Economic and Social Reports
Economic outcomes of government-assisted refugees in designated destinations: The effect of city size

Abstract

This study asks whether there were significant differences in the economic outcomes of government‑assisted refugees (GARs), based on the size of the city to which they were designated. The analysis was conducted for both those remaining in the designated cities (stayers) and those moving to other locations (movers). As supported by previous literature, the study found that the likelihood of moving was much higher among GARs who were assigned to smaller communities, compared with those assigned to large cities. Nonetheless, among GAR stayers, those assigned to Toronto had the lowest employment incidence and annual earnings, while stayers in medium-sized and small cities reported better economic outcomes. Similar patterns were observed among GAR movers, where those who chose to move to gateway cities (Montréal, Toronto and Vancouver) had the lowest employment incidence and annual earnings, while those who moved to a smaller city tended to report stronger labour market outcomes. The regression analysis found that the majority (89%) of these differences across city size could be explained by the city’s employment rate, as well as the source region of GARs. These explanatory variables also accounted for about half of the differences in annual earnings by city size.

Keywords: refugees, secondary migration, regional retention, earnings

Authors

Yasmin Gure and Garnett Picot are with the Research and Evaluation Branch at Immigration, Refugees and Citizenship Canada. Feng Hou is with the Social Analysis and Modelling Division, Analytical Studies and Modelling Branch, at Statistics Canada.

Acknowledgments

This study was conducted by Immigration, Refugees and Citizenship Canada and Statistics Canada. The authors would like to thank Stéphane Arabackyj, Chris Hamilton and Humaira Somra for their advice and comments on an earlier version of this paper.

Introduction

A key component of Canada’s immigration program is to promote the balanced geographic distribution of immigrants and refugees across Canada (IRCC, 2020). Through various policy tools, Immigration, Refugees and Citizenship Canada (IRCC) has invested in steering newcomers away from the nation’s largest cities to medium-sized and small metropolitan areas. It is believed that through these initiatives, the benefits of immigration, including economic growth and ethnocultural diversity, can be shared throughout the country. Unlike other immigrant or refugee admission categories, IRCC plays a direct role in determining the initial resettlement destination of government-assisted refugees (GARs) before arrival. Under this refugee stream, the department balances the specific resettlement needs of GARs with IRCC’s commitments to resettle refugees to cities across Canada.

The prevailing research on regional retention of newcomers in Canada tends to focus on immigrants broadly and is often limited to data at the provincial level. While this research can help provide some context on secondary migration patterns, it can limit the understanding of migration trends of refugees that may be occurring at a lower geographic level. More recently, studies have begun to analyze the retention and secondary migration patterns of refugees, at the city level (Gure & Hou, 2022; Kaida et al., 2020). However, there is a significant knowledge gap in the literature in terms of how retention and secondary migration decisions affect the economic outcomes of refugees. From a policy perspective, this is particularly relevant, since dispersion policies cannot be efficient without successful economic integration.

This study seeks to understand how the economic outcomes of GARs vary based on the size of the city of resettlement. Moreover, this study compares the economic outcomes of GARs who choose to remain in their initial destination and those who choose to move. This study also aims to understand the extent to which these economic outcomes can be explained by differences in the sociodemographic characteristics of GARs, as well as the conditions of the city of resettlement. The answers to these questions can provide valuable information on the retention and secondary migration decisions of refugees in Canada and may provide further evidence to inform how resettlement destinations may be selected for GARs.

Background

Resettlement of refugees in Canada and destination matching

Refugees who are selected abroad for resettlement in Canada are admitted through one of three resettlement streams: GARs, privately sponsored refugees (PSRs) and blended visa office-referred refugees (BVORs). While PSRs’ and BVORs’ resettlement destination is determined, in part, by their sponsors in Canada, GARs are “matched” or assigned to specific communities in Canada before arrival. The decision about where a GAR may be resettled is often determined by factors such as whether they have family or friends living in the area; the presence of ethnic, cultural and religious communities; and the availability of specific settlement services. Additionally, IRCC also determines where to resettle GARs based on discussions with provinces, which help inform the department’s pre-established annual plans and quotas to resettle refugees across Canada (IRCC, 2016).

