Economic and Social Reports
Poverty among racialized groups across generations

Abstract

This study uses data from Canada’s 2021 Census of Population to examine the differences between the poverty rates of racialized groups and the White population. The analysis examines whether these differences recede or persist across generations and the extent to which the sociodemographic composition of racialized groups explains these differences. Among the 11 racialized groups compared, 10 had a higher poverty rate than the White population. For all but the Japanese group, compositional differences did not entirely explain the difference in the poverty rate from the White population, although the size of the difference generally decreased after considering compositional differences. The Filipino had a lower poverty rate than the White population in the first generation, second generation, and third generation or more of Canadians. The differences in the poverty rate between most racialized groups and the White population persisted into the second generation. For South Asian, Chinese and Japanese groups, the poverty rate difference from the White population receded in the third generation or more. For Black, Latin American, Arab, and West Asian groups, the difference in the poverty rate persisted into the third generation or more.

Keywords: poverty, market basket measure, racialized groups, generation status

Authors

Christoph Schimmele, Feng Hou and Max Stick are with the Social Analysis and Modelling Division, Analytical Studies and Modelling Branch, at Statistics Canada.

Acknowledgments

The authors thank Rubab Arim and Xuelin Zhang for their advice and comments on an earlier version of this paper.

Introduction

The socioeconomic outcomes of recent cohorts of immigrants and their children will determine the long-term impact of non-European immigration and demonstrate the capacity of Canadian society to incorporate newcomers (Reitz & Somerville, 2004; Zhou & Gonzales, 2019). Recent studies have reported disparities in employment income and an uneven distribution of wealth across racialized groups in Canada (Banting & Thompson, 2021; Qiu & Schellenberg, 2022). One of the primary indicators of racial inequality is the persistence of disparities in the poverty rate between groups (Baker et al., 2022; Banting & Thompson, 2021). A long-standing concern is whether these inequalities endure across generations and become a permanent feature of stratification (Hou & Balakrishnan, 1996).

Poverty rates (or low-income rates, depending on the measure used)^Note are higher for most racialized groups than for the White population, and this disparity persists across generations (Banting & Thompson, 2021). Furthermore, the size of the difference from the poverty rate of the White population varies across racialized groups (e.g., Black, Chinese, South Asian) (Block, Galabuzi, & Tranjan, 2021). These disparities in poverty rates are at least partially linked to disadvantages in employment and earnings (Block, Galabuzi, & Tranjan, 2021; Chen & Hou, 2019).

Differences in the sociodemographic composition of racialized groups are expected to account for some of the variation in poverty rates. For example, most racialized people are immigrants, and a large proportion are recent arrivals. This contributes to the difference in the poverty rate from the White population because first-generation status and recent arrival in Canada increase the risk of poverty (Crossman, 2013; Murphy, Zheng, & Dionne, 2012). In addition, a large proportion of Canadian-born racialized people are children and youth who live with their parents, so their poverty rates are linked to their immigrant parents. Other compositional factors that may account for differences in poverty rates include the share of households with a university-level education, the number of earners in the family and the share of one-parent families.

Using Canada’s Official Poverty Line (defined below), this study examines generational differences in the poverty rate between 11 racialized groups and the White population. This study also considers the extent to which group differences in sociodemographic characteristics contribute to the observed poverty differences with the White population in the first generation, second generation and third generation or more of Canadians. This analysis disentangles the independent role of racialized status from other sociodemographic predictors of poverty.

Study background

Assessments of the socioeconomic success of immigrants (first-generation Canadians) and their Canadian-born children (second-generation Canadians) are based on achieving parity with the third generation or more of the White population, the predominant group in Canada and other settler societies (Park & Myers, 2010; Zhou & Gonzales, 2019). For immigrants, the imperfect transferability of foreign education and work experience, deficiencies in literacy skills, and unfamiliarity with local labour markets are explanations for a lack of economic parity with Canadian-born people (Bonikowska, Green, & Riddell, 2008; Chiswick & Miller, 2002; Skuterud, 2010). These are not plausible explanations for disparities in the second generation and third generation or more because they were educated and socialized in Canada. Disparities that persist in the second generation and third generation or more corroborate the association between racialized status and economic inequality (Park & Myers, 2010).

