Analytical Studies Branch Research Paper Series
Labour Market Outcomes Among Refugees to Canada

Warning 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.

by Garnett Picot, Yan Zhang and Feng Hou
11F0019M No. 419
Release date: March 11, 2019

Abstract

Canada welcomed over 830,000 refugees from the 1980s to 2000s. However, their economic outcomes, especially the variation among major refugee groups, have not been examined comprehensively. Using the Longitudinal Immigration Database, this paper examines the labour market outcomes of refugees from 13 source countries with large inflows to Canada over the 1980-to-2009 period. The analysis first compares employment rates and earnings among refugees from the 13 source countries. It further compares each refugee group with economic-class and family-class immigrants who arrived during the same period. The results reveal a very large variation in employment rates and average earnings among the 13 refugee groups. Groups with low employment rates tended to have low earnings levels among the employed. Groups with low (high) employment rates and earnings among the men also tended to have low (high) rates among the women. Very little of the variation in earnings among refugee groups could be accounted for by differences in observable human capital characteristics, economic conditions or the program of entry to Canada. Privately sponsored refugees earned more than comparable government assisted refugees during the initial years in Canada. However, this advantage disappeared after a decade in the country.

Executive summary

Although not selected for economic reasons, refugees must survive economically in the receiving country and can make a significant economic contribution. Hence, their labour market outcomes are of interest. However, relatively few studies of the economic outcomes of refugees have been undertaken in either Canada or the United States. In particular, virtually no studies have been conducted on the variation in the economic outcomes of refugees from a range of source countries. This is the topic of this paper.

The focus is on the labour market outcomes of refugees from the 13 countries (or groups of countries) with the largest inflows to Canada over the 1980-to-2009 period. These groups include refugees from Afghanistan, China, Colombia, El Salvador, Ethiopia, the former Yugoslavia (as one group), Iran, Iraq, Pakistan, Poland, Somalia, Sri Lanka, and Viet Nam, Cambodia and Laos (the three countries as one group). In addition to determining the economic outcomes of refugees entering Canada during the three decades leading up to 2009, this analytical framework will be useful when data for more recent refugees (e.g., from Syria) become available.

With some exceptions, employment rates after five years in Canada were substantial among refugees. Male refugees from 7 of the 13 countries had employment rates over 75% five years after entry. However, those from Iran and Somalia had very low employment rates. Female refugees from Iraq, Afghanistan, Pakistan and Somalia also displayed very low employment rates.

Among the employed, the earnings of refugees from different countries varied greatly. Ten years after entering Canada, the refugee groups with the highest earnings (i.e., from the former Yugoslavia, Poland and Colombia) earned roughly double what those with the lowest earnings did (i.e., from Somalia, Afghanistan, Pakistan and China). Differences in outcomes were accentuated by the fact that groups with low (high) employment rates tended to have low (high) earnings levels among the employed, and groups with high employment rates tended to have high earnings among the employed. Furthermore, male and female refugees from the same country tended to have similarly low (high) relative earnings. These high correlations in men’s and women’s earnings across groups tended to exacerbate any poverty issues for refugees with low earnings.

Differences among the groups in observable human capital characteristics (notably educational attainment, age and knowledge of an official language), as well as in economic conditions, years in Canada and the program of entry to Canada, explained either none or very little of the much lower than average earnings of the refugee groups with the lowest earnings. Other unknown and unobserved factors accounted for the very low earnings. One possible explanation is that the education that these groups received in their home countries may have been of lower quality or have been perceived to be of lower quality by employers, thereby affecting the wages paid. The paper includes some evidence to support this notion. Other unobserved factors that may have contributed to the earnings differences among the groups could include

Refugees’ economic outcomes are often compared with those of family-class immigrants—the group they most closely resemble. Immediately after entry, refugees from all countries earned less than family-class immigrants. However, the five refugee groups with the highest earnings (i.e., from the former Yugoslavia; Poland; Colombia; Viet Nam, Cambodia and Laos; and El Salvador) displayed much more rapid earnings growth than did the family- and economic-class immigrants entering during the same period. After 15 years, they earned from 80% to 110% of what family-class immigrants did. The story was very different for refugees in the five lowest earnings groups (i.e., from Iraq, Somalia, Afghanistan, Pakistan and China). Even after 15 years in Canada, they earned from 50% to 70% of what family-class immigrants did. For these groups, observable differences in educational attainment, language ability, economic conditions and the other observed factors mentioned above accounted for very little of the earnings gap with family-class immigrants. Once again, unobserved and unknown factors played a major role.

Finally, the program under which refugees entered Canada seemed to matter, at least in the initial years. Even after differences in observable human capital (e.g., education and language), source country, and economic conditions were accounted for, privately sponsored refugees earned more than government-assisted refugees in the first year in Canada. While differences in human capital factors did not account for the initial earnings gap, they did account for most of the earnings gap 10 years after immigration. This suggests that the program of entry has little effect on earnings after a decade.

1 Introduction

Refugees to Canada are not admitted for economic reasons. Rather, they are people who have fled their home country because of “a well-founded fear of persecution based on race, religion, nationality, political opinion or membership in a particular social group, as well as those at risk of torture or cruel and unusual treatment or punishment” (Immigration and Refugee Protection Act 2001). Nonetheless, they must survive economically in the receiving country in the long run, and they can make a significant economic contribution to that country. Hence, their labour market outcomes are of interest. However, relatively few studies of the economic outcomes of refugees have been undertaken in either Canada or the United States. In particular, virtually no studies have been conducted on the variation in the economic outcomes of refugees from a range of source countries.

Refugees from different source regions experience different economic outcomes after arrival for a number of reasons. Human capital characteristics—notably education and labour market experience—may differ. The quality—actual or perceived—of the human capital and labour market experience acquired in different source countries may vary. Different refugee groups face different types of humanitarian crises in their home countries. Some flee their home countries and stay in refugee camps before arriving in Canada; others come to Canada directly. Some groups are resettled directly by government agencies, while some are sponsored (at least in part) by private organizations and individuals (Painter and Kokallaj 2017). Depending on their racial background and their national socioeconomic condition at the time of their arrival, refugee groups may be accommodated differently when they start their resettlement in Canada. Theory suggests that the context of the exit from their home country and the reception encountered in the receiving country, together with a group’s human capital resources, likely affect the path and pace of integration among immigrants and refugees (Alba and Nee 2003; Portes and Zhou 1993). Cultural differences regarding the role played by different family members in the labour market may also be important.

As humanitarian crises rise and recede around the world, the countries from which refugees flee shift. And while refugees enter Canada from many countries at any given time, some countries are numerically dominant with respect to refugee flows. This changes over time. During the 1980s, refugee flows to Canada were dominated by Viet Nam, Poland and El Salvador. During the early 1990s, Somalia, Sri Lanka and Iraq were the major source countries for refugees to Canada. By the mid-1990s, the major sources shifted to Sri Lanka and the former Yugoslavia. In the late 1990s and early 2000s, these two countries were joined by Afghanistan. Colombia, China, Afghanistan, Pakistan and Sri Lanka were the largest refugee source countries in the mid-2000s, and, by the late 2000s, the largest groups of refugees came from Colombia, Iraq and Sri Lanka. In the early 2010s, Iraq and Haiti were the largest source countries of refugees.

Over the 30 years from 1980 to 2009—the period of interest in this study—Canada accepted 831,000 refugees, an average of about 28,000 per year. Of course, the number varied greatly annually, ranging from a low of 14,000 in 1983 to a high of 54,000 in 1991 (Chart 1). In general, fewer refugees entered during the 1980s, with an average of about 22,000 per year, and more entered in the 1990s, with an average of 33,000 per year. The 2000s saw an average of about 28,000 per year.

Chart 1

Data table for Chart 1 
Data table for Chart 1
Table summary
This table displays the results of Data table for Chart 1. The information is grouped by Landing year (appearing as row headers), Refugees (appearing as column headers).
Landing year Refugees
number
1980 40,349
1981 14,980
1982 16,930
1983 13,969
1984 15,361
1985 16,772
1986 19,204
1987 21,470
1988 26,765
1989 36,869
1990 40,215
1991 54,069
1992 52,338
1993 30,590
1994 20,433
1995 28,092
1996 28,472
1997 24,307
1998 22,842
1999 24,398
2000 30,091
2001 27,915
2002 25,110
2003 25,982
2004 32,686
2005 35,774
2006 32,499
2007 27,954
2008 21,859
2009 22,849
2010 24,697
2011 27,873
2012 23,079
2013 23,831
2014 23,286

This paper focuses on the labour market outcomes of refugees from 13 countries (or groups) with large inflows to Canada over the 1980-to-2009 period. These groups include refugees from Afghanistan, China, Colombia, El Salvador, Ethiopia, the former Yugoslavia (as one group), Iran, Iraq, Pakistan, Poland, Somalia, Sri Lanka, and Viet Nam, Cambodia and Laos (the three countries as one group). In addition to determining the economic outcomes of refugees entering Canada during the three decades leading up to 2009, this analytical framework will be useful when data for more recent refugees (e.g., from Syria) become available.

Cohorts of entering refugees from these countries were identified, and their outcomes were tracked for up to 15 years after entering Canada. This study first addresses the variation in earnings and employment rates among refugees from the 13 countries, and then examines whether human capital characteristics, program of entry or local labour market conditions can explain such variation. It further compares the earnings of each refugee group with those of the family- and economic-class immigrants who arrived during the same period.

The paper also addresses the outcomes of three different streams of refugees: government-assisted refugees, refugees with private sponsorship and refugees in the In-Canada Asylum Program. Some previous studies have shown considerable variation in socioeconomic outcomes among refugees depending on the entry program, at least in the initial years after arriving in Canada (Beiser 2003; IRCC 2016). However, these studies did not carefully examine how the differences in labour market outcomes by program type change as refugees stay longer in Canada.

