increasing numbers of skilled migrants in each country however had struggled to achieve appropriate employment, particularly in the first 5 years of arrival, regardless of the state of the host economy.

Business cycles have been remarkably similar in Canada and Australia in the recent decade, providing an excellent ‘laboratory’ for contrastive research concerning employment outcomes for skilled migrants. According to Richardson & Lester, from an economic perspective Australia and Canada ‘look very alike… During the 1996-2001 period, during which migrants included in (both countries’ Longitudinal and Census surveys) arrived and were looking for employment, both economies performed equally well. Thus (any) superior labour force performance of migrants to Australia… cannot be explained simply in terms of economic performance’ (Richardson & Lester 2004: 10). It is within this context that the comparative analysis of labour market integration for degree-qualified migrants (below) was undertaken.

Selection Systems for Economic Category Migrants: Canada and Australia Compared

While both Canada and Australia use points-based selection criteria to select economic migrants, it is important to note that there has been sharp divergence since 1996 on the values and priorities informing these programs. The primary goal for Canada remains nation building, based on sustained high-level intakes regardless of economic cycles, and informed by a human capital model of immigrant selection. According to Hiebert (2006), the prevailing Canadian view is that ‘well-trained flexible individuals… who have experience in the labour force’ should be able to ‘adapt to rapidly changing labour market circumstances’. In consequence ‘general’ rather than ‘specific’ competence is sought — Canadian selection criteria admitting Principal Applicants (PAs) with limited or no host country language skills, non-recognised qualifications, and in fields of minimal labour market demand on an equal basis to those with more immediately sought after attributes (Birrell, Hawthorne & Richardson 2006).

Australia by contrast progressively abandoned the human capital model from 1996 as ‘out of balance and out of control’ following the election of the current conservative government. While the family and humanitarian intakes were endorsed as serving broad social purposes, serious unemployment among recently arrived skilled migrants was perceived to undermine the effectiveness of the economic migration program — one explicitly devised in 1988 to support Australia’s economic goals, and in which it was legitimate to prioritise national interests.

The Australian government in 1996 defined six attributes as making ‘a good skill(ed) applicant’, most notably ‘obtaining a job soon after arrival that uses their skills… become quickly established’ and ‘(n)ot require benefits’. In commissioning successive reviews of the program in 1997, 1999, and 2006, the government determined to use the research evidence to finetune economic selection criteria, in the process optimising immediate as well as long-term employment outcomes.

Reflecting the findings of these reviews, since 1999 an increasing number of Principal Applicants at perceived risk of delayed or de-skilled employment have been excluded from migration to Australia at point of entry, through rigorous expansion of pre-migration English language testing (extended to family-skill categories), mandatory credential screening, assessment of labour market demand, plus a range of additional modifications to the points selection process. In the decade since, Australia has secured early and increasingly positive employment outcomes, with results at six months strongly correlated to longer-term labour market integration rates. The latest available data (May 2006) confirm additional benefit from the most recent policy refinements (Birrell, Hawthorne & Richardson 2006). These post-1999 results far exceed the level of benefit attributable to Australia’s improved domestic business cycle.

B. Differences in work outcomes for degree-qualified migrants in Canada and Australia: 2001 census data

A major goal of the current study was to assess factors associated with positive or negative employment outcomes for degree-qualified migrants in Canada and Australia, in the first instance through examination of 2001 Census data for 1996-2001 compared to earlier arrivals.

By definition Census data contain all degree-qualified groups, including those reaching Canada and Australia via economic, family and humanitarian migration categories. The aim of the preliminary data analysis was therefore to define ‘typical’ labour market integration rates for migrants with otherwise comparable characteristics, in order to:

• Compare the level of Canadian and Australian receptiveness to overseas qualified professionals in the context of very similar economic cycles; and
• Provide a ‘control’ group for the subsequent analysis of outcomes for points-tested economic category Principal Applicants (in order to define whether the Canadian or Australian programs currently perform better).

Labour market outcomes for professionals in 10 qualification fields were examined (engineering, IT, accounting, medicine, nursing, teaching, architecture and building, social sciences, creative arts and humanities, natural and physical sciences, and management and culture). The employment impact of major variables including birthplace, gender, age, field and place of qualification, settlement location and date of arrival was systematically assessed.

The key Census research findings are summarised in turn below.

1. The analysis confirmed there to be negligible differences in employment outcomes for recently arrived degree-qualified migrants in Canada and Australia (all migration categories).

By 2001, 65% of degree-qualified 1996-2001 arrivals had found work of some kind in Canada, compared to 66% in Australia. 29.8% held professional positions, compared to 31.4% in Australia. 5.0% of migrants had switched to administrative or managerial work, compared to 7.6% in Australia, with very substantial additional numbers clustered in lower skilled positions.

Unemployment represented a more significant problem in Canada, with 14.7% of degree qualified arrivals unemployed compared to 7.8% in Australia. Australia by contrast had larger migrant numbers categorised as ‘not in the labour force’ (typically learning English or trying to secure credential recognition): 26.2% of 1996-2001 arrivals compared to 20.4% in Canada. This higher proportion almost certainly reflected Australia’s sustained investment in the establishment of language and labour market training programs since the mid 1980s.

2. Degree and higher-degree qualified migrants enjoyed a substantial labour market advantage in both countries.

Relative to recently-arrived migrants holding diploma level qualifications, recent degree qualified migrants are generally able to secure double or more the rate of professional employment in the host country, despite results for diploma-level migrants being somewhat better in Australia.