From 1980 to 2000, the low-income rate of immigrants increased, and, while this rate declined in the 2000s, the gap between them and Canadian-born people continued to grow (Picot & Hou, 2003, 2014). This coincided with the shift in source regions from Europe and the United States to Asia, Africa, and South and Eastern Europe (Picot & Hou, 2003), and the rapid increase in Canada’s ethnocultural diversity after 1980 (Banting & Thompson, 2021). The increase in the low-income rate of immigrants from 1980 to 2000 cannot be attributed to a deterioration of macroeconomic conditions, because the low-income rate of Canadian-born people declined over this period. Low income has become a more acute and persistent experience for post-1980s immigrants. Among immigrants who have been in Canada for five years or less, the low-income rate increased from 1.4 times that of Canadian-born people in 1980 to 2.5 times in 2000 (Picot & Hou, 2003). Immigrants in Canada aged 6 to 20 years had low-income rates similar to or lower than those of Canadian-born people in 1980, but relatively higher rates in 2000.

This suggests that post-1980s immigrants have encountered increased difficulties in economic adjustment (Picot & Hou, 2014). The gap in the low-income rate between immigrants and the Canadian-born population rose, despite a large increase in immigrants’ average levels of education. The proportion of immigrants aged 25 to 65 with a university education in 2000 was more than double that in 1980 (Picot & Hou, 2003). Because low-income rates are lower for university-educated people, the increase in the low-income rates of immigrants from 1980 to 2000 would have presumably been higher without the large shares of those with a university education, although higher education reduces the risk of poverty less for immigrants than for Canadian-born people (Crossman, 2013).

Variation in low income across racialized groups has been observed since at least the 1991 Census (Crossman, 2013; Kazemipur & Halli, 2001). Data from the 2001 Census showed that the Arab and West Asian groups had a low-income rate of 35%, the East and Southeast Asian group had a rate of 22%, and the South Asian group had a rate of 19% (Kaida, 2015). In 2015, the low-income ratio of first-generation groups with the third generation or more of the White population ranged from 1.4 times to 3 times higher, although for the first-generation Filipino population it was lower (Banting & Thompson, 2021). Among the first-generation White population, the low-income ratio was slightly higher (1.1 times), providing a referent for gauging the role of generation status versus racialized status in low-income disparities. These disparities are consistent with a broad pattern of socioeconomic inequality.

Among racialized groups, the correlation between age and poverty has implications for differences within and across generations (Crossman, 2013). In 2021, about one-half of racialized people in the second generation were aged 0 to 14 years and one-fifth were aged 15 to 24 years.^Note This ties the poverty rate of a large proportion of the second generation directly to the poverty rate of the first generation (i.e., their parents). Accordingly, the age structure of racialized groups in the second generation is bound to have a large impact on their poverty disparity with the White population.

Given these trends, it is important to examine whether poverty rates decrease for racialized groups across generations, how their poverty rates compare with those of the White population, and the extent to which sociodemographic factors can explain poverty disparities between racialized groups and the White population.

Data and methods

This study uses microdata from the 2021 Canadian Census of Population long-form sample (25% of the population). The census provides comprehensive sociodemographic information for a very large, nationally representative sample, allowing disaggregated analyses of the poverty rates of racialized groups by generation status. Because this article focuses on racialized groups, all tabulations exclude individuals who self-identified as Indigenous people. The study excludes residents of collective dwellings and temporary foreign residents.^Note The analysis also excludes residents of the territories because the poverty measure was not developed for those areas in the 2021 Census.

The measure of poverty is based on Employment and Social Development Canada’s market basket measure (MBM, 2018 base), Canada’s Official Poverty Line. MBM thresholds are based on “the cost of a specific basket of goods and services representing a modest, basic standard of living” (Statistics Canada, 2015). The basket consists of specified qualities and quantities of food, clothing, footwear, transportation, shelter and other expenses for a reference family of two adults and two children. The MBM basket is delineated for 53 different communities, considering potential differences in the cost of the basket between similar-sized communities in different provinces, and between different geographic regions within provinces (Statistics Canada, 2022). For instance, in 2020 (the income reference year for the 2021 Census), the threshold of disposable family income^Note was $49,727 for the Toronto census metropolitan area (CMA) and $38,685 for population centres with a population of 30,000 to 99,999 people in Quebec (Statistics Canada, 2022). People with disposable family income below MBM thresholds are considered to be in poverty.