An important pillar of successful destination matching for GARs is retention. In other words, the effectiveness of the destination assignment process may be best measured by the proportion of GARs who choose to remain in their initial destination, following arrival in Canada. In addition to considerations for determining the ideal resettlement location for GARs, efficient resettlement also relies on the articulated preferences of GARs before their arrival. For instance, Simich et al. (2002) found that over 90% of GARs who changed their destination upon landing at the airport believed that they were not accommodated according to their expressed preferences. In more recent years, evidence suggests that the destination-matching process may have improved. A recent study by Gure & Hou (2022) found that since the early 2000s, overall retention rates have increased among successive GAR landing cohorts. This increase may indicate, among other factors, that there is greater consideration of GARs’ preferences.

Retention and secondary migration of refugees

Upon arrival in Canada, a subgroup of GARs may ultimately decide that they no longer wish to remain in their assigned destination. Although GARs may be resettled to cities across Canada, of varying sizes, existing literature supports the notion that refugees tend to prefer to reside in larger cities, rather than smaller communities or rural areas (IRCC, 2011; Derwing & Krahn, 2008; Hyndman et al., 2006; Sherrell et al., 2005). For GARs who have been resettled to these smaller regions, out-migration in the years following resettlement is high. Kaida et al. (2020) found that the proportion of GARs who chose to leave their initial destination increased with smaller size of the initial city they were destined to. This was supported in a 2022 study by Gure and Hou, where it was observed that GARs resettled to smaller communities were significantly more likely to leave their initial destination. Moreover, the study found that while the majority of secondary migration occurred in the first year after arrival, the share of GARs resettled to these smaller communities continued to decline over the 10-year study period.

Prior research has extensively explored the factors that may motivate newcomers to settle in large cities. One of the most frequently cited reasons for secondary migration among immigrants is the pursuit of economic opportunities (Derwing & Krahn, 2008; Hou, 2007; Hyndman et al., 2006). Compared with larger cities, smaller or more rural communities may have a more difficult time retaining immigrants, because of limited job prospects (Sherrell et al., 2005; Abu-Laban et al., 1999). This may be particularly salient among GARs, who possess lower levels of human capital than other immigrant groups, which may exacerbate adverse labour market outcomes. These GARs may be motivated to move to larger cities, where they may expect more favourable employment opportunities.

The potential for better labour market outcomes based on city size may also be underpinned by the presence of co-ethnic communities, which tend to be more established in Canada’s larger cities (Hiebert, 2015; Hyndman et al., 2006; Qadeer & Kumar, 2006). The literature has examined the role of large ethnic communities in settlement through the lens of the group affinity hypothesis, which suggests that new immigrants prefer to move to and remain in areas where co-ethnic communities are concentrated (Derwing & Krahn, 2008; Hou, 2007; Nogle, 1994). For GARs, the desire to resettle in a large city among co-ethnic communities may be a significant pull factor for secondary migration; the difficult and traumatic resettlement circumstances may further emphasize the need for familiar social and emotional support. Moreover, these established communities may serve as social networks for GARs, connecting them with valuable resettlement resources, including access to job opportunities, housing and other economic prospects.

While it is evident that refugees may prefer to resettle in larger cities for their perceived better economic outcomes, research is limited on whether these aspirations are actualized in the long term, especially when compared with refugees who reside in smaller communities. In the context of GARs, whose resettlement destinations are selected before arrival, this lack of information presents a critical knowledge gap. This study seeks to understand how the economic outcomes of GARs varies based on the size of the city of resettlement.

This research makes several contributions to the existing literature.

Firstly, this study is among the first to analyze the economic outcomes of GARs using census metropolitan areas (CMAs) and census agglomerations (CAs) as the geographic unit of analysis. Existing empirical evidence on the economic outcomes of refugees is often confined to the national level, which can obscure potentially large variations between cities across Canada.

Secondly, this study uses the Longitudinal Immigration Database (IMDB) to identify refugees at the admission category level. This allows for the analysis to track the labour market outcomes of refugee subcategories over time and, as a result, provides a more comprehensive and systematic examination. This research fills existing knowledge gaps, where previous analysis may have grouped refugees or immigrants together.

Lastly, this study analyzes how the economic outcomes of GARs may differ between GARs who remained in their designated destination and those who moved elsewhere, and whether such a difference may vary by the size of the designated destination. To the authors’ knowledge, this is the first Canadian study to explore the association between secondary migration and labour market outcomes among refugees.