2 Previous research

A few Canadian studies have focused on the earnings and employment outcomes of refugees. Comparison with family-class immigrants is the most common benchmark for refugees’ earnings. Different researchers have reached different conclusions. Some have found that refugees earned less than family-class immigrants, including Aydemir (2011) and, for male refugees, Abbott and Beach (2011). Some have concluded that refugees earned more, including Bevelander and Pendakur (2012) and, for female refugees, Abbott and Beach (2011). DeVoritz and Pivnenko (2004) found that the two groups earned about the same. However, all the studies compared refugees as a group with family-class immigrants as a group. None focused on refugees from particular countries.

A number of studies, notably a study by Abbott and Beach (2011), found that although refugees often had lower earnings than economic- or family-class immigrants during the first few years in Canada, they consistently had the highest earnings growth over the first decade in Canada. Earnings converged considerably among the immigrant classes over the first decade or more in Canada. This result was also observed in papers by De Silva (1997), Li (2003), and Abbott and Beach (2008).

Poverty among some refugees, particularly those with poor employment outcomes, has been seen as an issue. DeVoritz and Pivnenko (2004) concluded that economic poverty was endemic and a growing problem for refugees. Whether refugees fared well in Canada ultimately depended on their employment prospects. Only 52% of the refugees in their sample aged 20 to 64 found employment. Thus, the refugee population’s earnings performance was bimodal.

A paper by Bevelander and Pendakur (2012) is one of the few that examined the outcomes of refugees from different countries. The authors concluded that labour market outcomes were significantly different among refugees from the four countries that they studied. In a recent government report, Painter and Kokallaj (2017) provided a high-level statistical overview of the refugee groups from the 17 countries with the highest number of former refugees in 2011, according to the 2011 National Household Survey. They compared the demographic characteristics (i.e., average age and refugee categories) and socioeconomic outcomes of these refugee groups in terms of low income, educational level, employment and unemployment, and occupational skill level. The purpose of their report was to present a snapshot, descriptive analysis with comprehensive indicators. No attempt was made to take into account differences in length of residence in Canada among these groups or the extent to which differences in sociodemographic characteristics may account for differences in labour market outcomes among these groups.

Some other studies focused on one particular refugee group. Pendakur (2017), concentrating on immigrants—mainly refugees—from Afghanistan, concluded that refugees experienced poor labour force outcomes compared with other immigrants. Hou (2017) found that, in spite of very low levels of human capital, Vietnamese refugees to Canada registered very high employment rates and over time closed the initial earnings gap with other immigrants. He also found that their children had better education and earnings outcomes than those of other immigrants and the Canadian-born population.

Focusing on business ownership by immigrants in 2010, Green et al. (2016) found that self-employment tended to be marginally higher among refugees than among economic- and family-class immigrants and the Canadian-born population. However, ownership of a private incorporated company—the second type of business included in the study—was less likely among refugees. Picot and Ostrovsky (2017) concluded that the main difference between refugees and economic-class immigrants was not their tendency to be business owners but rather the type of company they owned. Economic-class immigrants tended to own companies in “knowledge-based” industries, whereas refugees tended to own companies in more traditional immigrant sectors such as food and accommodation, retail trade, taxis, janitorial work, and landscaping. This difference was likely related, at least in part, to differences in education background.

For the United States, relatively few studies of the earnings and employment outcomes of refugees during the first few years in the country are available, largely because of data limitations.

O’Connor (2010) found poorer employment, occupation and earnings outcomes for refugees, compared with other immigrants. This was the case for both the unadjusted (actual) and adjusted results, although the differences were minimized in the latter case. Following outcomes over a 20-year period, Evans and Fitzgerald (2017) found that refugees who entered the United States had much lower levels of education and poorer language skills than the native-born population, and they initially experienced low employment, high welfare use and low earnings. Outcomes improved considerably as refugees spent more time in the United States.

3 Data and methods

The Longitudinal Immigration Database (IMDB) is the principal data source for this research. It is a longitudinal microdata file of all immigrants (including refugees) who have landed in Canada since 1980 and have filed at least one tax return. The objective of this paper is to compare refugees from various major source countries who gained permanent resident status in Canada from 1980 to 2009. The following criteria were used to select major source countries for this study. If a country was the largest source of refugees for at least one year or if the number of refugees from that country exceeded 4,000 in total over three adjacent landing years, then the country was included in the analysis. This resulted in 13 countries as listed earlier.

For each group, the study included only refugees who became landed immigrants in the main arrival period (cohort), which is defined as all adjacent years when the number of refugees exceeded 1,000. The selected refugee cohorts encompassed from 4 to 15 entry years, depending on the source country (Table 1). The number of refugees from these source countries ranged from 9,500 to 94,800. For the purpose of studying the labour market outcomes of refugees, this study further restricted the sample to those who landed at age 20 to 49 and filed a tax return at least one year since landing. This resulted in sample sizes ranging from 6,965 to 57,185 for the selected countries.


Table 1
Number of refugees from major source countries
Table summary
This table displays the results of Number of refugees from major source countries. The information is grouped by Country of birth and main
arrival periods (appearing as row headers), Landed in main arrival period , Landed in
1980 to 2009, Number in the selected period as percent of full period, Landed at
age 20 to 49 in selected period and Landed at age 20
to 49 in selected
period and filed taxes at least once, calculated using number , percent and number units of measure (appearing as column headers).
Country of birth and main
arrival periods
Landed in main arrival period Landed in
1980 to 2009
Number in the selected period as percent of full period Landed at
age 20 to 49 in selected period
Landed at age 20
to 49 in selected
period and filed taxes at least once
number percent number
Viet Nam, Cambodia and Laos
1980 to 1994 94,830 95,820 99.0 57,210 57,185
Poland
1981 to 1992 66,020 67,845 97.3 52,130 52,105
Sri Lanka
1992 to 2009 43,275 45,820 94.4 28,285 28,195
Former Yugoslavia
1993 to 2001 40,090 41,555 96.5 24,560 24,515
Afghanistan
1995 to 2008 28,160 36,095 78.0 14,910 14,890
Iran
1988 to 1992; 1995 to 2004 22,005 30,805 71.4 14,325 14,280
Iraq
1992 to 1997; 2000 to 2001; 2008 to 2009 16,685 25,860 64.5 10,535 10,515
El Salvador
1983 to 1993 22,940 25,315 90.6 12,655 12,640
Colombia
2001 to 2009 23,440 24,635 95.1 14,945 14,900
Ethiopia
1988 to 1993 9,540 23,800 40.1 7,880 7,875
Somalia
1991 to 1995; 1998 to 2000 11,725 20,815 56.3 7,010 6,965
Pakistan
2000 to 2007 13,115 19,820 66.2 7,490 7,465
China
2002 to 2007 10,850 18,550 58.5 7,405 7,380
Total number 402,675 476,735 84.5 259,340 258,910

With regard to earnings, the focus is on annual earnings at 1, 5, 10 and 15 years after obtaining permanent residence in Canada. The average annual earnings of refugees from a particular country were compared with the earnings of immigrants in three classes: (1) all the refugees from the 13 countries included in the study, (2) family-class immigrants entering during exactly the same years as the refugee cohort and (3) economic-class immigrants entering during the same period. Average annual earnings were produced for refugees as well as for family-class and economic-class immigrants for up to 15 years after they entered Canada. In addition, unadjusted and adjusted estimates of earnings during the first 15 years in Canada were produced for each selected refugee group. The adjusted estimates control for differences in observable characteristics and economic conditions and are based on the linear regression models described below. Two different models were run to facilitate the earnings comparison—first among refugees from the 13 different countries, and second between refugees from one of the 13 countries and either family- or economic-class immigrants entering during the same period.

For the first comparison, the sample consisted of refugees aged 20 to 49 entering from 13 countries. To prevent any single group from dominating the model estimates, weights were assigned so that each group contributed the same weighted number of observations (refugees) to the model.Note The dependent variable is the annual earnings level for each individual. The independent variables include

Model 1 includes only the constant term, the dummies for source country, years since landing, years since landing squared and years since landing interacted with source country. This model was used to estimate unadjusted earnings.

Model 2 adds the human capital characteristics, program type and regional economic conditions to Model 1. This model was used to estimate adjusted earnings, assuming each refugee group had the same values in the control variables. The models were run separately for men and women.

For each refugee group, two different earnings gaps (differences) were estimated: the unadjusted gap, which is the difference in average unadjusted earnings between the refugees from a particular country and the refugees from all 13 countries; and the adjusted gap, which is produced in the same way but using adjusted earnings. For any particular refugee group, the difference between the adjusted and unadjusted gaps is an estimate of the extent to which the unadjusted gap in earnings is accounted for by differences in the observable characteristics included in the regression.

The second set of comparisons was between each refugee (source country) group and family-class immigrants entering during the same years as the refugee group. As well, a similar comparison was made between each refugee group and economic-class immigrants entering during the same years. The population was restricted to those aged 20 to 49 at entry. The dependent variable was annual earnings for individual i MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqiVu0Je9sqqrpepC0xbbL8F4rqqrFfpeea0xe9Lq=Jc9 vqaqpepm0xbba9pwe9Q8fs0=yqaqpepae9pg0FirpepeKkFr0xfr=x fr=xb9adbaqaaeGaciGaaiaabeqaamaabaabaaGcbaaeaaaaaaaaa8 qacaWGPbaaaa@3704@ . The independent variables include

Model A includes the constant term and dummies for immigrant class (i.e., the focal refugee group, economic- and family-class immigrants, and other refugees), years since landing and its squared term, and the interaction between immigrant class and years since landing and the squared term. This model was used to estimate unadjusted annual earnings for the four immigrant class groups in various years after immigration.

Model B adds in age at immigration, education, official language and regional economic conditions to Model A. This model was used to estimate adjusted annual earnings, which take into consideration differences among immigrant classes in the control variables.

The models were run separately for men and women and for each of the 13 selected refugee groups included in the analysis, for a total of 26 regressions. The sample for any particular regression consisted of all female or male immigrants (including refugees) who entered Canada during the entry years covered by the refugee group and who filed a tax return in any year in the observation period.

The gap in both unadjusted and adjusted earnings between, say, refugees entering from a particular country and family-class immigrants was calculated. The difference between the unadjusted and adjusted earnings gap is an estimate of the extent to which differences in the observable characteristics and economic conditions explain or account for the unadjusted earnings gap between the two groups.