This study compares the poverty rate across 11 racialized groups^Note (South Asian, Chinese, Black, Filipino, Latin American, Arab, Southeast Asian, West Asian, Korean, Japanese and other racialized groups^Note ) with the White population group. The poverty rates for these groups are estimated for the group as a whole and separately for the first generation, second generation and third generation or more. The first generation refers to foreign-born individuals (excluding temporary residents). The second generation refers to those who were born in Canada, with at least one foreign-born parent. The third generation or more refers to those who were born in Canada, with all parents Canadian born.

This study further examines the extent to which the observed differences in the poverty rate are associated with differences in sociodemographic characteristics. For this purpose, linear probability models^Note are estimated, with poverty (being poor = 1, otherwise = 0) as the outcome, and include racialized groups as the focal independent variable. For the total population (i.e., all generations) and each generation, the models control for sex, age,^Note education,^Note language,^Note generational co-residence pattern,^Note household type,^Note number of earners in the economic family,^Note population size of the economic family^Note and geographic distribution.^Note For the total population, the models control for generation status. For the first generation, the model controls for years since immigration.^Note For the second generation, the model controls for whether just one parent or both parents were born outside Canada. The regression models are presented in Appendix Table A.1.

Based on the model estimates, the adjusted poverty rates are estimated for each racialized group, assuming that each group had the same distributions in the selected sociodemographic characteristics as the White group in the total population and within each generation. The difference between the observed gap in the poverty rates of a racialized group and the White group and the corresponding adjusted gap reflects the portion of the observed gap that is accounted for by group differences in sociodemographic factors.

The portion of the observed gap accounted for depends on whether there are large differences between a racialized group and the White group in some sociodemographic factors, and whether these factors are strong correlates of poverty. If a factor is positively associated with poverty and a racialized group has a higher value in the factor than the White group, this factor would reduce the observed gap (i.e., account for part of the observed gap). If a racialized group has a lower value in this factor than the White group, this factor would increase the observed gap (i.e., not account for any of the observed gap). Conversely, if a factor is negatively associated with poverty and a racialized group has a higher value in the factor than the White group, this factor would increase the observed gap. If a racialized group has a lower value in this factor, the factor would reduce the observed gap. For a given racialized group, it is possible that some factors tend to reduce the observed gap and others tend to increase it. These potentially offsetting effects of different factors jointly determine whether, and how much of, the observed gap can be accounted for. A decomposition analysis is used to examine the relative roles of the selected sociodemographic factors in accounting for the observed gap (Hou, 2014).

Results

Observed poverty rates of racialized groups by generation status

The poverty rate differed across racialized groups (Table 1, observed rates, first column), and there was wide variation in the difference from the White group. While the Filipino group (3.9%) had a lower poverty rate than the White group (6.1%), all other racialized groups had higher rates, and the Arab group had the highest poverty rate. For the South Asian and Japanese groups, the difference from the White group was small—around 1 percentage point. The Black, Latin American and Southeast Asian groups had a poverty rate that was 3.2 to 4.3 percentage points higher than that of the White group. The Arab, West Asian, Korean and Chinese groups had a rate over twice that of the White group.



Given that most racialized groups consist primarily of first-generation Canadians, the pattern of group differences in the poverty rate among the first generation was similar to that of the total population (all generations), although for most groups, the poverty rates were somewhat higher for the first generation than the overall rate for their group (Table 1, observed rates, second column). Among the second generation, most racialized groups had a higher poverty rate than the White group. Only the second-generation Filipino population (4.0%) had a lower poverty rate than the second-generation White population (5.9%). The Arab, Korean and West Asian groups had the highest poverty rates (10.5% to 11.6%) observed in the second generation. Among the third generation or more, the poverty rates of the Black, Latin American and West Asian groups were over twice as high as that of the White group.

The poverty rate was lower in each successive generation for some racialized groups, but not for others. Compared with the first generation, the South Asian, Chinese, Arab, Korean and Japanese groups had a lower poverty rate in the second generation and third generation or more. For example, among the Chinese group, the poverty rate was 6 percentage points lower in the second generation (8.3%) and 9 points lower in the third generation or more (5.1%), compared with the rate for the first generation (14.1%). Conversely, among the Black, Latin American and West Asian groups, the third generation or more had the highest poverty rate of all generations.^Note Among the White group, the first generation had a slightly higher poverty rate than the second generation and third generation or more.