In Canada, where the data allow differentiation of employment outcomes by degree type, superior outcomes were secured by migrants holding Masters or PhD degrees, for all periods of arrival. For example 61% of pre-1991 arrivals with higher degrees held professional positions by 2001, in contrast to 45% of bachelor-qualified arrivals. The comparable data for 1991-96 was 53% (versus 31%), and for 1996-2001 44% (versus 24%). This outcome compared reasonably with employment outcomes for the Canada born (64% of Masters or PhD qualified local workers holding professional positions by 2001, compared to 52% of those with bachelor degrees.)

3. Degree-qualified migrants however were slightly more de-skilled in work in Canada than in Australia.

While 64.9% of degree-qualified 1996-2001 arrivals to Canada were employed by 2001, as we have seen just 29.8% were in professional work, 5.0% in administration/management and 6.3% in associate professional roles. The comparable rates for Australia were 31.4%, 7.6% and 6.2%. Overall 23.8% of degree-qualified recent arrivals had accepted very de-skilled work in Canada, compared to 20.7% in Australia.

4. Trade qualified migrants also performed well in both countries.

In Canada 65% of degree-qualified 1996-2001 arrivals had secured work by 2001, compared to 62% of migrants holding post-secondary diplomas. In Australia 66% of degree-qualified migrants had found work in the first 5 years, compared to an impressive 73% for those with vocational qualifications. These findings justify the current levels of interest in more trade migration to each country.

5. In the past decade, it is important to note that Canada and Australia have diverged markedly in terms of source countries for degree-qualified migrants, in particular the level of English-speaking background country migration.

Prior to 1991 25% of degree-qualified migrants to Canada were derived from English-speaking source countries (the UK/Ireland, US, Australia, New Zealand and South Africa). This plummeted to just 7% from 1991-1996 and to 5% from 1996-2001. Australia by contrast maintained far higher levels of degree-qualified English Speaking Background migrant flows — reduced from 38% to 20% from 1991-1996, but reverting to 28% from 1996-2001 following consistent evidence of the inferior labour market integration rates secured by select non-English speaking background groups.

From 1996-2001, for example, just 6% of doctors, 4% of nurses, 2% of engineers and 2% of IT professionals migrating to Canada were derived from ESB source countries. This compared to 30%, 43%, 22% and 18% respectively migrating in these fields to Australia. The latest available advice suggests UK flows to Canada to remain negligible, despite their clear attractiveness to employers.

6. Within the knowledge economy birthplace exerts major influence on employment outcomes for degree-qualified professions, particularly in the first settlement years.

The following degree-qualified 1996-2001 arrivals to Canada were the most likely to have secured professional work by 2001: South Africa (over 60% of migrants working in their own or another profession), Australia and New Zealand (close to 60%), UK/Ireland (over 50%), North West Europe (ditto) and the US (close to 50%). This compared to an almost identical rank order for 1996-2001 arrivals securing professional employment in Australia: UK/Ireland (51%), South Africa (50%), New Zealand (48%), US/Canada (40%), and North West Europe (39%).

Within both Canada and Australia the likelihood of degree-qualified migrants securing professional work in the first five years of migration dropped substantially for other birthplace groups, with many migrants at severe risk of de-skilled employment. Similar birthplace groups faced the greatest level of disadvantage in each country. Degree-qualified migrants from Iraq, Taiwan and Other North and South East Asia ranked lowest in terms of work access, followed by the Lebanese in Canada and the Vietnamese in Australia. Large numbers of degree-qualified migrants from the Philippines, India, Vietnam and Other South/Central Asia had gained employment only at the cost of taking low-skilled work — a significant migration policy issue in the context of the scale of these arrivals.

7. The critical determinant of migration employment outcomes between Canada and Australia proved to be the level of UK/Ireland migration.

The UK/Ireland was the source of 1.6% of degree qualified migrants to Canada from 1996-2001, compared to 15% to Australia. Had UK/Ireland migration been excluded from both countries, Australia’s rate of recently arrived degree-qualified migrants employed in any profession would have dropped from 31% to 28% (compared to Canada’s 30%).

8. ‘Visible minority’ status did not appear to unduly influence labour market integration rates for degree-qualified migrants in Canada or Australia within this period.

Recently arrived Commonwealth Asian and African migrants were reasonably well accepted by Canadian and Australian employers — in terms of labour market integration generally ranking well ahead of other Asian and Middle Eastern groups (a long established pattern in Australia). It is worth noting that Hong Kong, Malaysian and Singaporean migrants had performed particularly well — in Canada far more likely to secure work in their own profession than similarly qualified Indian arrivals (19% versus 12%).

9. Newly arrived degree-qualified migrants secured highly comparable representation in the professions in each country within the first few years post-arrival.

Overall 31.4% of migrants with degrees were employed in their own or some other professional position in Australia by 2001, compared with 23.5% of migrants in Canada with Bachelor degrees, and 43.9% holding Masters degrees. 7.6% of such migrants had found managerial or administrative positions in Australia, compared with 5.2% (Bachelor degrees) and 4.6% (Masters degrees) in Canada. As noted, 6.2% had slipped to associate professional positions in Australia, compared to 6.3% in Canada.

In terms of outcomes by birthplace, it is worth noting that a range of recently arrived degree qualified migrants had performed better in Canada than in Australia. For example 25.0% of North West Europe migrants were employed in their profession by 2001 (compared to 21.7% in Australia), 19.1% from Hong Kong, Malaysia % Singapore (versus 17.9%), 17.9% from Central and South Americas (versus 11.2%), 14.9% from China (versus 12.7%) and 12.2% from India (versus 15.5%). Factoring employment in ‘other professions’ or managerial positions into the analysis intensified this pattern.