In addition to earnings, the proportion of the refugee cohort employed (i.e., with positive earnings) is estimated for up to 10 years after entering Canada. It is simply the number of refugees who entered during year X MathType@MTEF@5@5@+= feaagKart1ev2aqatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqiFu0Je9sqqrpepC0xbbL8F4rqqrFfpeea0xe9Lq=Jc9 vqaqpepm0xbba9pwe9Q8fs0=yqaqpepae9pg0FirpepeKkFr0xfr=x fr=xb9adbaqaaeGaciGaaiaabeqaamaabaabaaGcbaGaamiwaaaa@36C4@ who are employed Y MathType@MTEF@5@5@+= feaagKart1ev2aqatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqiFu0Je9sqqrpepC0xbbL8F4rqqrFfpeea0xe9Lq=Jc9 vqaqpepm0xbba9pwe9Q8fs0=yqaqpepae9pg0FirpepeKkFr0xfr=x fr=xb9adbaqaaeGaciGaaiaabeqaamaabaabaaGcbaGaamywaaaa@36C5@ years later, divided by the total number of refugees from that country who entered during year X MathType@MTEF@5@5@+= feaagKart1ev2aqatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqiFu0Je9sqqrpepC0xbbL8F4rqqrFfpeea0xe9Lq=Jc9 vqaqpepm0xbba9pwe9Q8fs0=yqaqpepae9pg0FirpepeKkFr0xfr=x fr=xb9adbaqaaeGaciGaaiaabeqaamaabaabaaGcbaGaamiwaaaa@36C4@ . These results were estimated for refugees from a particular country who entered during any individual year, and the results were aggregated across all entry years included in the cohort to obtain estimates for a country’s entering cohort as a whole. The results were produced separately for men and women.

4 Results

4.1 The characteristics of refugees and other immigrants

Differences between refugee groups in earnings-related observable characteristics could explain some or all of the earnings gap between various groups. In particular, differences in educational attainment and language spoken at landing can have a significant effect. The characteristics of refugees from different countries were significantly different (Table 2). Quite low levels of education were found for refugees from Viet Nam, Cambodia and Laos; Sri Lanka; Ethiopia; El Salvador; Somalia; and China. Less than 10% had a university degree. Those from the former Yugoslavia, Poland, Colombia and Pakistan registered fairly high levels: 18% to 26% had a university degree. The variation in language was similarly large. Over 70% of refugees from the following countries spoke neither English nor French at landing: the former Yugoslavia; Poland; Viet Nam, Cambodia and Laos; and El Salvador. In contrast, those from Sri Lanka, Somalia and Pakistan had very high rates of speaking English in particular: over three-quarters spoke this language at landing. Only refugees from Viet Nam, Cambodia and Laos had both low levels of education and a low proportion of English or French speakers. However, these refugees did quite well economically.


Table 2
Characteristics of refugees from major source countries who became permanent residents in Canada
Table summary
This table displays the results of Characteristics of refugees from major source countries who became permanent residents in Canada Former Yugoslavia, Poland , Colombia , Viet Nam, Cambodia and Laos, El
Salvador, Sri Lanka , Iran , Ethiopia, Iraq , Somalia , Afghanistan, Pakistan, China and Overall , calculated using percent units of measure (appearing as column headers).
Former Yugoslavia Poland Colombia Viet Nam, Cambodia and Laos El
Salvador
Sri Lanka Iran Ethiopia Iraq Somalia Afghanistan Pakistan China Overall
percent
Gender
Male 52.0 61.4 48.3 59.6 55.4 61.4 60.5 60.5 61.2 48.9 50.7 60.2 51.3 57.7
Female 48.0 38.6 51.7 40.4 44.6 38.6 39.5 39.5 38.8 51.1 49.3 39.8 48.7 42.3
Age at landing
20 to 29 35.4 45.6 34.8 66.6 48.6 44.1 44.1 75.8 45.5 59.2 43.7 35.3 30.4 49.0
30 to 39 42.9 43.5 36.2 25.5 36.8 37.4 38.7 21.0 37.1 31.6 33.4 39.6 36.9 35.6
40 to 49 21.7 10.9 29.0 7.9 14.6 18.5 17.2 3.1 17.4 9.2 23.0 25.1 32.7 15.4
Education level at landing
Less than high school 10.7 3.5 18.4 61.2 39.9 11.5 13.2 19.5 35.1 24.6 32.3 22.4 40.9 26.6
High school 38.7 29.3 38.4 31.8 38.2 72.2 61.9 62.0 32.9 57.3 38.6 38.1 35.9 41.0
Some postsecondary education 32.4 47.3 17.4 4.9 16.3 12.2 13.8 13.1 15.5 9.3 13.5 13.2 17.1 20.5
University or above 18.1 19.9 25.8 2.0 5.6 4.1 11.1 5.4 16.5 8.7 15.6 26.2 6.1 11.9
Official language at landing
English only 25.7 19.4 48.5 12.9 14.5 85.1 50.1 52.1 33.7 76.8 30.6 77.3 54.1 35.3
French only 2.3 1.1 3.2 2.0 2.3 0.1 1.7 1.4 0.3 2.8 0.7 0.0 0.0 1.4
English and French 1.6 0.9 1.4 1.2 0.6 0.3 1.4 1.8 0.6 1.7 5.8 0.4 0.0 1.3
Neither English nor French 70.4 78.6 46.9 83.9 82.5 14.5 46.8 44.6 65.3 18.6 62.9 22.2 45.5 62.0
Refugee category
Government-assisted 80.1 28.7 38.2 56.4 81.6 0.4 43.0 37.6 47.1 11.5 49.4 2.0 0.9 40.7
Privately sponsored 14.7 71.0 4.0 43.6 7.5 1.0 8.7 56.6 43.1 7.9 35.8 0.3 0.0 32.3
Landed in Canada 4.8 0.2 52.6 0.0 10.5 86.5 44.0 5.7 9.2 79.4 12.9 74.9 68.3 23.4
Dependants 0.4 0.2 5.3 0.0 0.4 12.1 4.3 0.0 0.6 1.2 1.8 22.8 30.6 3.6

In general, the educational attainment of refugees is lower than that of family-class immigrants, and, particularly, than that of economic-class immigrants. This would tend to result in lower earnings among refugees than among the other groups. About 12% of the refugees from the 13 countries (combined) held a university degree upon entry to Canada, compared with 25% of family-class immigrants who entered Canada during the same years as the selected refugee groups and 51% of economic-class immigrants (table not shown). In addition, about 32% of the refugees had some form of postsecondary education, compared with 43% of family-class immigrants and 72% of economic-class immigrants.

The difference in the ability to speak English or French at entry between refugees and family- and economic-class immigrants was large. A much smaller share of refugees (38%) spoke English or French than did family-class (61%) or economic-class (78%) immigrants who entered during the same years.

4.2 The employment rate

The employment rate for a refugee group is defined as the number of refugees with positive earnings in any given year after entering Canada, divided by the number of refugees who arrived in that particular refugee cohort. For example, the employment rate among Polish refugees one year after entering Canada is the number from that country with positive earnings one year after entry divided by the size of the Polish entering cohort. An individual who is either employed as a paid worker or self-employed at any time during the year is included in the rate. With some exceptions, the employment rate for refugees from most countries increased with the number of years spent in Canada (Table 3). Five years after entering Canada, male refugees from most regions has a fairly high employment rate: the employment rate was 75% or higher for refugees from 7 of the 13 countries. Male refugees from Iran (at 56%) and Somalia (at 44%) were at the lower end.

The employment rate was lower among female refugees. Refugee women from four countries had employment rates at or over 65% five years after becoming permanent residents in Canada. Female refugees from five different countries had employment rates around 46% to 56%. Those from Iraq, Afghanistan, Pakistan and Somalia had the lowest employment rates five years after entering Canada. Only about one in five Somali women were working at any time during the year five years after entering Canada.


Table 3
Percentage of refugees with earnings, by years since landing, gender and major source country
Table summary
This table displays the results of Percentage of refugees with earnings. The information is grouped by Source country and gender (appearing as row headers), Years since landing, 1, 5 and 10, calculated using percent units of measure (appearing as column headers).
Source country and gender Years since landing
1 5 10
percent
Former Yugoslavia
Male 56.7 80.4 77.7
Female 38.9 72.1 73.2
Poland
Male 84.0 75.0 73.5
Female 69.8 64.5 68.3
Colombia
Male 64.1 76.5 70.1
Female 45.1 65.8 67.9
Viet Nam, Cambodia and Laos
Male 58.8 75.2 71.7
Female 43.9 55.7 53.8
El Salvador
Male 68.4 71.1 68.5
Female 47.1 51.4 57.2
Sri Lanka
Male 76.9 80.8 79.7
Female 36.5 53.6 58.5
Iran
Male 47.6 55.9 58.8
Female 28.6 47.5 54.3
Ethiopia
Male 69.2 62.5 61.8
Female 51.4 46.5 52.8
Iraq
Male 58.6 64.6 58.8
Female 30.0 37.9 38.3
Somalia
Male 45.2 44.0 41.5
Female 18.2 21.4 30.3
Afghanistan
Male 57.1 67.0 67.1
Female 27.2 39.1 42.8
Pakistan
Male 71.6 74.7 74.4
Female 31.4 37.9 35.4
China
Male 76.8 81.6 80.8
Female 66.3 73.0 74.3

The same general pattern of employment rates for both men and women was observed 10 years after entering Canada (Table 3).

Depending on their culture and financial needs, women in refugee groups with high male employment rates may choose not to work. Conversely, employment rates among women may be higher in groups where men are unable to find employment. That was not the case. The refugee groups with high employment rates among men tended to have high rates among women. Conversely, those with low male employment rates tended to have low female employment rates (Table 3). Across the 13 groups, the correlation coefficient between the employment rates of men and women five years after entering Canada was 0.8.Note To the extent that men and women marry within refugee groups, this high degree of correlation between the employment rates of men and women would exacerbate lower-income rates and economic inequality between refugee groups.