Some racialized groups achieved parity with the White group by the third generation or more, but for other groups, the poverty rate was higher across generations. Except for the Filipino group, no racialized group achieved parity with the White population in the second generation. In the third generation or more, the South Asian, Chinese, Filipino and Japanese groups had poverty rates similar to or lower than that of the third generation or more White group. The poverty rate was higher for the Black, Latin American, Arab, West Asian, Korean and Southeast Asian groups in each generation.

To what extent did sociodemographic characteristics account for gaps in the poverty rate between racialized groups and the White population overall?

The gap in the poverty rate between each racialized group and the White population could be partly explained by the sociodemographic differences. There were large compositional differences between racialized groups and the White group in terms of age structure, shares of first-generation Canadians and shares not having English or French as a mother tongue (Table 2). For instance, the share of people younger than 25 years was particularly high among the Black (42%) and Arab (42%) groups, compared with the White group (25%). The share of first-generation Canadians ranged from 56% for the Black group to 79% for the West Asian group, compared with 11% among the White group.



There was also large variation in family structure, number of earners in the family and educational attainment across racialized groups. For instance, almost 21% of the Black group consisted of one-parent families, compared with 9% among the White group. The share of one-parent families ranged from 5% to 12% among other groups. About 81% of the Filipino group and 73% of the South Asian group lived in families with two or more earners, compared with 55% of the White group. Having more earners in the family was strongly associated with lower poverty rates.

For the South Asian group, a higher share of immigrants, a higher share of people without English or French as a mother tongue, a higher share of children and youth and a relatively high concentration in the three largest CMAs were factors associated with their higher poverty rate. But the negative effect of these factors was offset by the group’s larger number of earners in the family and higher educational level. As a result, the selected sociodemographic characteristics jointly accounted for little of the observed gap between their poverty rate and that of the White population, as the adjusted gap was the same as the observed gap (Table 1).

For the Filipino group, the positive effect of a larger number of earners and a higher educational level more than offset the disadvantages associated with age structure, immigrant status and language profile. If the Filipino group had a similar sociodemographic profile as the White group, its poverty rate would be about 1 percentage point higher than the observed rate for the group (Table 1, adjusted rates). The differences in sociodemographic profiles accounted for about 40% of the Filipino group’s lower poverty rate, relative to the White group.

For other racialized groups, compositional differences in age structure, immigrant status, language use and geographic distribution tended to offset a small advantage in the number of earners in the family. The differences in selected sociodemographic characteristics together accounted for about one-third of the observed gap in the poverty rate for the Arab, West Asian and Korean groups; one-half of the gap for the Chinese, Black and Latin American groups; close to two-thirds of the gap for the Southeast Asian group; and the entire gap for the Japanese group.^Note

Overall, even with a similar sociodemographic profile, most racialized groups had higher poverty rates than the White group (Table 1, adjusted rates). The Japanese and Filipino groups were the only ones that had an adjusted poverty rate similar to or lower than the White group. For other groups, the adjusted gap in the poverty rate with the White group ranged from a low of +1 percentage point for the South Asian group to a high of over +5 percentage points for the Arab group.

To what extent did sociodemographic characteristics account for group differences in the poverty rate among the first generation?

Sociodemographic composition played different roles in accounting for the gaps in the poverty rate between racialized groups and the White group in the first generation. For the first-generation South Asian, Southeast Asian and Latin American groups, sociodemographic characteristics did not account for any of the gaps as their observed gaps with the White population did not decrease when compositional differences were taken into account (Table 1). Relative to the White population, these groups had more earners in the family—a sociodemographic factor that reduces poverty. For instance, 72% of first-generation South Asian families had at least two earners, compared with 46% among White families (Table 3). If they did not have this advantage, their poverty rates would have been higher. The difference in sociodemographic factors, mostly having more earners in the family, accounted for 80% of the Filipino group’s lower rate relative to the White population.



For other racialized groups, disadvantages associated with higher shares of children and youth, recent immigrants and people without English or French as a mother tongue suppressed any advantages related to a larger number of earners in the family or higher education. Together, the selected sociodemographic characteristics accounted for about one-quarter (for the Korean group) to three-quarters (for the Japanese group) of the observed gap with the first-generation White population (Table 1).