Data, measures and methods

Data

This study uses the IMDB, which combines the landing records and annual tax information of immigrants (Statistics Canada, 2020). Immigrants who have filed at least one tax return since 1982 are included in the database. The landing records contain immigrant characteristics at the time of landing, such as intended destination,^Note admission category (e.g., economic class, family class or refugee), age, education, marital status, source country and official language capabilities. Tax records provide information on annual income, current marital status and place of residence. The IMDB data in this study cover landing information up to 2021 and tax information up to 2020.

The analysis focused on 77,340 GARs aged 20 to 54 at the time of immigration and who arrived in Canada from 2000 to 2019. The 20-to-54 age group was selected because they are prime-aged workers at entry, and after 10 years in Canada, a follow-up period used in the study, they would be aged 30 to 64, still mostly of working age.

Measures

CMAs and CAs are the main geographic units for intended destinations and measuring subsequent mobility. Some CMA or CA boundaries change from one census to another, and this change poses a challenge when using CMAs and CAs to study economic outcomes of movers and stayers. To overcome this problem, this study created consistent CMA and CA boundaries in the tax file using 2016 Census boundaries as the base.^Note

The main outcome variables in this study are annual earnings from paid employment and employment incidence. Annual employment earnings are derived from tax files. Employment incidence for any given year is the number of individuals earning more than $500 in that year, divided by the number of individuals filing taxes in the same year. Over the study period, roughly 97% of GARs and 96% of PSRs filed taxes during their first full year in Canada. Filing rates were marginally lower among family class immigrants, at approximately 93%, and economic immigrants, at roughly 87%.

In the multivariate analysis predicting annual earnings and employment incidence of GARs, three sets of explanatory variables are used.

The first set includes the geographic measures. For GARs who are designated to a city and “stay” there, six city types are used, including the three gateway cities of Toronto, Montréal and Vancouver; medium‑sized CMAs (Ottawa, Calgary, Edmonton, Hamilton, Winnipeg and Québec); small CMAs (e.g., Victoria, Saskatoon and Halifax); and small urban areas (CAs). In addition, 11 variables capturing the movement of GARs among the city types are employed. These 17 variables, involving “stayers” and “movers,” are the primary independent variables of interest in this study. Refugees who intended to settle in rural areas were excluded, as it is difficult to create consistent boundaries for scattered rural areas. Moreover, less than 1% of refugees were intended to settle in rural areas.

The second group of explanatory variables relates to the individual-level sociodemographic characteristics of the GARs, including age at landing, gender, marital status, number of children, source region, language at landing, level of education at landing, employment status after landing and school attendance after landing. Source region is coded into nine categories: South America, Southern Europe, Eastern Europe, Africa, Southern Asia, Southeast Asia, East Asia, West Asia and others. Language skills are based on the combination of mother tongue and self-reported knowledge of official languages at the time of landing: English mother tongue, French mother tongue, other mother tongue but spoke English, other mother tongue but spoke French, other mother tongue but spoke English and French, or did not speak English or French. The education variable is coded into four categories: not stated, high school education or less, some postsecondary education, or bachelor’s degree or higher. The school attendance variable is coded as 1 if an immigrant attended full-time postsecondary education in the last tax year or as 0 if otherwise.^Note

The third set of explanatory variables provides city-level characteristics for intended destinations for stayers and current city of residence for movers, including the presence of Resettlement Assistance Program service provider organizations (RAP SPOs);^Note the economic conditions, as measured by the local employment incidence and median annual earnings among the adult (aged 18 to 64) Canadian-born population and long-term immigrants (those who arrived in Canada more than 20 years earlier); the presence of ethnic enclaves; and cluster resettlement. The RAP SPO variable was coded into four categories: none, 1 to 5 years in operation, 6 to 10 years in operation, or over 10 years in operation. This variable was used to capture the effects of not just the existence of RAP SPOs but also their accumulated experience. The ethnic enclave variable was measured by the percentage of immigrants from the same source region residing in the CMA or CA where a refugee was assigned to settle or currently residing. The cluster resettlement variable was measured by the number of GARs who arrived at the same intended destination in the same year. In the multivariate analysis, the logarithm transformation of this variable was used.