The group variation was larger in the female employment rates than in the male employment rates. Five years after entering Canada, men in the group with the highest employment rate (from the former Yugoslavia) saw a rate 1.8 times that of the group with the lowest rate (from Somalia). Among women, the rate between these same two groups varied by a factor of 3.4. The male employment rate had an average across the 13 groups of 70% (unweighted) and a standard deviation of 10; women had an average of 51% with a standard deviation of 15. This large variation in employment rates among female refugees from different countries is an important part of the overall variation in the economic outcomes of refugee families.

4.3 Earnings

4.3.1 Earnings differences among refugees from the 13 source countries

The unadjusted (i.e., raw) earnings data provide a number of interesting observations.

First, just as with the employment rate, the variation in earnings among the 13 refugee groups was tremendous. Ten years after entering Canada, employed male refugees had average annual earnings ranging from around $45,000 among the top three groups (i.e., from the former Yugoslavia, Poland and Colombia) to under $28,500 among the bottom four source-country groups (i.e., from Somalia, Afghanistan, Pakistan and China). The groups are ranked by the average earnings of male refugees 15 years after entering Canada in Table 4. Refugees in the top three earnings groups had average earnings 1.9 times those of their counterparts in the lowest three earnings groups (Table 4). Earnings were lower among female refugees. Five years after entry, female refugees had average earnings of $19,100, compared with $27,000 for males across all 13 groups. However, the variation was similar.


Table 4
Average earnings of refugees, family-class immigrants and economic-class immigrants aged 20 to 49 at landing, by years since landing, major source country and gender
Table summary
This table displays the results of Average earnings of refugees. The information is grouped by Source country and gender (appearing as row headers), Refugees , Family-class immigrants, Economic-class immigrants, Years since landing , Ratio (15th to 1st year), 1 , 5 , 10 and 15 , calculated using thousands of dollars units of measure (appearing as column headers).
Source country and gender Refugees Family-class immigrantsTable 4 Note 1 Economic-class immigrantsTable 4 Note 2
Years since landing Ratio (15th to 1st year) Years since landing Ratio (15th to 1st year) Years since landing Ratio (15th to 1st year)
1 5 10 15 1 5 10 15 1 5 10 15
thousands of dollars
Former Yugoslavia
Male 19.26 37.50 45.81 52.96 2.75 25.19 36.12 42.82 46.27 1.84 35.32 50.41 60.42 65.36 1.85
Female 11.49 24.74 33.34 40.42 3.52 16.60 23.14 28.82 33.11 1.99 20.79 30.81 39.48 43.55 2.09
Poland
Male 24.28 36.68 46.44 52.78 2.17 25.40 35.01 43.35 49.83 1.96 40.32 48.11 57.84 65.90 1.63
Female 13.86 23.80 32.37 38.37 2.77 17.40 23.26 29.19 34.18 1.96 22.25 28.81 35.06 39.79 1.79
ColombiaTable 4 Note 3
Male 23.10 33.62 44.69 51.41 2.23 28.68 39.39 45.98 48.73 1.70 35.41 51.34 62.61 67.81 1.91
Female 15.47 22.77 30.27 36.49 2.36 17.89 24.58 30.28 32.69 1.83 22.66 33.23 42.38 46.12 2.04
Viet Nam, Cambodia and Laos
Male 18.77 28.64 35.25 38.83 2.07 24.66 34.98 43.08 48.84 1.98 39.18 48.86 58.14 65.65 1.68
Female 14.14 20.86 24.36 26.53 1.88 16.98 23.06 28.78 33.71 1.99 21.74 28.64 34.96 39.92 1.84
El Salvador
Male 16.04 25.38 32.04 38.15 2.38 24.73 34.47 42.87 49.29 1.99 38.14 46.69 56.92 65.33 1.71
Female 10.49 17.31 22.70 26.82 2.56 17.07 23.14 28.96 34.03 1.99 21.94 28.86 35.24 40.31 1.84
Sri Lanka
Male 26.48 31.44 34.97 38.06 1.44 26.56 37.47 43.60 46.62 1.76 35.62 50.97 61.17 65.31 1.83
Female 14.78 18.53 21.86 25.08 1.70 17.20 23.85 29.23 33.22 1.93 21.94 32.05 40.08 43.20 1.97
Iran
Male 15.70 22.75 31.54 36.12 2.30 25.80 35.92 43.85 47.39 1.84 34.80 49.46 60.74 64.79 1.86
Female 11.35 18.98 26.12 31.18 2.75 17.06 23.56 29.56 33.68 1.97 21.29 31.27 39.92 42.81 2.01
Ethiopia
Male 17.30 22.11 29.59 33.38 1.93 23.48 32.80 42.38 47.69 2.03 35.93 44.67 56.35 63.86 1.78
Female 12.16 17.16 22.43 27.47 2.26 16.79 22.66 28.80 33.61 2.00 22.03 28.54 35.60 40.59 1.84
Iraq
Male 17.82 26.78 30.27 31.83 1.79 24.92 36.55 42.55 46.19 1.85 35.49 50.82 59.81 64.34 1.81
Female 12.43 17.79 23.36 25.48 2.05 16.61 23.34 28.57 32.94 1.98 21.86 31.32 38.56 41.83 1.91
Somalia
Male 15.67 20.16 27.05 31.06 1.98 23.75 34.61 42.22 46.67 1.97 36.14 49.65 60.05 65.90 1.82
Female 12.55 16.46 20.59 23.56 1.88 16.35 22.82 28.70 33.35 2.04 21.14 29.89 38.39 42.93 2.03
Afghanistan
Male 16.94 23.38 28.46 29.11 1.72 27.84 38.29 44.50 46.82 1.68 35.54 51.11 61.86 65.64 1.85
Female 11.00 16.77 20.79 23.33 2.12 17.57 24.14 29.73 33.49 1.91 21.91 32.47 41.20 44.49 2.03
PakistanTable 4 Note 3
Male 19.45 22.43 23.99 28.11 1.45 28.87 38.89 45.34 47.40 1.64 34.87 50.61 62.52 68.74 1.97
Female 12.59 16.27 17.97 20.02 1.59 17.86 24.41 30.24 33.82 1.89 21.82 32.83 42.44 47.47 2.18
ChinaTable 4 Note 3
Male 17.08 20.41 22.02 22.76 1.33 28.93 39.18 46.50 46.74 1.62 34.90 51.12 63.54 65.91 1.89
Female 15.32 17.71 19.43 19.75 1.29 17.91 24.69 30.49 31.63 1.77 21.98 33.19 42.51 44.93 2.04

Second, refugee groups that displayed high employment rates also tended to have higher earnings, and groups with low employment rates tended to have lower earnings. Male refugees from Somalia, Iran and Ethiopia had both low unemployment rates and low earnings among the employed, as did female refugees from Pakistan, Afghanistan, Iraq and Somalia (Chart 2). The one major exception was refugees from China; they had very low average earnings five years after entering Canada but the highest employment rate. They had a very high propensity to work but mainly in low paying jobs. Excluding Chinese refugees, the correlation between the employment rate and average earnings five years after entry across the 12 groups was 0.73 for men and 0.88 for women. These high correlations may be because the differences among refugees in human capital characteristics affect both the ability to find a job and the earnings in those jobs. However, as will be seen later, the difference among groups in human capital characteristics explained relatively little of the variation in earnings outcomes.

Chart 2

Data table for Chart 2 
Data table for Chart 2
Table summary
This table displays the results of Data table for Chart 2. The information is grouped by Gender, age at landing and source country (appearing as row headers), Five years after landing, Average earnings and Employment rate, calculated using thousands of dollars and percent units of measure (appearing as column headers).
Gender, age at landing and source country Five years after landing
Average earnings Employment rate
thousands of dollars percent
Men aged 20 to 49 at landing
Former Yugoslavia 37.5 80.4
Poland 36.7 75.0
Colombia 33.6 76.5
Viet Nam, Cambodia and Laos 28.6 75.2
El Salvador 25.4 71.1
Sri Lanka 31.4 80.8
Iran 22.8 55.9
Ethiopia 22.1 62.5
Iraq 26.8 64.6
Somalia 20.2 44.0
Afghanistan 23.4 67.0
Pakistan 22.4 74.7
China 20.4 81.6
Women aged 20 to 49 at landing
Former Yugoslavia 24.7 72.1
Poland 23.8 64.5
Colombia 22.8 65.8
Viet Nam, Cambodia and Laos 20.9 55.7
El Salvador 17.3 51.4
Sri Lanka 18.5 53.6
Iran 19.0 47.5
Ethiopia 17.2 46.5
Iraq 17.8 37.9
Somalia 16.5 21.4
Afghanistan 16.8 39.1
Pakistan 16.3 37.9
China 17.7 73.0

A straightforward way to assess the effect of these high correlations on group differences is to determine the employment earnings per person for each refugee group rather than the employment earnings per employed individual. The average earnings per person aged 20 to 64 are simply the employment rate times the average earnings. Given the average earnings per adult, male refugees from the top three countries—the former Yugoslavia, Poland and Colombia—had access to 2.3 times the economic resources from employment compared with those in the bottom three regions—Somalia, Afghanistan and Iran. Among women, this difference was three times greater.Note Notably, Somali refugees had access to extremely low levels of employment earnings, at $8,800 per adult man and $3,500 per adult woman 5 years after entering Canada, and $11,000 per man and $6,100 per woman 10 years after entering.

Third, just as with the employment rate, the earnings of male and female refugees from the same countries were highly correlated. The groups with low earnings among the men tended to have low earnings among the women. The correlation coefficient between men’s and women’s average earnings across the 13 groups was over 0.9 both 5 and 10 years after entry. There is little evidence that the earnings of women in any way offset the low earnings of men. This high correlation would tend to exacerbate poverty among refugee groups with low earnings. The same pattern was observed earlier with employment rates. The high correlation across groups between the economic outcomes of men and women would tend to concentrate high levels of employment and earnings in some groups and accentuate low levels in other groups.