The first-generation Filipino group had a slightly lower adjusted poverty rate than the first-generation White group (Table 1, adjusted rate). The Japanese, South Asian, Black, Latin American and Southeast Asian groups had adjusted poverty rates that ranged from 7.6% to 10.5%, about 0.6 to 3.5 percentage points higher than the rate of the first-generation White population (7.0%). The first-generation Chinese, Arab, West Asian and Korean groups had adjusted poverty rates that were about 4 to 6 percentage points higher than the rate of the first-generation White group.

To what extent did sociodemographic characteristics account for group differences in the poverty rate among the second generation?

Among the second generation, differences in age structure were a key factor associated with the gaps in the poverty rate for racialized groups. People younger than 25 accounted for two-thirds to three-quarters of most second-generation groups, and as much as 88% of the West Asian group, compared with one-quarter of the White group (Table 4). Conversely, except for the Japanese group, the share of people aged 65 years or older was 1% or less for second-generation racialized groups, compared with 21% for the White group. For the second generation as a whole, the poverty rate was 8.0% for those younger than 25, 6.1% for those aged 25 to 64 and 3.7% for those aged 65 years or older. Thus, second-generation racialized groups were expected to have higher poverty rates than the White group because of their younger age structures.



Except for the Filipino group, the disadvantage associated with a higher share of children and youth offset any advantages, such as a larger number of earners and higher family educational level, which second-generation racialized groups had over the White group. Consequently, among the second generation, the selected sociodemographic characteristics jointly accounted for about 40% of the gap between the White group and the Arab and Korean groups; about 60% of the gap between the White group and the Chinese, Southeast Asian, Latin American and West Asian groups; and 80% of the gap between the White group and the South Asian, Black and Japanese groups.

About 82% of second-generation Filipino people lived in families with at least two earners, the highest share among all second-generation racialized groups, and much higher than the share for the second-generation White population (56%, Table 4). This large advantage in the number of earners, as well as a higher educational level, mostly offset the disadvantage in the age structure relative to the White group. As a result, sociodemographic factors accounted for about 30% of the second-generation Filipino group’s lower poverty rate relative to that of the White group.

Overall, when sociodemographic composition was accounted for, the second-generation Filipino group had a lower adjusted poverty rate than the White group (Table 1, adjusted rate). The second-generation South Asian, Chinese, Black, Latin American and Southeast Asian groups had adjusted poverty rates similar to or up to 1 percentage point higher than the rate of the White group. The gaps in the adjusted poverty rate relative to the White group were larger for the second-generation Korean (3.2 points), Arab (2.6 points) and West Asian (2.0 points) groups.

To what extent did sociodemographic characteristics account for group differences in the poverty rate among the third generation or more?

Similar to the age structure of the second generation, third generation or more racialized groups (except the Japanese group) consisted mostly of individuals younger than 25, with the share of this age group in the population ranging from 63% (West Asian) to 89% (Filipino), compared with 27% of the White group (Table 5). However, the differences in the poverty rate by age group (6.7% for those younger than 25, 6.7% for those aged 25 to 64 and 3.5% for those aged 65 or older) in the third generation or more were smaller than in the second generation. As discussed, a factor’s effect on the poverty gap between racialized groups and the White group depends jointly on how strongly the factor is associated with the poverty rate and the size of the difference in the sociodemographic factor between the two groups. For these reasons, the effect of age structure on the poverty gap would be smaller in the third generation or more than in the second generation.



In addition to having a large share of children and youth, the third generation or more Black, Latin American and West Asian groups had relatively large shares of one-parent families (28%, 26% and 20%, respectively)—much higher than the White group (9%) and other groups (Table 5). They also lacked the advantage over the White group in the number of earners and family educational level that other groups had. Differences in sociodemographic characteristics accounted for two-thirds to three-quarters of the observed gaps between those groups and the White group.

Differences in sociodemographic characteristics accounted for little of the observed differences in the poverty rates for the third generation or more South Asian, Chinese, Filipino, Korean and Japanese groups relative to the White group. Their disadvantages in age structure were offset by the higher number of earners in the family and higher educational level relative to the White group. For the Arab and Southeast Asian groups, differences in sociodemographic characteristics—mostly a larger share of children and youth—accounted for about two-thirds of the gap between their poverty rate and that of the White group.