Methods

Linear probability models are used to estimate the effect of the explanatory variables on employment incidence. Regression models were run separately for 1, 5 and 10 years after landing. The sample includes GARs aged 20 to 54 at landing who arrived from 2000 to 2019. For each observation period, two models were run. Employment incidence was the dependent variable in both models. The independent variables in Model 1 include only the first set of independent variables described above, which are the 6 city types (for stayers) and 11 “mover” categories (for movers). The category of stayers in Toronto is used as the reference group. This means the coefficients on the mover and stayer variables indicate the percentage point difference in employment incidence between each category of these stayers or movers and stayers in Toronto. Model 2 includes the independent variables in Model 1 and adds the second and third sets of explanatory variables described above. The difference in the coefficients between Model 1 and Model 2 on the mover or stayer independent variables indicates the extent to which the control variables accounted for the gap in employment incidence between Toronto stayers and any other mover or stayer types. An example is given below.^Note

A set of ordinary least squares models is used to estimate the effect of the explanatory variables on annual earnings of GARs. The dependent variable is the level of annual earnings of the GARs. The other details regarding the modelling are identical to those outlined for the employment incidence estimates, except with the level of median annual earnings in each city added as an additional explanatory variable. For this analysis, stayers in Toronto are used again as the reference group.

Results

Intended destinations of government-assisted refugees and other immigrants

Before arrival in Canada, GARs are assigned a “designated” destination, referred to as an intended destination in the IMDB. Compared with the family class and economic immigrants, GARs were much less likely to intend to resettle in Montréal, Toronto and Vancouver, and more likely to intend to settle in medium-sized and small metropolitan centres (Table 1). GARs were also somewhat less likely than PSRs to intend to settle in the three largest cities, and more likely to have an intended destination of a small metropolitan or urban area.



Stayers and movers

Immigrants may not reside in their intended (designated) destination by the end of their first full year in Canada,^Note or in years thereafter, either because they never resided in their intended destination, or because they landed in that destination but moved shortly after. These individuals were referred to as movers. Possible reasons why such a move may take place were discussed by Gure and Hou (2022). Stayers were those immigrants who resided in their intended destination in any given year.^Note The economic outcomes of both movers and stayers residing in cities of different sizes were assessed at the 1st full year, the 5th year and the 10th year following landing.

The tendency of immigrants to move varied significantly by immigrant class. By the 5th and, particularly, 10th year following landing, a much higher percentage of refugees were not found in their intended destination, compared with family or economic class immigrants (Chart 1). By the 10th year, 42% of GARs and 38% of PSRs had moved from their intended destination, compared with 19% and 27% of the family and economic classes, respectively.

Data table for Chart 1



Data table for Chart 1
Table summary
This table displays the results of Data table for Chart 1 Government-assisted refugees, Privately sponsored refugees, Family class and Economic class, calculated using percent units of measure (appearing as column headers).

	Government-assisted refugees	Privately sponsored refugees	Family class	Economic class
	percent
Year 1	24.1	27.0	10.9	20.8
Year 5	34.9	34.2	16.1	25.9
Year 10	42.2	38.0	19.2	27.7
Source: Statistics Canada, 2020 Longitudinal Immigration Database.

The likelihood of GARs moving was higher among those designated to small metropolitan or urban communities, compared with those in larger gateway cities or medium-sized CMAs. For example, by the fifth year, about half (51%) of GARs designated to settle in a small urban community had moved, compared with 29% of those designated to a medium-sized CMA.

Economic outcomes in cities of different sizes

Employment incidence

Employment incidence, which is defined by the number of individuals with earnings over $500 in a year divided by the number of tax filers in the city in that year, was estimated for stayers in six different city sizes and for movers in different combinations of intended destination and current city type.

For stayers, employment incidence was lower among GARs than among PSRs. These employment gaps were reduced as time spent in Canada increased, but persisted nonetheless. This pattern was observed in all city sizes (Chart 2). It was also observed that movers had a slightly higher employment incidence than stayers. For example, five years after landing, GARs moving from their designated city had an employment incidence of 63%, compared with 59% among those who stayed in their designated city. This is a relatively small difference, but it may suggest that movers, who are often motivated by economic prospects, did better than if they had stayed in their designated city. However, there were far larger differences in employment incidence among city types than between movers and stayers. For instance, five years after landing, which may be enough time for some economic integration to have occurred, employment incidence among GAR stayers was lowest among stayers in Toronto and, to a lesser extent, Montréal (49% and 56%, respectively). Stayers in small urban communities, medium-sized CMAs and Vancouver had the highest employment incidence, at 72%, 63% and 65%, respectively (Chart 2). These patterns tended to hold 1 year and 10 years after landing as well (Table 2).