Fourth, as is the case with virtually all immigrants, the annual earnings of employed refugees increase significantly with the number of years spent in Canada (Chart 3). The average earnings of employed refugees from some countries grew at a high rate. Male refugees from 6 of the 13 countries (i.e., the former Yugoslavia; Poland; Colombia; Viet Nam, Cambodia and Laos; El Salvador; and Iran) saw their earnings more than double during the first 15 years in Canada (Table 4). This rapid growth in annual earnings was not observed among family-class and economic-class male immigrants who arrived over the same period as the refugee groups; none saw their earnings more than double over the first 15 years in Canada. A similar story held for female refugees; those from 8 of the 13 countries saw their earnings more than double over the first 15 years, again something not observed among female family-class immigrants. This result was observed in earlier studies such as that by Abbott and Beach (2011). Of course, refugees from some countries did not see their earnings grow as fast as family- and economic-class immigrants entering at the same time. This was particularly true for refugees, both male and female, from Sri Lanka, Pakistan and China (Table 4).

Chart 3

Data table for Chart 3 
Data table for Chart 3
Table summary
This table displays the results of Data table for Chart 3. The information is grouped by Years since landing (appearing as row headers), Poland, Viet Nam, Iran, Iraq, Somalia and Pakistan, calculated using thousands of dollars units of measure (appearing as column headers).
Years since landing Poland Viet Nam Iran Iraq Somalia Pakistan
thousands of dollars
Men aged 20 to 49 at landing
1 24.3 18.8 15.7 17.8 15.6 19.5
2 29.4 23.6 18.2 21.5 16.9 20.9
3 32.4 25.3 19.7 23.6 17.8 21.7
4 34.5 27.1 21.1 25.3 19.1 22.1
5 36.7 28.6 22.7 26.8 20.2 22.4
6 38.9 30.0 24.4 26.8 22.0 23.1
7 41.3 31.4 27.3 27.4 23.5 23.8
8 43.1 33.4 28.0 28.4 25.4 24.1
9 44.9 34.7 29.7 29.4 26.7 24.2
10 46.4 35.3 31.5 30.3 27.0 24.0
11 48.2 35.7 32.2 30.7 27.4 24.3
12 49.3 36.4 33.7 30.9 29.4 25.2
13 50.6 37.6 35.2 31.0 29.7 25.9
14 51.5 38.7 35.0 31.3 30.2 28.1
15 52.8 38.8 36.1 31.6 31.1 Note ..: not available for a specific reference period
Women aged 20 to 49 at landing
1 13.9 14.1 11.4 12.4 12.5 12.6
2 17.6 17.4 13.2 14.6 13.7 14.1
3 19.8 18.8 15.3 15.9 15.3 15.0
4 22.0 19.8 17.0 16.9 15.7 15.8
5 23.8 20.9 19.0 17.8 16.5 16.3
6 25.7 21.5 20.5 19.8 17.3 17.1
7 27.5 22.3 21.9 21.5 17.6 17.1
8 29.3 23.3 23.2 22.3 18.5 17.5
9 30.5 24.2 25.1 23.2 19.9 17.9
10 32.4 24.4 26.1 23.4 20.6 18.0
11 33.9 24.5 27.0 24.2 21.2 17.6
12 35.2 25.1 27.9 24.1 22.4 18.2
13 36.2 25.7 29.4 24.2 22.2 20.5
14 37.2 26.3 30.0 25.1 23.1 20.0
15 38.4 26.5 31.2 25.4 23.6 Note ..: not available for a specific reference period

4.3.2 Why the large difference in earnings among refugees from the 13 source countries?

This study first shows how much the unadjusted average earnings of refugees from a particular country differ from the unadjusted average earnings of all refugees from the 13 countries. This gap varied substantially. Male refugees from Poland earned $13,200 more than the average for all refugees 10 years after entry, while those from China earned $11,300 less than the average (Table 5). The gap for female refugees varied from $9,600 more than the average to $6,200 less than the average. The most obvious reason for these earnings gaps is a difference in human capital characteristics. To determine the extent to which these differences “explain” the difference in earnings between a refugee group and the overall average, adjusted earnings were estimated using regression Model 2 described in the “Data and methods” section. The model takes into account

The difference between the unadjusted earnings gap of a refugee group relative to the overall average (based on Model 1) and the adjusted gap (based on predicted earnings from Model 2) indicates the extent to which the variables mentioned above explain the actual earnings gap (Table 5).


Table 5
Earnings, earnings gapsTable 5 Note 1 and percentage of gaps explained, by major source country, gender and years since landing, for refugees aged 20 to 49 at landing
Table summary
This table displays the results of Earnings. The information is grouped by Source country and gender (appearing as row headers), Unadjusted earnings, Adjusted earnings, Percentage of gap explained, Average predicted, Gap to be explained, Gap remaining, Years since landing, 1 , 5 , 10 , 1, 5 and 10, calculated using dollars and percent units of measure (appearing as column headers).
Source country and gender Unadjusted earningsTable 5 Note 2 Adjusted earningsTable 5 Note 3 Percentage of gap explainedTable 5 Note 4
Average predicted Gap to be explained Average predicted Gap remaining
Years since landing Years since landing Years since landing Years since landing Years since landing
1 5 10 1 5 10 1 5 10 1 5 10 1 5 10
dollars percent
Former Yugoslavia
Male 22,400 36,200 46,900 2,500 9,200 13,600 24,000 36,400 46,200 3,700 9,300 13,100 none none 3.7
Female 12,900 24,100 34,000 -500 5,000 9,600 14,100 24,500 33,500 700 5,400 9,000 240.0 none 6.3
Poland
Male 25,600 36,700 46,600 5,700 9,700 13,200 27,000 37,100 45,500 6,700 10,000 12,400 none none 6.1
Female 14,700 23,700 32,400 1,300 4,600 7,900 15,200 24,000 31,700 1,800 4,900 7,200 none none 8.9
ColombiaTable 5 Note 5
Male 23,400 33,700 44,900 3,500 6,700 11,500 22,200 33,100 44,400 1,900 6,000 11,300 45.7 10.4 1.7
Female 15,600 23,000 30,700 2,200 3,900 6,300 14,100 21,500 29,400 700 2,400 4,900 68.2 38.5 22.2
Viet Nam, Cambodia and Laos
Male 20,200 28,600 35,500 300 1,600 2,100 23,200 31,800 38,900 2,900 4,700 5,800 none none none
Female 15,400 20,600 24,700 2,000 1,400 200 17,600 23,800 28,700 4,200 4,700 4,200 none none none
El Salvador
Male 17,600 24,900 32,300 -2,300 -2,100 -1,000 21,300 28,800 35,700 900 1,700 2,600 139.1 181.0 360.0
Female 11,500 17,100 22,700 -1,900 -2,000 -1,800 14,100 20,000 25,600 700 900 1,100 136.8 145.0 161.1
Sri Lanka
Male 27,000 31,200 35,200 7,100 4,200 1,800 25,400 29,900 34,600 5,100 2,800 1,500 28.2 33.3 16.7
Female 15,000 18,500 22,100 1,600 -700 -2,400 13,400 16,700 20,800 0 -2,400 -3,700 100.0 none none
Iran
Male 15,800 23,100 30,600 -4,100 -3,900 -2,700 16,400 23,400 30,400 -3,900 -3,700 -2,600 4.9 5.1 3.7
Female 11,300 18,800 26,000 -2,100 -300 1,600 10,900 18,200 25,400 -2,500 -1,000 800 none none 50.0
Ethiopia
Male 17,200 23,000 29,000 -2,700 -4,000 -4,300 20,000 23,200 26,600 -300 -3,900 -6,500 88.9 2.5 none
Female 12,800 17,100 22,300 -600 -2,000 -2,200 14,200 18,100 22,800 800 -1,000 -1,700 233.3 50.0 22.7
Iraq
Male 19,300 25,700 30,200 -600 -1,300 -3,100 20,000 25,000 29,100 -300 -2,100 -4,000 50.0 none none
Female 12,400 18,700 23,600 -1,000 -500 -900 12,300 18,600 23,500 -1,100 -500 -1,000 none 0.0 none
Somalia
Male 15,100 21,000 26,900 -4,800 -6,000 -6,500 15,400 19,200 24,200 -5,000 -7,900 -8,800 none none none
Female 12,800 16,400 20,400 -600 -2,700 -4,100 11,100 14,800 19,400 -2,300 -4,300 -5,100 none none none
Afghanistan
Male 17,400 23,500 28,200 -2,400 -3,400 -5,200 16,900 23,000 27,400 -3,500 -4,100 -5,600 none none none
Female 11,600 16,700 21,100 -1,800 -2,400 -3,300 11,800 17,200 21,300 -1,600 -2,000 -3,200 11.1 16.7 3.0
PakistanTable 5 Note 5
Male 20,100 22,500 24,700 200 -4,500 -8,700 16,600 19,800 22,000 -3,800 -7,300 -11,100 2,000.0 none none
Female 13,200 16,200 18,200 -200 -3,000 -6,200 11,000 13,200 15,100 -2,400 -5,900 -9,400 none none none
ChinaTable 5 Note 5
Male 17,400 20,300 22,000 -2,400 -6,700 -11,300 15,500 20,900 24,300 -4,900 -6,200 -8,800 none 7.5 22.1
Female 15,400 17,800 19,500 2,000 -1,300 -5,000 14,600 17,800 20,800 1,200 -1,400 -3,700 40.0 none 26.0
Overall
Male 19,900 27,000 33,400 ... ... ... 20,300 27,100 33,100 ... ... ... ... ... ...
Female 13,400 19,100 24,500 ... ... ... 13,400 19,200 24,500 ... ... ... ... ... ...