The third generation or more South Asian, Chinese, Filipino, Japanese and Southeast Asian groups had adjusted poverty rates lower than or close to the rate for the White group (Table 1, adjusted rates, last column). The Korean, Black, Arab groups had adjusted poverty rates about 1 percentage point higher than the White group, while the poverty rate was 2 percentage points higher for the Latin American group and 4 points higher for the West Asian group.

Conclusion

This study compared the poverty rates of 11 racialized groups with that of the White population group, based on the 2021 Census of Population. The information on the differences in poverty rates and how these vary across generations highlights the diversity in economic well-being and some underlying sociodemographic factors that account for these differences.

Overall, poverty was more prevalent for most racialized groups than the White population. Of the 11 racialized groups, 10 had higher poverty rates than the White group, although the gaps between rates were narrower for some groups (e.g., South Asian and Japanese) and wider for others (e.g., Arab, West Asian and Korean), before and after controlling for sociodemographic characteristics. Differences with the White group are summarized as follows:

• The Filipino group had a lower poverty rate than the White group in the first, second and third or more generations
• The South Asian group had a slightly higher poverty rate than the White group in the first and second generations and a similar rate of poverty in the third generation or more.
• The Chinese and Japanese groups had higher poverty rates than the White group in the first generation, but the differences declined in the second generation and disappeared (and reversed) in the third generation or more.
• The Arab and Korean groups had higher poverty rates than the White group in the first, second and third or more generations, although the gaps between their poverty rates and that of the White group decreased in each successive generation.
• The Southeast Asian group had higher poverty rates than the White group in the first, second and third or more generations. While the gap decreased from the first to the second generation, there was no further decrease in the third generation or more.
• The Black, Latin American and West Asian groups had higher poverty rates than the White group in the first, second and third or more generations. Decreases in their poverty rates were observed from the first to the second generation, but the gaps relative to the White group were largest in the third generation or more.

Compositional differences in the sociodemographic characteristics of racialized groups accounted for some of the observed differences in the poverty rate with the White population. The Filipino group had the highest number of earners in the family in all three generations, and this advantage tended to offset the disadvantage in immigrant status and age structure. The advantage in the number of earners in the family was also the key factor for the relatively lower poverty rates among first- and second-generation South Asian people. The gaps observed for the Chinese, Korean and Japanese groups were partially related to the recency (e.g., period of immigration and language use) of their first generations and the young age of their second generations. For the Arab group, compositional differences—mostly age structure and the recency of the first generation—accounted for about one-third to two-thirds of the gaps in the poverty rate with the White population, depending on the generation. Differences in sociodemographic characteristics accounted for two-thirds to three-quarters of the large gaps with the White group of third generation or more Black, Latin American and West Asian people. One factor common to these three population groups was their higher shares of one-parent families than other groups.

The results in this article at least partly reflect the special circumstances of 2020—the year when the poverty status was measured for the 2021 Census. The national poverty rate in 2020 was 44% lower than the 2015 rate. This decrease was driven largely by higher government transfers in 2020, including the enhanced Canada Child Benefit (CCB) and temporary COVID-19 pandemic relief benefits (Bernard & Zhang, 2022). The decrease was even larger among some racialized groups, including the Filipino, West Asian, Arab and Black groups. Clearly, the enhanced CCB and temporary pandemic relief benefits affected various racialized groups differently, likely because of their different demographic compositions and patterns of labour market engagement. Nevertheless, the general patterns of group differences in the poverty rate relative to the White population across generations observed in the 2021 Census were broadly similar to those observed in the 2016 Census (detailed tables for the 2016 Census not shown here), although the magnitudes of these differences changed.

In sum, racialized groups differed considerably in their poverty experiences and the associated factors. The effect of some influential factors, such as recency among the first generation and the very young age structure among the second generation and third generation or more of racialized groups, may diminish over time. The effect of other influential factors, such as one-parent families and disadvantages in the labour market, requires more concrete efforts to overcome.

Appendix



References

Baker, R. S., Brady, D., Parolin, Z., & Williams, D. T. (2022). The enduring significance of ethno-racial inequalities in poverty in the U. S., 1993 – 2017. Population Research and Policy Review, 41, 1049–1083.