10. Newly arrived degree-qualified migrants had higher unemployment rates in Canada, but higher ‘not in the labor force’ rates in Australia.

Overall, newly arrived degree-qualified migrants from disadvantaged groups reported lower unemployment levels in Australia than Canada, at a time when national unemployment rates for locals with degrees was 4% in Canada and 2% in Australia.

The unemployment rate for migrants from Other Middle East / North Africa was 21% in Canada (compared to 12% in Australia), 19% for China (compared to 9% in Australia), 17% for South and Central Asia (10%), 17% for South Eastern Europe (11%), 15% for Taiwan (6%), 14% for East Europe (10%), and 13% for India (10%).

Many birthplace groups also included high proportions reported as ‘not in the labour force’: most notably migrants from Taiwan, Other North and South East Asia, Indonesia, Lebanon, and Iraq. Substantial numbers of these migrants, it seems fair to presume, would have been learning English/French, or re-positioning to enter the labour market through study to achieve credential recognition.

In line with the literature, far better outcomes were achieved by long-established degree qualified migrants in both Canada and Australia (defined as resident 10 years or more in the host country). Encouragingly, this pattern included the most initially disadvantaged groups, with unemployment rates more nearly approximating host country norms. For example by 2001 English speaking background migrants in Canada 10 years or more had achieved equal or superior representation in the professions to the Canada-born. South Africans (as in Australia) represented the most elite migrant group overall, with 87% working compared to 85% of the Canada-born, and a high 71% occupying professional or managerial positions (compared to 64% of the Canada-born). Select Asian groups had also achieved representation in the professions equivalent to the Canada-born, including those from Vietnam and Hong Kong, Singapore and Malaysia, with the Indonesia-born and China-born also faring well. High rates of access to the professions had been achieved by additional visible minority groups, including migrants from Other Africa, the Middle East and North Africa — all with better representation in their own and other professions than migrants born in East and South East Europe, India, other parts of Asia, Taiwan and the Philippines.

Long-established degree-qualified migrants however had achieved somewhat higher levels of employment in their own professions in Australia than in Canada: most notably those qualified in Hong Kong, Malaysia and Singapore (34% versus 25%), Other South and Central Asia (27% versus 18%), India (26% versus 18%), South East Europe (24% versus 20%) and Iraq (18% versus 14%). This gives credence to the overall view that visible minorities may experience greater disadvantage in Canada, despite the fact that 1996-2001 arrivals from a range of such groups had secured less initial access to their professions in Australia.

11. The level of labour market demand exerted a powerful influence on employment outcomes for degree-qualified recent arrivals in both Canada and Australia.

Engineering

In engineering, for example, where there was an 18% growth in professional positions from 1996-2001 in Canada compared to a major period of contraction in Australia, new labour market entrants in Canada performed well. 66-75% of recently arrived ESB engineers had gained some form of professional or managerial work by 2001. A wide range of other migrant engineers had also fared positively, including those from select visible minority groups. For example 21% of degree-qualified engineers from Central & South America were working as engineers by 2001 (compared to 6% in Australia), 19% of Indian and East European arrivals (compared to 9% and 8% respectively in Australia), 16% of Iraqis (compared to 6% in Australia), and 15% from Other Middle East/North Africa (compared to 6%). Overall, 39-51% of recently arrived engineers had secured some form of professional or managerial work in Canada by 2001 — a positive outcome, and remarkable in the light of local regulatory hurdles. Outcomes proved less positive however for recently arrived engineers from China (12% employed in engineering positions compared to 8% in Australia) and the Philippines (5% compared to 6%). This represents a serious issue given the dominance of such source countries in contemporary Canadian engineer flows. The findings gain further significance given the scale of recent engineering migration: 26,639 Skilled Worker and 2,619 Assisted Relative arrivals between 2001-03.

Information Technology

In information technology, where dynamic labour market demand existed to 2001 in both Canada and Australia, analysis of the Census confirmed there to be profoundly beneficial impacts for recently arrived migrants. For example 57% of recently-arrived IT degree-qualified Eastern Europeans had secured professional work in their field by 2001 in Canada (compared to 63% in Australia), 49% from South Eastern Europe (68%), 43% from HK, Malaysia and Singapore (27%), 37% from India (34%), 41% from China (36%), 38% from Other Middle East/North Africa (37%), 37% from South and Central Asia (46%) and 32% from the Philippines (42%). In Australia virtually every birthplace group examined in IT achieved double or triple its standard employment level in ‘own profession’ within the first five years, employers appearing willing to overlook perceived deficits in terms of prior training or language ability in the context of very buoyant labour market demand.

Nursing

Excellent Australian labour market integration rates were also evident in the medical and nursing professions, in a context where demand constantly outstripped domestic and immigration supply. By 2001 virtually every category of 1996-2001 nurse arrivals had performed well in the labour market despite regulatory hurdles: 73% of UK/Ireland nurses working in their field, along with 66% of nurses from India and Hong Kong, Malaysia and Singapore, 63% from South Africa, and 49% from Other South/Central Asia. Even a high 52% of nurses qualified in China had found professional employment, in marked deviation from the typical China professional outcomes. The contrast with Canada where national demand was lower was stark: 22% of recently-arrived Indian nurses securing work in their field in Canada (compared to 66% in Australia), 22% of Filipino nurses (versus 35%), and 32% from North West Europe (versus 45%). Large numbers of recently arrived nurses from China remained unemployed in Canada (28%) or were categorised as ‘not in the labour force’ (25%).

Medicine

Comparable outcomes were characteristic of medicine in Australia, once again reflecting the strength of field-specific domestic demand. Doctors from South Africa and the UK/Ireland had moved seamlessly into medical employment, their rates close to or exceeding local graduate norms (a finding confirming the three mandatory pre-registration exams to present minimal barriers). Doctors qualified in India (66%), Hong Kong, Malaysia and Singapore (59%) and Taiwan (57%) had also achieved excellent integration rates in medicine within the first 5 years. By contrast recently arrived China-qualified doctors fared appallingly in Australia (just 5% working in medicine by 2001), followed by doctors from Eastern Europe (24%), the Philippines (33%), South Eastern Europe and Other Middle East/North Africa (36%) and Other South/Central Asia (39%). In the context of lower demand doctors in Canada generally had far more negative outcomes, despite those resident 10 years or more doing well (as in Australia). Just 3% of Filipino doctors had found medical positions by 2001, compared to 4% from China, 8% from Eastern Europe, 11% from Iraq, 12% from Other South and Central Asia (probably predominantly Pakistan), 19% from India, and 31% from Hong Kong, Malaysia and Singapore.

General Qualification Fields

It is important to note that inferior labour market outcomes for degree-qualified migrants were associated with generic degrees (examined for humanities, science and commerce graduates) in both countries. Policy makers should recognise that most generically-qualified migrants will fail to secure professional or managerial work, despite their possession of degrees and regardless of length of settlement. This form of credential also significantly worsens outcomes for migrants from non-English speaking background countries.

12. The gender of degree-qualified migrants had a major impact on employment patterns in both Canada and Australia — a significant finding in the context of the growing participation of women in economic migration, and their disproportionate qualification in traditionally ‘male’ professions.

In Canada female migrants performed significantly worse than migrant males of the same age group and field, as well as the Canada-born of either gender. For example 66% of recent migrant female IT professionals had found professional or managerial work by 2001, compared to 71% of migrant males, 78% of Canadian females and 81% of Canadian males. (The comparable figures for engineering were 51% compared to 62%, 72% and 80%; for architecture and building 45% cf 58%, 74% and 83%; for medicine 62% cf 81%, 87% and 93%; and for accounting 54% of 68%, 83% and 87%.)

Employment outcomes were also consistently worse for migrant women than males qualified by generic degrees. In management and commerce, for instance, 49% of migrant females had secured professional or managerial work, compared to 61% of migrant males, 68% of Canadian females and 73% of Canadian males.

Highly comparable trends were evident for recent migrant women resident in Australia. For example 57% of migrant female IT professionals had secured professional or managerial work, compared to 67% of migrant males, 67% of Australian females and 78% of Australian males. (The figures for engineering were 38% — far worse than in Canada — compared to 57%, 67% and 75%; for architecture and building 47%, versus 58%, 65% and 68%; for medicine 72%, versus 87%, 89% and 95%; and for accounting 49%, versus 67%, 71% and 78%.) Once again, outcomes were inferior for migrant women qualified in generic fields.

Despite this trend, the study found that young recently arrived migrant women in Australia had approximated or exceeded male migrants’ professional integration rates across a range of fields. Further, they had achieved higher professional or managerial employment rates than comparable males in the natural and physical sciences, medicine, nursing, teacher education, and management and commerce, with near equivalence in all other fields examined. Within this context, gender-related outcomes merit sensitive monitoring by immigrant-receiving governments.

13. The age of recently-arrived degree-qualified migrants significantly affected employment outcomes, although this pattern was also influenced by demand by field.

In Canada and Australia, in every professional field examined but IT, medicine and nursing, new locally-born graduates enjoyed greater access to work in their profession than 1996-2001 migrant arrivals (all ages).

Employment outcomes by age and gender were most similar to the Canada-born for degree qualified migrants from English-speaking background source countries (the UK/Ireland, USA/Canada and South Africa) in addition to those from North West Europe. A similar pattern was found in Australia.

25-44 year old degree-qualified male arrivals fared best of all recent migrants to Canada and Australia, outperforming the older age group (45-64 years), and doing infinitely better than newly arrived overseas-qualified graduates (15-24 years, see next section). For example 75% of recently arrived 25-44 year old male UK/Ireland graduates to Canada had secured work in their own or another profession in Canada by 2001, compared to 57% of 45-64 year olds. The comparable figures for Australia were 74% and 67%.

Young overseas-qualified graduates by contrast faced catastrophic levels of labour market rejection, if not derived from ESB source countries. For example just 9% of young accountants found work in their field in Canada within the first 5 years of arrival, compared to 8% of engineers, 11% of commerce graduates, 16% of teachers and 22% of IT professionals. The comparable figures for Australia were 20% (accounting), 26% (teaching), 5% (commerce) and 16% (IT). Given a choice between degree-qualified locals or new migrant graduates of identical age, Canadian and Australian employers emphatically preferred the former, including in the more challenging context for young graduates which appeared to be characteristic of Canada.

14. It is highly advantageous for young degree-qualified migrants to possess host country rather than overseas degrees — a finding affirming Australia’s growing selection of ‘onshore’ (ie former international student) migrants. However length of residence in the host country also matters.

Young migrant arrivals (1991-96) with Australian degrees had achieved less positive employment outcomes than those resident in Australia 10 years or more: employers clearly favouring graduates with strong English skills and local acculturation. For example 77% of Australia-born new accounting graduates had secured work in 2001, compared to 70% of overseas-born local graduates resident 10 years or more, and 55% of those in Australia just 5-10 years.

Overall employment rates by select country of origin for 1991-96 arrivals with Australian degrees were as follows: 34% for UK/Ireland born youth, 39% for Eastern Europe, 38% for Other Middle East/North Africa, 34% for North-West Europe, 31% for India and Other South/Central Asia, 28% for the Philippines, 26% for Hong Kong / Malaysia / Singapore and 19% for China.

The most recent available data (May 2006) reveal 83% of economic migrants with Australian degrees (all fields) to have found work within 6 months. However just 46% of these young graduates were using the credentials they had gained at this time, compared with 63% of offshore applicants (who had 82% overall employment rates). In line with this new graduates were also paid substantially less than more experienced migrants.

The Canadian data confirmed a generally tougher labour market entry for new graduates than in Australia. For each vintage of young migrant graduates examined the level of professional employment was lower, despite possession of a Canadian degree conferring definite advantage. For example 46% of Canada-born new accounting graduates had found work, compared to 44% of overseas-born local graduates resident 10 years or more, and 33% of those in Canada 5-10 years. Far lower employment outcomes, as in Australia, were secured by newly arrived young graduates who had qualified overseas, with just 6% of medical/medical science migrants gaining work in Canada in their field in the first 5 years, 8% of engineers, 9% of accountants, 22% of IT professionals, and 4-11% of those qualified in generic fields. The proportion engaged in work of any kind was also generally lower.

15. In Canada there appeared to be some labour market advantage for degree qualified migrants who had settled in the major immigrant-receiving location of Toronto, where substantial settlement services had developed.

The Census analysis confirmed Toronto to have attracted by far the largest number of degree qualified 1996-2001 arrivals to Canada: 381,232 migrants (compared to 141,245 in Vancouver, 112,234 in Montreal, 56,911 in Quebec City, 52,020 in Ottawa-Hull, and 257,816 in the rest of Canada). This was despite Vancouver receiving a disproportionate share of skilled migrants in terms of overall population.

For virtually every Census Metropolitan Area (CMA) examined, the best labour market integration rates were secured by the birthplace groups favoured by Canadian employers when selecting temporary foreign workers: degree-qualified arrivals from the UK/Ireland (eg 68% employed in Toronto in a professional or managerial position by 2001), the USA (69%), South Africa (70%), Australia/New Zealand (69%), and North West Europe (69%).

The following recently arrived degree-qualified groups were the least likely to have found any form of work in the examined CMA locations, with poor employment outcomes striking for select birthplace groups in Quebec City and Vancouver (compared to far lower levels of risk in Toronto and the ‘rest of Canada’). It is worth reprising in relation to this the numerical dominance of some of these groups, with those listed below among the least likely to be employed in the stated CMA, yet among the top 4 1996-2001 arrival groups for each location:

Toronto: Other South and Central Asia (42,264) and China (33,887)
Vancouver: China (17,857), Taiwan (11,554)
Quebec City: Eastern Europe (364)

Identical trends were found in terms of the impact of location on employment outcomes in Australia, where 47,985 degree-qualified migrants had settled in Sydney from 1996-2001, compared to 24,241 in Melbourne, and 34,572 across all other urban and regional sites. As in Canada, for virtually every CMA examined the best labour market integration rates were secured by those birthplace groups favoured by Australian employers when selecting temporary foreign workers: degree-qualified arrivals from the UK/Ireland, the USA, South Africa, Australia/New Zealand, and North West Europe. In line with Australia’s selection policy, English speaking background migrants were strongly represented in the top 4 birthplace groups for skilled arrivals in each location:

Sydney: UK/Ireland (8,077), China (5,579), India (5,418) and Other South & Central Asia (3,172)
Melbourne: India (3,635), UK/Ireland (3,019), China (2,235) and USA/Canada (1,332)
Rest of Australia: UK/Ireland (6,767), USA/Canada (2,960), South Africa (2,443) and North West Europe (2,432)

C. The impact of Canadian and Australian economic selection systems

As is clear from this Census data analysis, Canada and Australia represented highly comparable settlement sites by 2001 for degree-qualified migrants selected under all immigration categories (economic, family and humanitarian). Within the first five years of arrival, as noted, 65% of 1996-2001 arrivals had gained work of some kind in Canada compared to 66% in Australia. The modest Australian ‘advantage’ was entirely due to the proportion of English-speaking background migrants selected (28% of degree-qualified arrivals by 1996-2001 compared to just 5% in Canada) — the research evidence confirming their far more immediate acceptability to employers.

The Census thus provided control data for the study, which assessed employment outcomes for economic category Principal Applicants only. Three policy questions were addressed in this analysis:

1. To what extent do Canada and Australia’s economic migration selection criteria deliver degree-qualified migrants with a capacity to integrate quickly into host country labour markets?
2. How effective have Australia’s post-1999 economic migration reforms been in terms of improving employment outcomes there since the mid 1990s?
3. How does the labour market experience of economic migrants in each country compare with the patterns established as the norm for otherwise comparable groups (derived from the 2001 Census data), entering through all immigration categories?

Two major and directly comparable categories of economic migrants were examined for each country for the purpose of this research: ‘Other Skilled Workers’ and ‘Assisted Relatives’ in Canada, compared to ‘Independents’ and ‘Concessional Family’ (later ‘Skilled Australia-Linked’, then ‘Australian-Sponsored’) migrants in Australia. Three longitudinal data sources were used:

• The Longitudinal Survey of Immigrants to Australia (LSIA): Based on recurrent interviews commencing 6 months post-arrival of a representative sample (5%) of migrants and refugees entering Australia, administered to date three times (LSIA 1 in 1993/5, LSIA 2 in 1999/2000 and LSIA 3 in October 2005);
• The Longitudinal Survey of Immigrants to Canada (LSIC): Based on recurrent interviews with a representative sample of migrants and refugees to Canada conducted 6 months, 2 years and 4 years post-arrival, LSIC 1 administered in 2000/01; and
• The Longitudinal Immigration Data Base Canada (IMDB): A semi-longitudinal database which allows tracking of individual migrants through the filing of tax returns from the first year of lodgement. (For the purpose of the current study analysis of data for recently arrived economic migrants who had lodged their first tax file in 1994-95 and in 1999-2000 [up to a year post-arrival] was used, facilitating comparison with LSIA 1 and LSIA 2 data for similar vintages of degree-qualified migrants.)