The main finding is that in most cases, human capital characteristics, economic conditions and program of entry explain relatively little or none of the variation in earnings among refugees from different countries. Furthermore, both the size of the earnings gaps and the percentage that was accounted for in the regressions varied with years since landing. The variation in earnings among refugees from the 13 countries (measured by the earnings gaps between each refugee group and the mean value for all groups) increasedNote as refugees spent more time in Canada. And the percentage of this variation that was accounted for by the variables mentioned above declined between the year 1 and year 10 results. The average share (across the 13 groups) of the earnings gap that was accounted for fell from 30.7% in year 1 to 4.5% in year 10 for menNote and from 38.4% to 25.0% for women. This is not too surprising. The human capital variables included in the regression refer to differences that refugees bring to Canada, but post-arrival experiences, including differences in Canadian work experience, the acquisition of language skills and support from ethnic communities, could affect the size of the earnings gaps. These post-landing experiences are not captured in the model.

The model results were particularly interesting for refugee groups with earnings well below the average. For these refugee groups, including those from Ethiopia, Iraq, Somalia, Afghanistan and Pakistan, the human capital, economic conditions and program of entry variables included in the model accounted for virtually none of the earnings gap. The result for refugees from China was an exception; about 20% of the earnings gap in year 10 was explained by the independent variables in the model. In all the other cases, the refugees’ low earnings were not related to a deficiency (relative to the average) in observable characteristics.Note This held true for both men and women and for virtually all years since entry up to year 15 (Table 5).Note Other unknown and unobserved factors accounted for the low earnings.

In general, one has to look beyond educational attainment, language skills, economic conditions and the program under which they enter for an explanation as to why refugees from some countries earn more than those from other countries. Other possible reasons for the earnings gap are discussed in the conclusion.

4.3.3 Refugees’ earnings compared with those of family-class immigrants

During the first full year in Canada, refugees from all 13 countries had annual earnings below those of family-class immigrants entering during the same period. The refugee–family-class unadjusted earnings gaps were all negative (Table 6). This was true for both men and women. However, earnings for refugees from many countries displayed a higher growth rate than those of family-class or economic-class immigrants. The end result was that after 15 years in Canada, refugees from the former Yugoslavia, Poland and Colombia had earnings higher than their family-class counterparts entering during the same period; their unadjusted earnings gaps were positive. Although refugees from some other countries saw their earnings grow quite quickly, their earnings remained below those of family-class immigrants. As was the case 1 year after entry, refugees from countries such as China, Pakistan, Afghanistan, Somalia, Iraq and Ethiopia experienced large negative unadjusted gaps ranging from $6,000 to $23,000 compared with family-class immigrants, even after 15 years in Canada.


Table 6
Unadjusted and adjusted earnings gaps between refugees and family-class immigrants aged 20 to 49 at landing, by major source country, gender and years since landing
Table summary
This table displays the results of Unadjusted and adjusted earnings gaps between refugees and family-class immigrants aged 20 to 49 at landing. The information is grouped by Source country and gender (appearing as row headers), 1 year since landing, 5 years since landing, 10 years since landing, 15 years since landing, Earnings gap, Percentage explained , Unadjusted and Adjusted, calculated using thousands of dollars, percent and percent units of measure (appearing as column headers).
Source country and gender 1 year since landing 5 years since landing 10 years since landing 15 years since landing
Earnings gap Percentage explained Earnings gap Percentage explained Earnings gap Percentage explained Earnings gap Percentage explained
UnadjustedTable 6 Note 1 AdjustedTable 6 Note 2 UnadjustedTable 6 Note 1 AdjustedTable 6 Note 2 UnadjustedTable 6 Note 1 AdjustedTable 6 Note 2 UnadjustedTable 6 Note 1 AdjustedTable 6 Note 2
thousands of dollars percent thousands of dollars percent thousands of dollars percent thousands of dollars percent
Former Yugoslavia
Male -2.8 -0.1 98.0 0.9 4.2 none 4.0 7.6 none 5.2 8.9 none
Female -3.8 -1.8 53.6 1.1 3.3 none 5.0 7.3 none 6.5 8.7 none
Poland
Male -0.4 1.3 403.8 1.9 3.5 none 3.6 5.1 none 3.7 5.5 none
Female -3.2 -2.3 27.5 0.4 1.2 none 3.3 4.0 none 4.3 5.3 none
ColombiaTable 6 Note 3
Male -5.5 -3.0 45.9 -5.0 -1.6 68.0 0.2 2.7 none 5.7 6.7 none
Female -2.5 -1.1 55.8 -1.5 0.7 150.5 0.6 2.5 none 2.3 3.3 none
Viet Nam, Cambodia and Laos
Male -4.8 3.1 163.4 -6.1 2.4 139.9 -7.5 1.4 118.5 -8.9 0.1 101.3
Female -1.9 4.4 336.4 -2.4 4.1 268.3 -4.2 2.5 160.6 -7.1 -0.2 97.5
El Salvador
Male -7.7 -1.9 74.9 -9.3 -2.7 70.8 -10.2 -3.3 68.0 -9.9 -3.4 66.2
Female -6.0 -1.4 77.0 -6.0 -1.1 81.8 -6.3 -1.3 79.7 -7.0 -2.1 70.0
Sri Lanka
Male -0.2 1.2 612.3 -5.3 -4.3 19.3 -8.2 -7.5 8.1 -7.3 -6.8 7.5
Female -2.6 -0.6 76.0 -5.2 -3.3 35.1 -7.2 -6.0 16.3 -7.9 -7.8 0.7
Iran
Male -10.5 -8.9 16.0 -12.3 -9.9 20.1 -12.2 -9.7 20.9 -9.5 -7.8 17.3
Female -6.1 -5.1 16.9 -4.7 -3.0 36.7 -3.3 -1.4 57.6 -2.6 -1.1 55.6
Ethiopia
Male -6.2 -5.5 10.8 -10.0 -9.2 7.7 -12.6 -11.8 6.7 -12.9 -12.1 6.6
Female -4.2 -1.8 57.3 -5.6 -3.3 42.4 -6.4 -3.8 40.0 -6.1 -3.1 49.1
Iraq
Male -6.4 -3.0 52.7 -9.5 -6.6 30.9 -12.3 -9.4 23.2 -13.9 -10.6 23.8
Female -4.6 -2.6 44.7 -4.4 -2.5 44.2 -5.3 -3.3 38.7 -7.5 -5.0 32.6
Somalia
Male -9.1 -9.8 none -13.2 -14.5 none -15.2 -16.8 none -13.9 -15.2 none
Female -4.0 -4.5 none -6.3 -6.2 1.6 -8.2 -7.7 6.1 -9.2 -8.7 5.6
Afghanistan
Male -10.9 -7.4 32.2 -14.0 -10.2 27.1 -15.6 -12.2 21.9 -14.9 -12.6 15.1
Female -6.4 -2.5 60.9 -7.2 -3.4 53.2 -8.4 -5.2 38.6 -9.8 -7.7 21.1
PakistanTable 6 Note 3
Male -9.1 -11.0 none -15.7 -17.3 none -19.7 -22.8 none -19.6 -25.2 none
Female -5.1 -4.9 3.0 -8.1 -7.9 2.9 -11.8 -12.0 none -14.7 -15.6 none
ChinaTable 6 Note 3
Male -11.8 -7.6 35.2 -18.4 -12.8 30.5 -22.8 -17.7 22.6 -23.4 -19.1 18.2
Female -2.8 1.2 143.9 -6.7 -2.1 68.7 -10.6 -6.2 41.3 -11.9 -7.9 33.7

To determine whether differences in educational attainment, ability in the official languages, age and economic conditions explain the earnings gap between refugees and family-class immigrants, the study turns to the regression technique described in the “Data and methods” section.Note

The results are reported for years 1, 5, 10 and 15 in Table 6. The discussion of the results focuses on the earnings gaps five years after entering Canada and begins with male refugees. For the higher earnings groups (including from the former Yugoslavia; Poland; Colombia; El Salvador; and Viet Nam, Cambodia and Laos), there was either no negative gap with family-class immigrants or the explanatory variables accounted for all of the small earnings gap between the refugees and the family-class immigrants. For the remaining groups, the explanatory variables accounted for a small share of the sometimes very large negative earnings gaps between the refugees and the family-class immigrants entering during the same period. This was particularly true for lower-earning groups from countries such as Iran, Ethiopia, Iraq, Somalia, Afghanistan, Pakistan and China. These refugee groups earned from $10,000 to $18,000 less than family-class immigrants did, and 0% (Pakistan and Somalia) to 31% of the gap was accounted for by the explanatory variables.

For female refugees five years after landing, the explanatory variables generally accounted for a larger share of the refugee–family-class earnings gap than was the case for male refugees, particularly for groups with lower earnings. For example, for refugees from China and Afghanistan, the explanatory variables accounted for over half of the earnings gap, something not seen for male refugees in the lower earnings groups. Otherwise, the results resembled those of the male refugees.

In summary, for the male refugee groups with lower earnings (from Iran, Ethiopia, Iraq, Somalia, Afghanistan, Pakistan and China) and hence larger earnings gaps with family-class immigrants, differences in the observable characteristics (i.e., education, language and age) and economic conditions between them and family-class immigrants accounted for relatively little of these gaps. In general, the explanatory variables were more successful in explaining the gap for female refugees, but, in most cases, they still explained less than half of the earnings gap. Refugees from these countries were earning less than family-class immigrants for other reasons. For two groups (from Somalia and Pakistan), virtually none of the gap was explained by the differences in human capital characteristics. This was true for men and women and in all years since landing. For refugee groups with higher earnings (including from the former Yugoslavia; Poland; Colombia; and Viet Nam, Cambodia and Laos), there was either no earnings gap or the earnings gap was entirely explained by differences in human capital characteristics. This was true for both men and women.

4.3.4 The earnings gap between refugees and economic-class immigrants

The earnings of economic-class immigrants were, on average, higher than those of family-class immigrants. Hence, the earnings gap between refugees and economic-class immigrants tended to be larger. Refugees (both men and women) in groups with relatively higher average earnings, such as from the former Yugoslavia, Poland and Colombia, earned from 55% to 66% of what economic-class immigrants did during the first full year in Canada. These percentages increased to from 75% to 96% 15 years later (Table 7). These employed refugees were doing fairly well after 15 years in Canada, and even 5 years after entry they were earning 70% to 80% of what economic-class immigrants did. Five years after entry, from 47% to 100% of the relatively small gap was associated with differences in observable characteristics or economic conditions between refugees and economic-class immigrants (Table 7). In other words, the majority of the gap is explained by differences in these characteristics.