Banting, K., & Thompson, D. (2021). The puzzling persistence of racial inequality in Canada. Canadian Journal of Political Science, 54(4), 870–891.

Bernard, A., & Zhang, X. (2022). Disaggregated trends in poverty from the 2021 Census of Population. Census in Brief. Statistics Canada, Catalogue no. 98-200-X, issue 2021009.

Block, S., Galabuzi, G-E., & Tranjan, R. (2021). Canada’s colour coded income inequality. Directions: Research and Studies on Race Relations in Canada, 9(Winter), 63–80.

Bonikowska, A., Green, D. A., & Riddell, W. C. (2008). Literacy and the labour market: Cognitive skills and immigrant earnings. Statistics Canada, Catalogue no. 89-552-M – No. 020.

Chen, W-H., & Hou, F. (2019). Intergenerational education mobility and labour market outcomes: Variation among the second generation of immigrants in Canada. Statistics Canada, Analytical Studies Branch Research Paper Series. Catalogue no. 11F0019M – No. 418.

Chiswick, B. R., & Miller, P. W. (2002). The complementarity of language and other human capital: Immigrant earnings in Canada. IZA Discussion Papers, No. 451. Bonn: Institute for the Study of Labor.

Crossman, E. (2013). Low-income and immigration: An overview and future directions for research. Ottawa: Citizenship and Immigration Canada.

Government of Canada. (2018). Opportunity for all: Canada’s first poverty reduction strategy. Ottawa: Employment and Social Development Canada.

Hou, F. (2014). A general approach to effect decomposition. Social Science Quarterly, 95(3), 894–904.

Hou, F., & Balakrishnan, T. R. (1996). The integration of visible minorities in contemporary Canadian society. Canadian Journal of Sociology, 21(3), 307–326.

Kaida, L. (2015). Ethnic variations in immigrant poverty exit and female employment: The missing link. Demography, 52(2), 485–511.

Kazemipur, A., & Halli, S. (2001). Immigrants and the “new poverty”: The case of Canada. International Migration Review, 35(4), 1129–1156.

Murphy, B., Zheng, X., & Dionne, C. (2012). Low income in Canada: A multi-line and multi-index perspective. Statistics Canada, Income Research Paper Series. Catalogue no. 75F0002M – No. 001.

Park, J., & Myers, D. (2010). Intergenerational mobility in the post-1965 immigration era: Estimates by an immigrant generation cohort method. Demography, 47(2), 369–392.

Picot, G., & Hou, F. (2003). The rise in low-income rates among immigrants in Canada. Statistics Canada, Analytical Studies Branch Research Paper Series. Catalogue no. 11F0019M – No. 418.

Picot, G., & Hou, F. (2014). Immigration, low income and income inequality in Canada: What’s new in the 2000s? Statistics Canada, Analytical Studies Branch Research Paper Series. Catalogue no. 11F0019M – No. 364.

Qiu, T., & Schellenberg, G. (2022). The weekly earnings of Canadian-born individuals in designated visible minority and White categories in the mid-2010s. Economic and Social Reports, 2(1), 1–21.

Reitz, J. G., & Somerville, K. (2004). Institutional change and emerging cohorts of the “new” immigrant second generation: Implications of the integration of racial minorities in Canada. Journal of International Migration and Integration, 5(4), 385–415.

Skuterud, M. (2010). The visible minority earnings gap across generations of Canadians. Canadian Journal of Economics, 43(3), 860–881.

Statistics Canada. (2015). Low income lines, 2013-2014: Update. Statistics Canada, Income Research Paper Series. Catalogue no. 75F0002M – No. 002.

Statistics Canada. (2022). Dictionary, Census of Population, 2021.

Statistics Canada. (2023). A portrait of educational attainment and occupational outcomes among racialized populations in 2021. Census in Brief. Statistics Canada, Catalogue no. 98-200-X, issue 2021011.

Zhang, X. (2010). Low income measurement in Canada: What do different lines and indexes tell us? Statistics Canada, Income Research Paper Series. Catalogue no. 75F0002M – No. 3.

Zhou, M., & Gonzales, R. G. (2019). Divergent destinies: Children of immigrants growing up in America. Annual Review of Sociology, 45, 383–399.