Data table for Chart 2



Data table for Chart 2
Table summary
This table displays the results of Data table for Chart 2 Government-assisted refugees, Privately sponsored refugees, Family class and Economic class, calculated using percent units of measure (appearing as column headers).

	Government-assisted refugees	Privately sponsored refugees	Family class	Economic class
	percent
Montréal	55.6	72.9	70.3	82.1
Toronto	48.8	65.5	75.0	81.1
Vancouver	65.1	69.5	79.6	76.0
Medium-sized metropolitan areas	63.4	81.2	79.0	89.2
Small metropolitan areas	56.2	70.4	78.3	84.5
Small urban areas	71.8	86.5	80.0	91.1
Source: Statistics Canada, 2020 Longitudinal Immigration Database.

Among GAR movers, five years after landing, those who had moved to a gateway city (either from another gateway city or a medium-sized metropolitan area) tended to have the lowest employment incidence, at around 50% (Table 2). Employment incidence was highest among GARs moving from a larger city to a smaller community. For instance, those who moved from a gateway city to a medium-sized CMA or smaller community had an employment incidence of 65%. The incidence of employment was even higher for those who moved from a medium-sized CMA to a smaller community, or from a small CMA to a small urban community (68% and 79%, respectively). GARs designated to Toronto and Montréal, as well as those moving from a smaller community to a gateway city, tended to have the lowest employment rates. However, it is important to note that not all individuals designated to a small urban community did exceptionally well; by the fifth year, one-half of GARs designated to a small urban community had moved, with the vast majority (88%) choosing to move to a larger community. The average employment incidence of these movers was 61%.^Note

There may be several reasons for the patterns observed above, including differences in the characteristics of the GARs designated to or moving to a community, and differences in the characteristics of the cities themselves. Multivariate analysis was conducted to assess the importance of these factors.



Multivariate analysis accounting for differences in employment incidence among city types

The discussion in this section focuses on results five years after landing since this may provide enough time for GARs to be somewhat integrated into the local economy. Model 1 results are identical to those reported above in Table 2, since they represent the actual data, but reported in a different way.^Note

Model 2 results indicate the employment incidence (relative to Toronto stayers) after accounting for the effects of the control variables. The control variables accounted for, or “explained,” most of the difference in employment incidence between Toronto stayers and the other categories. The differences between Toronto stayers and other locations were significantly reduced and often became statistically non‑significant^Note in Model 2 (Table 3). On average,^Note across all five stayer categories, the control variables accounted for 75% of the difference in employment incidence with Toronto. Among movers five years after landing, 10 of the 11 categories registered statistically significant higher incidence than stayers in Toronto in Model 1, but this was reduced to 5 of the 11 categories in Model 2 (Table 3). On average, the control variables accounted for 89%^Note of the difference in employment incidence between the 16 categories of stayers or movers and stayers in Toronto five years after landing.

It is important to understand which variables accounted for most of this effect. If the characteristics of the immigrants are most important in explaining employment outcomes, then choosing to designate refugees to smaller rather than larger communities will not in itself necessarily lead to better economic outcomes, since it is the characteristics of the refugees that matter most. Conversely, if the characteristics of the city are most important, then designating GARs to one type of city rather than another will make a significant difference in outcomes.

Additional decomposition analysis found that the largest part of the “explained” component (about three‑quarters of the difference between Toronto stayers and other city types) was attributable to the source region of the refugee. A significant portion was also attributable to the employment rate in that city. In some cases, the number of refugees designated to a community in any given year also contributed to explaining differences in employment incidence: the larger the number, the lower the employment success of the refugees.

Two examples of the decomposition are illustrated here.