Table 7
Unadjusted and adjusted earnings gaps between refugees and economic-class immigrants aged 20 to 49 at landing, by major source country, gender and years since landing
Table summary
This table displays the results of Unadjusted and adjusted earnings gaps between refugees and economic-class immigrants aged 20 to 49 at landing. The information is grouped by Source country and gender (appearing as row headers), 1 year since landing, 5 years since landing, 10 years since landing, 15 years since landing, Earnings gap, Percentage explained , Unadjusted and Adjusted, calculated using thousands of dollars and percent units of measure (appearing as column headers).
Source country and gender 1 year since landing 5 years since landing 10 years since landing 15 years since landing
Earnings gap Percentage explained Earnings gap Percentage explained Earnings gap Percentage explained Earnings gap Percentage explained
UnadjustedTable 7 Note 1 AdjustedTable 7 Note 2 UnadjustedTable 7 Note 1 AdjustedTable 7 Note 2 UnadjustedTable 7 Note 1 AdjustedTable 7 Note 2 UnadjustedTable 7 Note 1 AdjustedTable 7 Note 2
thousands of dollars percent thousands of dollars percent thousands of dollars percent thousands of dollars percent
Former Yugoslavia
Male -12.1 -0.2 98.6 -12.4 -0.8 93.3 -12.5 -1.3 89.4 -12.4 -1.5 88.1
Female -7.6 -0.3 96.0 -6.7 0.5 106.8 -5.6 1.4 124.4 -4.4 2.2 149.5
Poland
Male -14.1 -8.3 41.0 -10.9 -5.8 46.9 -8.9 -4.2 53.2 -9.4 -4.4 53.2
Female -8.1 -5.0 37.8 -5.1 -2.3 54.1 -2.5 0.2 106.5 -1.3 1.7 226.6
ColombiaTable 7 Note 3
Male -11.1 -3.0 72.7 -16.1 -7.2 55.4 -16.4 -7.5 54.1 -13.3 8.6 165.0
Female -7.2 -2.0 71.4 -10.0 -4.2 58.3 -11.4 -5.8 49.1 -11.0 -6.2 43.7
Viet Nam, Cambodia and Laos
Male -18.5 -6.1 67.1 -19.4 -6.9 64.6 -20.8 -8.2 60.7 -22.6 -9.8 56.4
Female -6.6 2.2 133.7 -8.0 0.8 110.5 -10.3 -1.4 86.8 -13.3 -4.1 69.0
El Salvador
Male -19.9 -10.1 49.2 -21.0 -11.0 47.6 -22.0 -11.9 45.9 -22.7 -12.6 44.5
Female -10.9 -3.8 65.3 -11.7 -4.4 62.1 -12.5 -5.2 58.4 -13.2 -6.0 54.8
Sri Lanka
Male -8.1 1.2 114.3 -18.1 -9.7 46.0 -24.5 -16.9 31.1 -24.4 -16.9 30.7
Female -7.0 -0.5 92.6 -13.4 -7.2 46.3 -17.8 -12.4 30.6 -18.3 -14.0 23.3
Iran
Male -18.6 -9.0 51.4 -25.0 -15.3 38.9 -28.0 -18.9 32.5 -25.2 -17.8 29.6
Female -10.0 -4.7 53.6 -12.5 -6.5 47.7 -13.5 -7.5 44.6 -12.0 -6.8 43.6
Ethiopia
Male -17.7 -13.0 26.2 -21.5 -17.6 18.2 -24.7 -21.0 14.8 -26.0 -21.9 16.0
Female -9.4 -4.1 56.1 -11.6 -6.5 44.0 -13.1 -7.9 40.0 -13.1 -7.4 43.9
Iraq
Male -15.8 -4.5 71.3 -22.8 -12.4 45.9 -28.4 -18.5 34.7 -30.3 -20.6 31.8
Female -9.4 -2.6 72.5 -12.3 -5.8 53.3 -15.2 -8.8 42.0 -17.3 -10.9 36.9
Somalia
Male -20.4 -12.3 39.6 -27.2 -20.8 23.5 -31.5 -26.3 16.7 -31.2 -25.9 16.7
Female -8.3 -4.1 50.5 -13.6 -9.0 34.0 -17.8 -13.0 27.0 -19.3 -14.7 23.8
Afghanistan
Male -17.5 -6.7 61.6 -26.0 -15.2 41.7 -32.1 -21.3 33.6 -32.9 -22.4 31.9
Female -10.4 -2.2 79.0 -15.6 -7.5 51.7 -19.8 -12.2 38.2 -21.3 -14.6 31.4
PakistanTable 7 Note 3
Male -14.0 -9.8 29.9 -26.7 -22.4 16.2 -36.7 -33.3 9.3 -40.0 -38.2 4.6
Female -8.9 -4.8 46.1 -16.5 -12.4 24.6 -24.0 -20.3 15.5 -28.3 -25.1 11.2
ChinaTable 7 Note 3
Male -16.6 -6.8 59.2 -29.5 -18.2 38.4 -39.5 -28.2 28.5 -41.6 -31.0 25.4
Female -6.9 1.0 114.5 -15.2 -6.8 55.0 -22.5 -14.5 35.7 -24.5 -16.9 31.2

For the refugee groups with the lowest earnings, such as those from Iran, Ethiopia, Iraq, Somalia, Afghanistan, Pakistan and China, the story was very different. These refugees (both men and women) earned from 40% to 60% of what economic-class immigrants did during the first year in Canada. Refugees from Iran and Ethiopia saw their earnings rise to from 60% to 70% of what economic-class immigrants earned 15 years after entering Canada. Refugees from the other countries saw no relative improvement. In fact, the earnings gap increased significantly for some refugee groups. After 15 years in Canada, those from Pakistan and China saw their relative earnings decline to 35% to 45% of what economic-class immigrants earned. For all the groups with low earnings, something other than observable characteristics accounts for these sometimes very large earnings gaps with economic-class immigrants. Five years after landing, the percentage of the gap explained rarely exceeded 50% and was sometimes as low as 20%. Other factors must be examined to learn why both male and female refugees from these countries earned so much less than economic-class immigrants did even after 10 or 15 years in Canada.

4.3.5 The difference in earnings among refugee program types

Refugees typically enter Canada one of four ways:

The average earnings among refugees entering through these programs varied considerably. For example, one year after entry, refugee men in the In-Canada Asylum Program earned the most, at around $24,000, followed by privately sponsored refugees at $23,000 and government-assisted refugees at $20,100 (Table 8, left panel). In the longer run, privately sponsored refugee men appeared to have the most economic success. After 10 years in Canada, they were earning $40,600, compared with $38,100 for government-assisted refugees and $33,200 for In-Canada Asylum Program refugees.


Table 8
Predicted average annual earnings of refugees from major source countries aged 20 to 49 at landing, by program of entry, gender and years since landing
Table summary
This table displays the results of Predicted average annual earnings of refugees from major source countries aged 20 to 49 at landing. The information is grouped by Gender and years since landing (appearing as row headers), Unadjusted annual average earnings, Adjusted annual average earnings, Government-assisted resettlement, Private
sponsorship, In-Canada Asylum Program, Gap between government-assisted resettlement and and Private sponsorship, calculated using thousands of dollars units of measure (appearing as column headers).
Gender and years since landing Unadjusted annual average earningsTable 8 Note 1 Adjusted annual average earningsTable 8 Note 2
Government-assisted resettlement Private
sponsorship
In-Canada Asylum Program Gap between government-assisted resettlement and Government-assisted resettlement Private
sponsorship
In-Canada Asylum Program Gap between government-assisted resettlement and
Private sponsorship In-Canada Asylum Program Private
sponsorship
In-Canada Asylum Program
thousands of dollars
Men
Years since landing
1 20.1 23.0 24.0 -2.9 -3.9 20.5 23.3 25.3 -2.8 -4.8
5 29.8 32.3 28.7 -2.5 1.1 30.0 31.2 31.0 -1.2 -1.0
10 38.1 40.6 33.2 -2.5 4.9 38.0 38.4 35.5 -0.4 2.5
15 42.3 45.3 36.2 -3.0 6.1 41.7 42.6 36.9 -0.9 4.8
Women
Years since landing
1 12.6 15.0 16.2 -2.4 -3.6 11.5 13.9 16.1 -2.4 -4.6
5 20.8 21.9 20.8 -1.1 0.0 19.2 20.1 21.6 -0.9 -2.4
10 27.9 28.6 24.8 -0.7 3.1 26.2 26.5 25.6 -0.3 0.6
15 31.7 33.4 26.6 -1.7 5.1 30.1 31.5 26.5 -1.4 3.6

Many reasons could explain these differences in earnings. Privately sponsored refugees have a closer relationship with their Canadian sponsors immediately after entry. This may assist in job searches, training and language acquisition and provide them with an earnings advantage. In-Canada Asylum Program refugees have been in Canada for some time before they are granted permanent residence. Their exposure to Canadian society and the labour market may provide an advantage in the early years after landing.

Given the data available for this study, little can be learned about these possible advantages. However, another set of factors could affect the earnings differences, including the observable human capital characteristics (including source country) and economic conditions in the regions where refugees settle. The regression models described in the “Data and methods” section were used to produce adjusted earnings controlling for these variables (Table 8, right panel).

The results show that the effect of a program on earnings tended to be greater during the early years in Canada than in the long run. In the longer run, refugees learned more about job opportunities and experienced language and occupational training and other integration programs. These factors may have reduced any advantage that refugees in one program held over others.

The actual (unadjusted) data support this idea. One year after entry, privately sponsored immigrants held a $2,900 advantage over government-assisted refugees. Although the absolute gap stayed from $2,500 to $3,000 after 5 to 10 years in Canada, the relative percentage difference decreased (Table 8, right panel). In-Canada Asylum Program refugees held a $3,900 annual earnings advantage over government-assisted refugees the first year after entry, but after 5 to 10 years in Canada, this earnings gap turned negative.