The research on employment patterns for recently arrived economic Principal Applicants in their first 6-12 months of residence in Canada and Australia ^{Footnote 1} — a far more immediate timeframe than accessible through 2001 Census data, and highly predictive of longer term outcomes (Birrell, Hawthorne & Richardson 2006). Key findings from this analysis are summarised in turn below.

1. Employment outcomes for economic migrants were comparable for Canada and Australia in the mid 1990s, but had greatly improved for only Australian economic PAs by 1999/2000.

In Australia, following the described change to migration selection criteria from 1999, Independent migrants’ work rates had surged from 57% in 1993-95 to 81% in 1999-2000, unemployment had dropped, and labour market participation rates had increased. Excellent employment gains were also evident for the Concessional Family category (rising from 46% to 69%). By contrast there appeared to have been a slight decline in employment rates for Other Skilled Workers in Canada (from 64% to 60%), with Assisted Relatives faring much worse — dropping from 57% in 1994/5 (IMDB approximation) to just 36% in the context of higher education levels but rapidly diversifying source countries.

Outcomes have continued to improve for PAs from both economic categories in Australia in the past 5 years: LSIA 3 data confirming 83% of Independent PAs to have found work within 6 months, compared with 72% of comparable Australian-Sponsored migrants by October 2005. 12% of economic PAs were unemployed, and just 5% were not in the labour force — a marked improvement over the 15% in LSIA 2.

2. Level of employment and job satisfaction had substantially improved for economic PAs in Australia since the mid 1990s, with select outcomes also improving in Canada.

In line with Australia’s generally superior employment outcomes, access to high skilled positions proved more common for Independent migrants in Australia, rising from 55% to 60% by 1999/2000. Concessional Family rates however remained comparable to those in Canada (reflecting these migrants’ lower points requirement) — just 32% of the employed located in high skilled positions by LSIA 2. In 1999/2000 85% of employed Independent PAs also stated they were actively using their credentials within 6 months, compared to just 64% of Concessional Family workers (a modest improvement over LSIA 1 findings).

The latest available Australian data (from LSIA 3) confirm 63% of all Independent arrivals to be using their qualifications, compared to 49% who were Australian-sponsored. This represents a positive outcome, given the figures now include very substantial numbers of recently graduated international students. (Comparable data were not available from Canada.) The data confirmed job satisfaction to be good but static for Concessional Family PAs in Australia, at around 45% for both LSIA 1 and 2. By contrast Independent PAs had become increasingly satisfied with their positions: 61% claiming to ‘love or really like’ their work, compared to 50% 6 years earlier.

LSIA 3 data confirmed this generally positive result, with 57% of Independent migrants liking their work and 31% considering it ‘OK’, compared to 50% and 37% respectively of Offshore Australian-Sponsored migrants, and 44% and 40% of former international students (the latter far more likely to be working in less skilled positions given their new graduate status).

It was not possible to categorise the type of positions secured by employed economic PAs in Canada for LSIA 1. IMDB data however suggested positive trends to have occurred from 1994/5 to 1999/2000: 32% of Other Skilled Workers in managerial or professional work in the mid 1990s, compared to 54% by the end of the decade. Growth in professional/managerial positions from 22% to 36% also occurred for Assisted Relatives.

The IMDB does not provide job satisfaction data. However LSIC 1 demonstrated a high 73% of Other Skilled Workers in Canada to be ‘positive’ regarding their employment by 2000/1, along with 70% of Assisted Relatives. Two years post-arrival, LSIC data confirmed 84% of those in work to be satisfied with their employment.

3. Salaries had risen markedly for economic Principal Applicants in Australia by 1999/2000 rather than in Canada.

In Australia weekly earnings had greatly improved for employed Independent PAs between LSIA 1 and LSIA 2, with 57% earning $A674 or more compared to just 39% 6 years earlier. By 1999/2000 76% of employed Independent migrants were earning above the median wage ($A154), along with 63% of Concessional Family migrants. Growth in income was also positive for Concessional Family workers, rising from 23% to 34% at this rate. While trend data for Canada were not available, earnings at the time of LSIC 1 were significantly less than in Australia: 33% of Other Skilled Workers paid $C618 per week or more, compared to just 11% of Assisted Relatives in current employment.

Australian data for LSIA 3 reveal income rates to have risen much further since 1999/2000. 65% of Independent Principal Applicants were earning over $A674 per week by October 2005, with their average weekly wage an extraordinary $A1,015. This compared to 46% of Off-Shore Australian-Sponsored PAs (on an average $A779 wage) and 40% of On-Shore former international students, who were new graduates lacking professional experience (averaging $A641 in terms of weekly earnings — an outcome likely to be remedied over time).

4. Australian labour market integration rates had improved for both male and female economic PAs, as well as across all age groups.

It is important to note that Australia’s changed selection criteria have neutralised female economic migrants’ disadvantage to a high degree — employment rates for female PAs rising from 49% to 71% between LSIA 1 and LSIA 2, with comparable male rates only slightly higher (from 53% to 78%). By contrast female economic PAs in Canada appeared to have experienced employment decline rather than gains across this period (assuming comparability between the LSIC and the IMDB). The 2000/01 work rate for migrant females 6 months post-arrival was 55% compared with 63% in 1994/5. For migrant males the comparable rates were 62% and 65%. This finding is problematic, given the increasing arrival of migrant women to Canada with relatively high qualifications.

Age-related policy changes have also been beneficial. In Australia by the late 1990s older migrants received no economic category points for age at all. This change is associated with clearly positive outcomes (older applicants being obliged to score very highly on other employment-related measures). By 1999/2001 77% of 25-44 year old economic PAs had secured work in Australia within 6 months, compared to 70% of 15-24 year olds and 59% of 45-64 year olds. These results for older workers are noteworthy, their work rate close to doubling within 6 years — reflecting their pre-migration filtering for credential recognition, English language and other employer-desired attributes which offset the disadvantage typically associated with greater age.

Canada continued to accept substantial numbers of economic PAs over 45 years of age from 1991-2003, despite some decline in this trend. However employers there strongly favoured young skilled workers across all three periods examined: 80% of 15-24 year olds employed by 2000-01 (compared to 79% in 1994), 61% of 25-44 year olds (compared to 65%), and 50% of 45-64 year olds (compared to 52%). This policy issue seems worth addressing.

5. Employment outcomes had also improved for economic PAs in Australia for all birthplace groups, including those typically disadvantaged.

As established by the Census data analysis, select ‘developing country’ birthplace groups are at risk of severe labour market disadvantage. The Australian data however confirmed tremendous gains in early employment outcomes for such migrants when filtered by changed selection criteria, with economic PAs proving far more acceptable to employers than their birthplace norm.

In 1993/95, for example, 85% of UK, 76% of South African and 73% of North West European economic migrants were employed within 6 months of arrival. This compared to 56% of economic migrants from India, 53% from Hong Kong / Malaysia / Singapore, 45% from China, 42% from the Middle East/North Africa and 31% from Eastern Europe. By the time of LSIA 2, this Australian ‘disadvantage gap’ had largely closed. 86% of UK, 89% of South African and 83% of North West European Principal Applicants were working within 6 months, compared to 73% from India, 72% from the Middle East/North Africa, 68% from Hong Kong / Malaysia / Singapore, 61% from China, and 79% from Eastern Europe (the improved outcomes for East European arrivals particularly notable, see Table 1).

In Canada improved labour market integration rates were also achieved by UK economic PAs (87% rising to 89% based on IMDB data), Filipinos (78% rising to 84%), Indians (72% rising to 76%), and Pakistanis (57% rising to 66%). However far more variable labour market integration rates remained the norm for other economic PA groups in 2000/01, with 47% from Hong Kong / Malaysia / Singapore employed (compared to 39% in the mid 1990s), and 48% from China (apparently declining from 73%).

6. Externally validated host country language screening was associated with greatly improved work outcomes.

Facility in the host country language/s represents a critical determinant of employment outcomes, as established by this and earlier reports. The LSIA showed a marked recent increase in the proportion of economic PAs using English well or very well in Australia (45% for LSIA 1 compared to 73% for LSIA 2). The employment gulf between migrants with high and low level English was also vast by 1999/2000: 73% of the former securing work compared to just 41% of the latter. The LSIA further allows us to assess English language ability for recent migrants by gender. Female economic PAs selected by Australia had matched or exceeded comparable males in terms of host country language ability — 93-98% of female Independent PAs speaking English well or very well, despite somewhat lower rates in the Concessional Family category. This seems certain to have contributed to females’ positive employment outcomes.

Australia’s 2005-06 skilled migration review confirmed the critical association between English language ability and migrants’ speed of labour market integration. 89% of economic PAs with English as ‘only or best’ language were employed 6 months post-arrival, compared to 86% knowing English ‘very well’, 76% ‘well’ and 59% ‘not well’. Principal Applicants with best command of English were also the most likely to be using their professional qualifications in work (61%, compared to 60%, 44% and 37%), as well as to be employed in professional or managerial positions.

The LSIC data confirmed migrants with knowledge of English and/or French to be highly advantaged in terms of access to employment. Given lack of language levels for LSIC and IMDB data however, it was not possible to make more meaningful comparisons in relation to this over time.

7. Improved work outcomes for economic migrants were found in major migration settlement locations for both countries.

In line with the 2001 Census data, employment outcomes for economic PAs varied across the Australian locations analysed: by 1999/2000 81% securing work in Sydney, compared to 73% in ‘Other Australia’ and 70% in Melbourne, and with markedly better outcomes in each location than for comparable groups in the LSIA 1. Similarly, analysis of the longitudinal data confirmed economic PAs located in Toronto (65%) and ‘Other Canada’ (69%) CMAs to have better immediate employment outcomes than migrants settling in Vancouver (56%) or Montreal (just 38%).)

8. Welfare dependence had also declined among economic PAs in both countries.

Welfare dependence had virtually disappeared in Australia for Independent and Concessional Family migrants by 1999/2000, in line with government policy barring access to it in the first 2 years post-arrival. While 12% of economic PAs were categorised as ‘unemployed’ at six months in Australia in LSIA 3, they would have been ineligible to receive benefits. In Canada reliance was also very low for Other Skilled Worker and Assisted Relative PAs in 2000-01 (just 8% of households), confirming it to be the non-economic categories most at risk of welfare dependence.

D. Conclusion

As established by the Census data analysis, Canada and Australia represent highly comparable settlement sites for degree-qualified migrants (all immigration categories). Economic migrants however perform indisputably better in Australia post-arrival — their immediate work outcomes strongly correlated to longer-term labour market integration rates (Birrell, Hawthorne & Richardson 2006). Far greater proportions of new arrivals in Australia now than in Canada secure positions fast, access professional or managerial status, earn high salaries, and use their credentials in work. In the process unprecedented numbers are avoiding the labour market displacement typically associated with select birthplace, language, age and gender-related groups. The latest available data (released May 2006) confirm the benefits of this policy refinement to be dynamic.

In redesigning its economic selection criteria, the Australian government from 1999 affirmed the program’s original intent — to select skilled migrants able to make an immediate contribution to the economy through use of their skills at an appropriate place in the labour market. A parallel goal was to reduce skills wastage among recent arrivals, together with the level of government investment required to support migrants’ labour market adjustment needs (by the mid 1990s some $A250 million of Federal funding per year for employment, credential recognition and English language bridging programs — even this level of resourcing proving inadequate).

Australia’s transformation of its economic migration program was viewed as legitimate and essential in a context where governments frame policy, but employers retain the power to offer or withhold work. The human capital model of selection had proven flawed — delivering Principal Applicants lacking the ‘knowledge economy’ attributes employers sought (sophisticated English language ability, recognised credentials, and qualification in fields associated with buoyant labour market demand).

Since 1999, in consequence of the research findings, perceived ‘employability’ has determined Principal Applicants’ capacity to proceed with skilled migration to Australia. In terms of credential recognition, economic PAs qualified in regulated fields have been required to apply for pre-migration screening by the relevant Australian national or state licensing bodies (typically a three month postal process) — a strategy designed to avoid years of forced labour market displacement due to non-recognition of skills. Given the existence of niche economies, priority processing and up to 20 bonus points have been awarded to applicants qualified in high-demand fields, a measure associated with clearly beneficial outcomes. Recognising the importance of host country language ability, candidates have been required to achieve ‘vocational’ or higher level scores on the independently administered International English Language Testing System (or approved equivalent), administered globally and monthly by the British Council for a modest fee. The level set has not been draconian — the minimum standard for economic eligibility defined as ‘Has partial command of the language, coping with overall meaning in most situations, though is likely to make many mistakes. Should be able to handle basic communication in own field’.

In terms of overall program impacts, it is essential to note that these 1999+ policy changes have not discouraged or distorted skilled flows to Australia. Economic intakes rose to 97,500 in 2005-06, from 77,800 in 2004-05 and a third that level in the mid 1990s. Racial and ethnic diversity have been maintained, including major flows from India and China. The latest available data show nine of the top ten sources for economic applicants remain in Asia —the rank order being India (19%), China (18%), the UK (16%), Malaysia (4%), the Philippines (3%), Indonesia (3%), Hong Kong (3%), Korea (3%), Sri Lanka (3%) and Singapore (2%). Rather than diminishing ethnic diversity, changed selection criteria have resulted in applicants being far more effectively screened.

International students had become strong economic program participants by 2005 (52% of the total). By definition such migrants had self-funded to meet local employers’ English language and credential needs, simultaneously supporting the development of Australia’s export education industry. Their selection was not viewed as ethically problematic, despite international debate concerning this issue. Parents rather than source countries had resourced these students’ tertiary education, with scholarship holders systematically excluded from selection. From an ethical perspective the recruitment of mature-age professionals who had been fully trained offshore could seem less defensible — despite this being the recruitment norm for Canada and Australia across many years.

In terms of Australia’s changed selection framework, it is finally important to affirm the positive employment gains achieved by traditionally disadvantaged groups (see Table 1). While labour market integration for all source countries had improved by 1999/2000, in the case of economic Principal Applicants from Eastern Europe, the Middle East/North Africa, India, the Philippines and China, the scale of this improvement had been dramatic. For example 79% of economic Principal Applicants from East Europe had found work within 6 months of arrival by 1999/2000, compared to 31% in 1993-95. The comparable rate for the Philippines was 76% (versus 57%), such gains further improving by the time of collection of LSIA 3 data (October 2005). Impressive employment rates had simultaneously been achieved by other at-risk groups, including women of all ages, and older skilled migrants.

Should Canada wish to improve labour market outcomes for the economic category in the period ahead, a more radical overhaul of economic selection criteria seems warranted than the gentle fine-tuning associated with the 2002 Immigration and Refugee Protection Act, from which benefits remain in doubt (Sweetman 2006). The stakes are high for both economic migrants and Canada. As established by a recent study, ‘If immigrants (escape) low income in their first year, the likelihood of entry in subsequent years (falls) dramatically to below 10% (2007: 5). However,

… (B)y the early 2000s, skilled class entering immigrants (to Canada) were actually more likely to enter low-income and be in chronic low-income than their family class counterparts, and the small advantage that the university educated entering immigrants had over, say, the high school educated in the early 1990s had largely disappeared by 2000, as the number of highly educated rose. What did change was the face of the chronically poor immigrant; by the late 1990s one-half were in the skilled economic class, and 41% had degrees (up from 13% in the early 1990s) (Pictor, Hou & Coulombe in 2007: 5-6).

Governments construct and promote economic migration programs to address national development goals. While substantial numbers of migrants at risk of delayed or de-skilled employment are certain to continue arriving in Canada as spouses, and within the family and humanitarian categories, it is legitimate to change selection criteria to secure more beneficial immediate as well as long-term economic outcomes.

Table 1: Impact of improved economic migration screening (1999+) on employment outcomes in 6 months for select birthplace groups, Australia (1993-1995 and 1999-2000)

Source: Longitudinal Survey on Immigrants to Australia