In the first example, the employment incidence gap between stayers in Toronto and those in medium‑sized CMAs was among the largest gaps observed, at 14.6 percentage points. The control variables explained almost all this difference (13.2 percentage points). Differences in source regions between stayers in Toronto and those in medium-sized CMAs accounted for 7.4 percentage points of the gap. Differences in the city employment rate also contributed 7.0 percentage points to the gap in employment incidence. Differences in some of the other variables tended to increase, not reduce, the employment incidence gap, but these effects were small. Of the explained component, source region and city employment rate were by far the most important explanations. The results for Vancouver were similar.^Note

The second example is about the 23.0 percentage point gap in employment incidence between stayers in Toronto and small urban communities. Differences in source region accounted for 10.6 percentage points of this difference, and differences in the number of GARs entering the city in a given year accounted for another 7.3 percentage points. Source region was the most important explanatory variable. This inevitably raises the question of why source region was important. The regression results indicate that, adjusting for other immigrant and city characteristics, GARs from West Asia had the lowest employment incidence, and those from South America had the highest, 23.7 percentage points higher than that of GARs from West Asia. Almost three-quarters of Toronto stayers were from West Asia, compared with 12% of stayers in small urban communities. Conversely, one-third of small urban stayers were from South America, compared with 3% of those in Toronto. Differences such as these explained much of the gap in employment incidence between city types.



Annual earnings

On average, employed GARs had lower annual earnings (in 2020 constant dollars) than PSRs and economic class and family class immigrants. This was particularly true during the first year following landing but held for all years examined. It also held across all city types, as noted for the fifth year following landing (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 Government-assisted refugees, Privately sponsored refugees, Family class and Economic class, calculated using in 2020 dollars units of measure (appearing as column headers).

	Government-assisted refugees	Privately sponsored refugees	Family class	Economic class
	in 2020 dollars
Montréal	19,500	23,900	27,700	40,300
Toronto	22,500	26,600	32,200	46,100
Vancouver	25,200	27,200	34,300	44,900
Medium-sized metropolitan areas	26,000	30,600	38,400	52,100
Small metropolitan areas	24,500	28,500	36,400	49,900
Small urban areas	25,700	33,900	39,500	55,100
Source: Statistics Canada, 2020 Longitudinal Immigration Database.

Among employed GAR stayers, by the fifth year following landing, annual earnings among stayers in medium-sized and small communities surpassed those in the three gateway cities, particularly Toronto and Montréal. This is similar to the pattern observed for employment incidence. On average, employed GAR movers tended to have marginally higher earnings five years after landing than stayers, at $26,200 and $24,500, respectively. But again, there was far more variation among different city types than between movers and stayers. For instance, five years after landing, the average earnings among GAR stayers ranged from $19,500 in Montréal to $26,000 in medium-sized CMAs (Chart 4).

GAR movers to gateway cities, including those from other gateway cities or from medium-sized metropolitan areas, had lower earnings than other movers, at around $21,000 five years after landing (Table 4). Movers who did relatively well by the fifth year tended to be those who moved down in community size, possibly because they were moving to a better-paying job. This included movers from a gateway community to a medium-sized CMA, from a medium-sized CMA to a smaller community, and from a small CMA to a small urban community (Table 4). Those who moved but remained in a medium‑sized CMA or a small urban centre also had relatively high earnings. These results resemble the patterns observed with employment incidence among GARs: those staying in smaller communities tended to do better than those staying in large gateway communities; movers tended to do marginally better than stayers, on average; and those moving down in community size tended to do better than those moving to the gateway cities.

Data table for Chart 4



Data table for Chart 4
Table summary
This table displays the results of Data table for Chart 4 Year 1, Year 5 and Year 10, calculated using in 2020 dollars units of measure (appearing as column headers).

	Year 1	Year 5	Year 10
	in 2020 dollars
Montréal	10,400	19,500	25,600
Toronto	12,700	22,500	27,500
Vancouver	13,900	25,200	30,600
Medium-sized metropolitan areas	12,800	26,000	33,600
Small metropolitan areas	12,900	24,500	31,100
Small urban areas	12,700	25,700	32,800
Source: Statistics Canada, 2020 Longitudinal Immigration Database.



Multivariate analysis accounting for differences in earnings among city types

The details of the multivariate regression approach are described in the Methods section. In the analysis that follows, two models were run. Model 1 results represent those observed in the actual data. Regarding stayers, comparisons between Model 1 and Model 2 results suggest that in most cases, adding the control variables (Model 2) accounted for relatively little of the actual difference in earnings between stayers in Toronto (the reference group) and those in other city types.^Note On average,^Note across the five stayer categories, five years after landing, the control variables accounted for 37% of the difference in annual earnings between different city types and Toronto. Regarding movers five years after landing, adjusting for the effects of the control variables in most cases reduced the earnings gap with Toronto substantially, at times to zero. The control variables were more important in explaining the earnings gaps with Toronto stayers for movers than for stayers. On average, across all 16 mover and stayer categories, adjusting for the control variables accounted for 47%^Note of the unadjusted earnings gap with Toronto stayers.

The decomposition results indicate that where the control variables explained a significant proportion of the earnings gaps with Toronto stayers, source region and the city’s employment rate played the key role. This result was similar to that observed for employment incidence. For example, differences in source regions between stayers in medium-sized metropolitan areas and Toronto would account for $2,700 of the earnings gap between those city types, and differences in regional employment rates would account for $1,800 of the gap. A number of other variables tended to increase the gap between stayers in Toronto and medium-sized CMAs, and thus offset some of the effects of source region and regional employment rates. Overall, the control variables, mainly source region and regional employment rate, accounted for $1,900 (or 54%) of the $3,500 gap.

Decompositions were also conducted for some mover categories, and the results were similar. For example, movers from gateway cities to medium-sized CMAs earned $5,500 more than GARs who stayed in Toronto, $3,400 (or 62%) of which was accounted for after adjusting for differences in the control variables (i.e., Model 2). Differences in the source region of these movers, compared with those who stayed in Toronto, accounted for $2,700 of the gap. About 67% of stayers in Toronto were from West Asia, compared with 19% among those who moved from a gateway city to a medium-sized CMA. GARs from West Asia had the lowest annual earnings, at $2,200 to $7,200 less than those from other regions, even after adjusting for differences in immigrant characteristics (e.g., education and age) and city-level characteristics (e.g., economic conditions and the existence of own-group enclaves). GARs from Africa had among the highest annual earnings, at $5,300 more than those from West Asia. An estimated 51% of the movers in this example were from Africa, compared with 11% of those staying in Toronto. These differences contributed significantly to the earnings gap. Additionally, differences in the employment rate between the CMAs to which the movers went and the Toronto region, also contributed to the earnings gap, by $2,200. Again, these effects were offset by the effect of some other control variables, but none of them were large. The decomposition of the earnings of movers from a medium-sized CMA to another medium-sized CMA, compared with GARs who stayed in Toronto, produced similar results. Differences in source region and the city employment rate accounted for most of the observed earnings gap.



Summary and conclusion

This study sought to determine whether there were significant variations in the economic outcomes of GARs who were designated to cities of different sizes. These outcomes were also compared between GARs who remained in their designated cities and those who undertook secondary migration and resettled elsewhere.^Note Additionally, the study aimed to determine the extent to which these differences in economic outcomes could be accounted for by the sociodemographic characteristics of GARs, as well as by the conditions of the designated or chosen city. The analysis focused on GARs aged 20 to 54 at entry who landed in Canada from 2000 to 2019.

The study observed that compared with PSRs and family and economic class immigrants, GARs were less likely to have one of the three gateway cities as their initial destination, and were more likely to be resettled to smaller communities. Over the study period, GARs had lower levels of employment incidence and annual earnings than PSRs or the family and economic classes. These gaps were reduced with years spent in Canada but persisted to at least 10 years after landing.

Among GAR stayers (those who remained in their designated city), those who were assigned to small urban communities had the highest employment incidence and annual earnings, while those who were assigned to Toronto had the lowest economic outcomes. Moreover, GAR stayers in medium-sized or small communities had stronger labour market outcomes than GAR stayers in Toronto. Among GAR movers (those who moved away from their intended destination), those who moved to a gateway city had the lowest employment incidence and annual earnings. Most notably, GARs who moved down in city size tended to have better economic outcomes than other movers. Nevertheless, city size was still negatively correlated with out-migration, with the likelihood of moving increasing as the size of the city decreased.

The regression analysis indicated that five years after landing, differences in individual sociodemographic characteristics and city-level characteristics accounted for the majority (89%) of the observed differences in employment incidence between Toronto GAR stayers and other GARs. Of the many explanatory variables in the regression, the source region of the GARs and the employment rate in the city were by far the most important in explaining the differences with Toronto GAR stayers in employment incidence. In particular, compared with other stayers and movers, a very high percentage of GARs designated to Toronto and who chose to stay there were from West Asia, which was the group with the lowest employment incidence even after adjusting for differences in age, education, city economic conditions and other explanatory variables. The regression results for annual earnings illustrated that differences in sociodemographic and city characteristics explained less (47% on average) of the difference in earnings between Toronto stayers and movers in other cities. Thus, there are other, unobserved factors that could explain these differences, but could not be captured in this analysis.

Since the differences in earnings across city sizes could not be fully explained in this study, there may be partial explanations as to why these variations may exist. In particular, the apparent success of smaller cities compared with Toronto and Montréal may relate to a phenomenon whereby those unable to find well-paid employment in small communities tend to move to larger centres, resulting in a higher employment rate and earnings among the stayers in the smaller communities. As supported by this study and existing literature, the observed preference that refugees have for larger cities could mean that even those who are struggling economically will still prefer to remain in a large city such as Toronto, and that those who struggle in smaller communities would prefer to move there, as well. Thus, the implication is that GARs who remain in small communities are the ones who are able to integrate economically.

The findings from this study may inform policy discussions in various ways. In terms of employment prospects, the study’s findings suggest that it is less about the size of the city to which refugees are assigned, but rather the strength of the labour market, as indicated by the regional employment rate. Hence, consideration of regional economic conditions should be an integral part of GAR resettlement decisions. Additionally, the findings in this study indicate that despite the general preference among GARs to resettle in large cities, relatively strong economic outcomes can be achieved in smaller communities, which can in turn increase retention rates.

References

Abu-Laban, B., Derwing, T., Krahn, H., Mulder, M., & Wilkinson, L. (1999). The settlement experience of refugees in Alberta: a study prepared for Citizenship and Immigration Canada. Edmonton: Prairie Centre of Excellence for Research on Immigration and Integration and Population Research Laboratory.

Derwing, T. M., & Krahn, H. (2008). Attracting and retaining immigrants outside the metropolis: Is the pie too small for everyone to have a piece? The case of Edmonton, Alberta. Journal of International Migration and Integration, 9(2), 185–202.

Gure, Y., & Hou, F. (2022). Retention of government-assisted refugees in designated destinations: Recent trends and the role of destination characteristics (statcan.gc.ca). Economic and Social Reports. Statistics Canada.

Hiebert, D. (2015). Ethnocultural minority enclaves in Montreal, Toronto and Vancouver. Institute for Research on Public Policy, 52, 1–52.

Hou, F. (2007). Changes in the initial destinations and re-distribution of Canada’s major immigrant groups: Re-examining the role of group affinity. International Migration Review, 41(3), 680–705.

Hyndman, J., Schuurman, N., & Fiedler, R. (2006). Size matters: Attracting immigrants to Canadian cities. Journal of International Migration and Integration, 7(1), 1–25.

IRCC [Immigration, Refugees and Citizenship Canada]. (2020). Departmental Plan 2020/2021. Government of Canada.

IRCC [Immigration, Refugees and Citizenship Canada]. (2016). The Matching Centre. Government of Canada.

IRCC [Immigration, Refugees and Citizenship Canada]. (2011). Evaluation of Government Assisted Refugees (GAR) and Resettlement Assistance Program (RAP). Government of Canada.

Kaida, L., Hou, F., & Stick, M. (2020). Are refugees more likely to leave initial destinations than economic immigrants? Recent evidence from Canadian longitudinal administrative data. Population Space and Place, 26(5),1–14.

Nogle, J. (1994). Internal migration for recent immigrants to Canada. International Migration Review, 28(1), 31–48.

Qadeer, M., & Kumar, S. (2006). Ethnic enclaves and social cohesion. Canadian Journal of Urban Research, 15, 1–17.

Sherrell, K., Hyndman, J., & Preniqi, F. (2005). Sharing the wealth, spreading the “burden”? The settlement of Kosovar refugees in smaller British Columbia cities. Canadian Ethnic Studies, 37(3), 76– 96.

Simich, L., Beiser, M., & Mawani, F. (2002). Paved with good intentions: Canada’s refugee destining policy and paths of secondary migration. Canadian Public Policy, 28(4), 597–607.

Statistics Canada. (2020). Longitudinal Immigration Database (IMDB) Technical Report, 2018 (Diversity and Sociocultural Statistics, No. 24).