The adjusted results suggest that the human capital, economic and source country characteristics accounted for very little of the earnings differential very early after immigration but accounted for a large portion of the differences after 5 to 10 years in Canada (Table 8, left panel). After observable differences were controlled for, privately sponsored refugees and government-assisted refugees had similar earnings about 10 years after immigration, while In-Canada Asylum Program refugee men earned $2,500 less than privately sponsored refugees after 10 years in Canada.

5 Conclusion

This study focuses on the employment and earnings outcomes of refugees to Canada from 13 major source countries over the 1980-to-2009 period. These outcomes were tracked for up to 15 years after entry. This paper discusses a number of salient findings.

First, male refugees from 7 of the 13 countries had employment rates over 75% five years after entry. But those from Iran and Somalia had very low employment rates. Female refugees from Iraq, Afghanistan, Pakistan and Somalia also displayed very low employment rates. Refugee groups that displayed relatively low male employment rates tended to display low female employment rates, and groups that displayed high male employment rates tended to display high female employment rates. To the extent that men and women marry within the same refugee group, the tendency of families to have both spouses not be employed would exacerbate poverty and economic inequality within that refugee group.

Second, the earnings of refugees from different countries varied greatly. Ten years after entering Canada, the refugee groups with the highest earnings (i.e., from the former Yugoslavia, Poland and Colombia) earned roughly double what those with the lowest earnings did (i.e., from Somalia, Afghanistan, Pakistan and China). Differences in outcomes were accentuated by the fact that groups with low employment rates tended to have low earnings levels among the employed, and groups with high employment rates tended to have high earnings among the employed. Furthermore, male and female refugees from the same country tended to have similarly low or high relative earnings. These high correlations in men’s and women’s earnings across groups tended to exacerbate any poverty issues for refugees with low earnings.

Third, differences in observable human capital characteristics, economic conditions and the program of entry to Canada explained either none or very little of the much lower than average earnings of the refugee groups with the lowest earnings. Other unknown and unobserved factors accounted for these low earnings. One possible explanation is that the education that these groups received in their source countries may have been of lower quality or have been perceived to be of lower quality, thereby affecting the possibility of finding jobs that matched their education level or the wages employers were willing to pay the university-educated in particular. Some evidence supports this notion. A separate regression analysis (not shown) revealed that university-educated refugees from Somalia, Afghanistan, Pakistan and China earned little more (from $1,100 to $4,400 per year) than their counterparts who graduated from high school. The returns to a university degree were quite low. In contrast, refugees from Poland, Colombia and Sri Lanka had an earnings advantage for a university degree in the $11,000-to-$12,000 range. Even if a significant share of the refugees from the countries with low returns had a university degree, this would have had little positive effect on overall average earnings. Other unknown factors may be related to some of the variation in earnings across source-country groups.

Distinguishing between the factors that are evident at landing and those that have an effect after landing is useful. Perceived or actual differences among refugees from different countries in the quality of education and labour market experience in the source country, cultural factors affecting labour market success, and discrimination may be some of the unknown factors at play immediately upon landing. This study found that the earnings variation among groups increased between year 1 and year 10 after landing. Differences among refugee groups in post-landing experiences that may have resulted in such an increase could include differences in

Such factors were not included in the models. In this context, it is not surprising that the models explained more of the variation in earnings among refugees from different countries at 1 year after landing than at 10 years after landing.

Fourth, immediately after entry, refugees from all countries earned less than family-class immigrants. However, the five refugee groups with the highest earnings (i.e., from the former Yugoslavia; Poland; Colombia; Viet Nam, Cambodia and Laos; and El Salvador) displayed much more rapid earnings growth than did the family- and economic-class immigrants entering during the same period. After 15 years, they earned from 80% to 110% of what family-class immigrants did. The story was very different for refugees in the five lowest earnings groups (i.e., from Iraq, Somalia, Afghanistan, Pakistan and China). Even after 15 years in Canada, they earned from 50% to 70% of what family-class immigrants did. For these groups, observable differences accounted for very little of the earnings gap with family-class immigrants, possibly for the reasons mentioned above. Once again, unobserved and unknown factors played a major role. The comparison with economic-class immigrants provided similar results, only the earnings gaps were larger. After 15 years in Canada, the groups with higher earnings earned from 60% to 90% of what economic-class immigrants did, and the lower earnings groups earned from 35% to 60%. Little of this latter gap could be explained in the analysis.

Fifth, the program under which refugees entered Canada seemed to matter, at least in the initial years. Even after differences in observable human capital, source country, and economic conditions were accounted for, privately sponsored refugees earned from $2,400 (for women) to $2,800 (for men) more than government-assisted refugees in the first year in Canada. This may be related to the assistance that the Canadian sponsors provided regarding job search, contacts and other integration services, as suggested by some previous studies (e.g., Beiser 2003; Roma 2016). Differences in human capital factors did not account for this initial gap, but they did account for most of the earnings gap 10 years after immigration. Refugees who claimed asylum within Canada initially earned $4,800 more (adjusted) than government-assisted refugees, but after 10 years they earned roughly $2,500 less. This may be because they had some Canadian work experience at entry; however, this advantage disappeared over time.

References

Abbott, M., and C.M. Beach. 2008. Immigrant Earnings Distributions and Earnings Mobility in Canada: Evidence for the 1982 Landing Cohort from IMDB Micro Data. CLSRN Working Paper no. 13. Vancouver: Canadian Labour Market and Skills Researcher Network.

Abbott, M.G., and C.M. Beach. 2011. Do Admission Criteria and Economic Recessions Affect Immigrant Earnings? IRPP Study no. 22. Montréal: Institute for Research on Public Policy.

Alba, R., and V. Nee. 2003. Remaking the American Mainstream: Assimilation and Contemporary Immigration. Cambridge, Massachusetts: Harvard University Press.

Aydemir, A. 2011. “Immigrant selection and short-term labor market outcomes by visa category.” Journal of Population Economics 24 (2): 451–475.

Beiser, M. 2003. “Sponsorship and resettlement success.” Journal of International Migration and Integration 4 (2): 203–215.

Bevelander, P., and R. Pendakur. 2012. The Labour Market Integration of Refugees and Family Reunification Immigrants: A Comparison of Outcomes in Canada and Sweden. IZA Discussion Paper no. 6924. Bonn: Institute of Labor Economics.

CIC (Citizenship and Immigration Canada). 2006. Facts and Figures: Immigration Overview: Permanent and Temporary Residents: 2005. Research and Evaluation Branch. Ottawa. 114 p. Available at: http://publications.gc.ca/collections/Collection/Ci1-8-2005E.pdf (accessed October 16, 2018).

CIC (Citizenship and Immigration Canada). 2009. Facts and Figures: Immigration Overview: Permanent and Temporary Residents: 2008. Research and Evaluation Branch. Ottawa. 108 p. Available at: http://publications.gc.ca/collections/collection_2009/cic/Ci1-8-2008E.pdf (accessed October 16, 2018).

CIC (Citizenship and Immigration Canada). 2015. Canada Facts and Figures: Immigration Overview: Permanent Residents: 2014. Research and Evaluation Branch. Ottawa. 53 p. Available at: http://publications.gc.ca/collections/collection_2015/cic/Ci1-8-9-2014-eng.pdf (accessed October 16, 2018).

De Silva, A. 1997. “Earnings of immigrant classes in the early 1980s in Canada: A re-examination.” Canadian Public Policy 23 (2): 179–202.

DeVoritz, D., and S. Pivnenko. 2004. The Economic Experiences of Refugees in Canada. IZA Discussion Paper no. 1088. Bonn: Institute of Labor Economics.

Evans, W.N., and D. Fitzgerald. 2017. The Economic and Social Outcomes of Refugees in the United States: Evidence from the ACS. NBER Working Paper Series, no. 23498. Cambridge, Massachusetts: National Bureau of Economic Research.

Green, D., H. Liu, Y. Ostrovsky, and G. Picot. 2016. Immigration, Business Ownership and Employment in Canada. Analytical Studies Branch Research Paper Series, no. 375. Statistics Canada Catalogue no. 11F0019M. Ottawa: Statistics Canada.

Hou, F. 2017. The Resettlement of Vietnamese Refugees Across Canada over Three Decades. WIDER Working Paper no. 2017/188. United Nations University World Institute for Development Economics Research.

Immigration and Refugee Protection Act. S.C. 2001, c. 27, paragraph 3(2)(d).

IRCC (Immigration, Refugees and Citizenship Canada). 2016. Evaluation of the Resettlement Programs (GAR, PSR, BVOR and RAP). Last updated March 8, 2017. Available at: https://www.canada.ca/en/immigration-refugees-citizenship/corporate/reports-statistics/evaluations/resettlement-programs.html (accessed October 11, 2018).

Li, P.S. 2003. “Initial earnings and catch-up capacity of immigrants.” Canadian Public Policy 29 (3): 319–337.

O’Connor, P. 2010. “Explaining the refugee gap: Economic outcomes of refugees versus other immigrants.” Journal of Refugee Studies 23 (3): 377–397.

Painter, C.V., and A. Kokallaj. 2017. Statistics Profile of Former Refugees by Crisis of Origin. Ottawa: Immigration, Refugees and Citizenship Canada.

Pendakur, R. 2017. Settlement and Labour Force Outcomes for Afghan Immigrants and Their Children in Canada. WIDER Working Paper no. 2017/189. United Nations University World Institute for Development Economics Research.

Picot, G., and Y. Ostrovsky. 2017. Immigrant Businesses in Knowledge-based Industries. Economic Insights, no. 69. Statistics Canada Catalogue no. 11-626-X. Ottawa: Statistics Canada.

Portes, A., and M. Zhou. 1993. “The new second generation: Segmented assimilation and its variants.” Annals of the American Academy of Political and Social Science 530: 74–96.

Roma, G. 2016. “The Indochinese refugee movement: An exploratory case study of the Windsor experience.” Refuge 32 (2): 81–89.

Date modified: