Introduction
While politics is often thought about in terms of representative institutions, most individuals have no direct experience with parliament or even their elected representatives. Rather, it is through everyday interactions with government agencies and concrete experiences of utilizing public services that people meet the “government in action” (Breidahl and Brodkin Reference Breidahl and Brodkin2024; Soss Reference Soss1999). These everyday interactions with government services occur in numerous ways, such as through interactions with front-line workers helping them access benefits; interactions with online platforms to fulfill duties to the state, such as filing taxes; or contacting a local office to help solve a problem. Whether using public infrastructure, obtaining official documents, or accessing social welfare and insurance programs, encountering different parts of government becomes an integral part of daily routines. In Canada, many—if not most—of these services are available in some form to non-citizen residents in a country.
A broad range of studies have revealed and documented that we should care about these everyday encounters because they can be consequential (Bell, Wright, and Oh Reference Bell, Wright and Oh2024; Bruch et al., Reference Bruch, Marx Ferree and Soss2010; Bruch and Soss Reference Bruch and Soss2018). These studies have stressed the important implications of encountering so-called welfare programs and institutions (social benefits and services) as the central sites where social rights are exercised and substantiated. As famously stated by Joe Soss (Reference Soss2002, 167), personal encounters with welfare programs can also involve broader feedback effects (e.g., on trust, political engagement, and civic virtues more broadly), which basically means that “clients draw political lessons from their program experiences because welfare agencies are usually the most accessible and consequential government institution in their life.” Therefore, to bring in one more well-known quote, it is important to recall that “[e]xperiences in these settings bring practical meaning to abstract political concepts such as rights and obligations, power and authority, voice and civic standing” (Soss et al., Reference Soss, Fording, Schram and Schram2011, 284).
In line with these arguments, the point of departure of this paper is the recognition that it is crucial to understand how individuals encounter the state and in particular the broad range of welfare programs that many people will encounter at some point in their lives—from student aid to social assistance or parental leave benefits—to understand attitudes towards government. However, this paper departs from previous policy feedback research by going one step back to consider heterogeneity in experiences: we unpack who has contact with the welfare state and the quality of that contact, including how it varies depending on someone’s citizenship background and racial identity.
Focusing on one’s citizenship background means, we approach the study of government interactions from a novel perspective that highlights the intersection of immigration history and racial background. We often think of interactions with the government as being relevant only to those who hold citizenship, yet in reality, both citizens and non-citizens interact with the state in myriad ways, and citizens themselves vary in their immigrant backgrounds. How migrants and first-generation citizens interact with the welfare state is specifically important for understanding social and political integration (Breidahl et al. Reference Breidahl, Fage Hedegaard, Kongshøj and Albrekt Larsen2021). Therefore, we consider multiple categories with varying levels of immigration background, specifically using both citizenship status (citizen versus non-citizen permanent residents) and immigrant generation (first-generation citizens who naturalized, second-generation who have at least one immigrant parent, and third-generation plus for other citizens)Footnote 1 .
As we will elaborate below, we have a particular interest in examining the gap between the “status” of the racial majority (Whites) and racial minorities. Whiteness is a socially constructed racial category that highlights significant privilege granted to individuals identified as White (Meister Reference Meister2024). While we acknowledge the need to move beyond the traditional black-and-white racial discourse and strive for a more nuanced understanding of race, this binary coding enables us to take an initial look at the potential government interaction advantages associated with racial majority status.
To capture analytically who has contact with government and the quality thereof, and how it varies across racial privilege and citizenship status, we bring forward an analytical framework that distinguishes between “need,” “capability” and “perceived quality of contact.” Citizens with various immigrant backgrounds and non-citizens alike access the state for various reasons based on their individual needs, but their likelihood of doing so is also influenced by their capacity to navigate the state and their beliefs about whether such contact will be positive. Our analysis draws on a unique series of large, cross-sectional surveys collected in Canada from 2020–2023 that ask about the type and quality of interaction with government offices. The studies include both citizens and non-citizen permanent residents, providing an excellent opportunity to investigate how one’s citizenship category is related to government interactions. Note that the non-citizens and citizens with immigrant backgrounds may have come to Canada via multiple avenues in our study (humanitarian, economic, and family class migrants), but all have acquired either permanent residency or citizenship. In addition to the appropriateness of the geographic and temporal scope of the data, Canada is an excellent case for this research particularly in recent years, due to its diverse patterns of immigration and racial backgrounds. We focus specifically on contact with different welfare programs and thereby focus on government contact related to social benefits. Our findings document how actual contact as well as experiences differ across citizenship status, immigrant generation, and racial background. Our results are striking. Among non-citizens, compared to our baseline of White 3rd or higher generation citizens, White respondents reported a similar experience, and Non-White respondents evaluated the experience to be more helpful and faster. However, among Canada-born citizens, Non-White individuals rated their experiences more unfavorably than their White counterparts, specifically noting a lower level of perceived respect. These findings point to a promising new avenue of research that can complement investigations of immigrant social integration as well as document how racial identity intersects with immigration background to structure how government is experienced.
Contact with the Welfare State
There is a long line of literature that addresses how different welfare state regimes and programs determine who can access welfare benefits and services. Notably, classic welfare state research has developed various typologies over the past decades, emphasizing how the nature of welfare state institutions can influence the inhabitants of a given society (Breidahl and Fersch Reference Breidahl and Fersch2018). Esping-Andersen’s (Reference Esping-Andersen1990) The Three Worlds of Welfare Capitalism classifies Western European welfare states into three ideal-typical welfare state regimes: liberal, conservative–corporatist, and social democratic. The regimes are defined by the types of programs that are most prevalent: liberal regimes have need-based programs, conservative-corporatist models have contribution-based programs, and social democratic models focus on universal programs. Therefore, there are cross-country variations in the design of welfare programs that affect political debates, citizens’ interpretations and preferences, notions about solidarity, and norms and beliefs in general (Kumlin Reference Kumlin2004; Larsen Reference Larsen2006; Rothstein Reference Rothstein2005; Sjöberg Reference Sjöberg2004).
However, one needs to bear in mind that even if a country is categorized as one type of regime, in reality the label may mask a combination of different program types. For example, while the geographic focus of this paper, Canada, is often categorized as a liberal welfare regime, it is a combination of means-tested programs like social assistance or income-based support to seniors and students, contribution-based programs like employment insurance and parental leave, and universal programs such as healthcare. This complexity would be overlooked if one only considers regime categorization; ignoring heterogeneity in the consequences that arise from different types of government contact experiences. For example, the paternalistic authority structures of means-tested programs—which score high on conditionality—tend to have a stigma attached to them, and importantly, anticipation of poor treatment, as well as prior poor treatment, may reduce the likelihood of accessing programs in the future (Stuber and Schlesinger Reference Stuber and Schlesinger2006). Moreover, means-tested programs also tend to have a repressive effect on different dimensions of political engagement, while universal programs are shown to have more positive effects (Mettler and Soss Reference Mettler and Soss2004; Bruch et al., Reference Bruch, Marx Ferree and Soss2010; Soss Reference Soss2002). Further, even within a single type of policy, there can be variations in experience within the same welfare regime across programs and policy designs (Bruch, Ferree, and Soss Reference Bruch, Marx Ferree and Soss2010). Yet, little research looks at differential experiences with government programs by racial identity, citizenship status, and immigration background.
Differences in Need, Capability, and Perceived Quality of Contact
While the existing research provides important insights into the link between policy designs and a number of civic outcomes, in this paper we take a step back to think about who is having government interaction experiences and what these experiences are like. We propose a framework for assessing contact with the government that emphasizes needs, capacity, and perceptions of the nature or quality of the contact. This synthesizes and brings together previous studies, especially Yang and Hwang (Reference Yang and Hwa Hwang2016), who studied immigrant health service utilization with four sets of factors—contextual, predisposing, resources, and need.
First, we assume people access benefits when they have a need, and we know that policy designs can make that need more or less onerous to meet. Needs can arise from a variety of situations. For instance, seniors may require more services due to health and retirement issues; the newly unemployed may seek unemployment insurance, but when those benefits run out or for those who are unable to seek employment, they may require social assistance, and parents may pursue childcare benefits. In some cases, such as a loss of employment, the need can be acute and sudden. In others, such as renewing a healthcare card, the process can be more procedural.
Second, some people may be more aware of what is offered by the state and what they are eligible to receive than others. In this sense, an individual’s capability may also influence how much they interact with the state. People need to be aware of service availability and have the knowledge, skills, and resources required to utilize it, such as language proficiency and digital literacy. For example, Gidengil (Reference Gidengil2020) has shown how there are information costs involved in making a claim, such as the time and energy needed to obtain information about the claiming process. Moreover, research on administrative burdens has revealed how it is a more challenging and costly process for vulnerable and disadvantaged groups to access social services compared to other groups (Herd and Moynihan Reference Herd and Moynihan2019). Social networks may also play a crucial role in providing relevant information and practical assistance in navigating complex bureaucratic processes (Ratzmann Reference Ratzmann2023). In liberal welfare states, in particular, social welfare policies may be designed in ways (due to their means-tested nature) that are particularly onerous and require more capacity among individuals to navigate, which in turn puts burdens on those who are less well-resourced (Herd and Moynihan Reference Herd and Moynihan2019).
Finally, quality of contact can also contribute to an individual’s accessibility. Having previous experience may make a second interaction with the state easier if it leads to familiarity with the process, such as where they need to apply, the types of information they may need, and how the process unfolds. Yet, encounters with the welfare state can differ, and how they are perceived by the individual matters for the effect of previous contact. For that reason, perceived quality of contact also matters. Negative experiences with the state may make people less likely to seek help or feel less efficacious during the process. If one tried to get help and was treated poorly or failed to have their needs met, this may discourage future contact. This may be particularly the case if those negative experiences were interpreted in a way that leaves someone with the impression that they were denied help for arbitrary reasons, or especially because the processes themselves are viewed as discriminatory toward people like them. Poor experiences with front-line workers, for example, can discourage people from seeking services, and minority clients feel additional stigma when seeking out services (Stuber and Schlesinger Reference Stuber and Schlesinger2006; Moynihan and Soss Reference Moynihan and Soss2014). In Soss’s (Reference Soss1999, 366) study, clients of the Aid to Families with Dependent Children program in the USA experienced interactions with caseworkers as “a pervasive threat in their life,” defined by the agency having power over them and individual caseworkers having personal discretion.
In this paper, we focus on better understanding how the amount and quality of contact differs across citizenship categories, taking into account how needs and capabilities will structure overall patterns of contact. We empirically test our expectation that individuals with greater needs will have more contact with welfare state programs. For example, those with lower incomes should have more pressing needs and be more likely to initiate contact to receive contribution-based and means-tested programs. This is simply a function of what those policies are designed to do. We also expect that individuals with greater capabilities will be likely to initiate more contact because they may be more likely to know about programs and be able to seek out the proper channels to access them. For example, people with higher levels of education may be more capable of finding and utilizing information about programs and, therefore, be more inclined to contact the government. Given our framework of government contact, we expect that any remaining variation in contact levels across citizenship categories will be an indication of how the quality of previous contact may matter. Thus, we also examine reports of the quality of contact experiences, to shed light on this intermediate point in the framework that can be difficult to measure and yet potentially impacted by one’s citizenship category due to prejudice or unfair treatment. To do so, we are particularly interested in the intersection of immigrant generation and racial identity.
The Intersection of Racial Identity and Immigration Background
While observing individual differences in need, capability, and perceived quality of contact offers an initial glimpse inside the “black box,” it provides only a partial understanding of government interaction experiences. We argue that fully grasping the nature of contact with the welfare state requires considering how experiences with the state may be racialized, and importantly how majority or minority racial background may intersect with migration background and citizenship status.
There is a longstanding literature linking racial inequality to social welfare in the U.S. context (e.g., Lieberman Reference Lieberman2001; Brown Reference Brown2013; for those outside the U.S. context, see De Micheli Reference De Micheli2018; Lewicki Reference Lewicki2022; Eseonu Reference Eseonu, Gregory and Iafrati2024). Poverty in the USA—and the programs that aim to alleviate it—has long been racialized (Mendelberg Reference Mendelberg2001; Schram et al., Reference Schram, Brian Soss and Carl Fording2003). The public regularly overestimates the number of racial minorities on social welfare programs and is less willing to support programs when cued about recipients being Black (Gilens Reference Gilens1999). In Michener’s (Reference Michener2019) important work on Medicaid recipients, for example, she shows that race is an important component of how Black Americans experienced the program. Clients of the program were well aware of racial stereotypes and believed it influenced how they were treated while also influencing when and how they protested poor treatment (Michener Reference Michener2019, 99–101). Michener develops a framework through which to understand policy feedback as a racialized process. She notes:
Policies often channel resources unevenly and inequitably across racial groups; racial stratification is a key determinant of the advent, alignment, and power of interest groups; and race is a fundamental prism through which experiences of policy are understood and interpreted (Michener Reference Michener2019, 425).
Racialized structural inequalities and long-term systemic racial discrimination in the USA have contributed to a persistent racial wealth gap (Derenoncourt et al. Reference Derenoncourt, Hyun Kim, Kuhn and Schularick2022). As a result, communities of color are likely to experience higher levels of need for programs like income assistance. However, while prior experiences with the state may make the program more familiar, if that experience is negative it may also lead people to feel less efficacious and more discriminated against, influencing the types of demands people make on these programs.
In the Canadian context, relatively less is known about how race structures interaction and experiences with the welfare state. What research does exist has shown similar dynamics of racialization in public opinion, where less support is received for programs when minority beneficiaries are the recipients—at least or especially among those with racial prejudice (Harell, Soroka, and Ladner Reference Harell, Soroka and Ladner2014; Harell, Soroka, and Iyengar Reference Harell, Soroka and Iyengar2016; Ford Reference Ford2016). There is also evidence of wage gaps between the white majority and some minority communities. For example, Fearon and Wald (Reference Fearon and Wald2011) find a significant wage gap between white and black workers that increases when considering immigrants, partly rooted in discrimination. Building on these studies of racial prejudice, this study provides an opportunity to explore differences between racial majority and minority groups in Canada. By examining the amount and quality of government contact, we can assess whether a certain group requires more assistance, is better able to navigate the system, or perceives equal treatment.
Compared to the research studying government interaction through the lens of race, very little research in Canada or elsewhere explores how citizenship status and immigrant background shape contact with the welfare state. However, migration holds both empirical and theoretical significance for the nature of government contact. Empirically, Canada has a long history of immigration and currently has one of the highest immigration rates in the world, with 23% of its population composed of immigrants (Statistics Canada 2021). Theoretically, citizenship and immigration background add a layer of needs, capabilities, and perceived quality of the contact. In this study, we focus on non-citizens holding permanent residency, who, in the Canadian context, have access to the same benefits and programs as citizens. Immigration status may structure the distribution of resources in Canada (need) as well as access to resources (such as higher education). Compared to the USA, foreign-born individuals in Canada tend to have relatively high levels of resources (such as higher education levels than the native-born populationFootnote 2 ) due in part to the points-based immigration system. However, these skills do not always translate into knowledge about the Canadian system or the language capabilities to navigate it.
In particular, we expect that citizenship status and immigration history may shape perceived quality of contact. A recent study based on survey data collected in Denmark among newly arrived refugees from Syria and Ukraine accentuates how evaluations of government interactions vary depending on which institutions/programs/policies they encounter. This research reveals that people with refugee status do not necessarily have more negative experiences with welfare state programs than native-born citizens and that experiences with host country institutions may be shaped by experiences with institutions in the country of origin (Breidahl et al. Reference Breidahl, Djuve, Kavli, Larsen and Sandy Harboesgaard Nielsen2025). There are also studies documenting (over)-confidence among migrants in destination country institutions. A central finding is that migrants display higher trust in destination state institutions than native-born citizens. Mexicans in the USA (Weaver Reference Weaver2003; Wenzel Reference Wenzel2006), migrants from non-democratic countries in Canada (Bilodeau and Nevitte Reference Bilodeau and Nevitte2003), and Caribbeans and South Asians in the UK (Maxwell Reference Maxwell2009) are found to have higher trust in destination country institutions than natives. The same has been found in the European context (Maxwell Reference Maxwell2010; Röder and Mühlau Reference Röder and Mühlau2012a; Reference Röder and Mühlau2012b). This interpretation is supported by the empirical finding that the difference between native-born citizens and migrants is largest upon arrival and diminishes with the length of stay in the destination country (Maxwell Reference Maxwell2010; Röder and Mühlau Reference Röder and Mühlau2012a; Weaver Reference Weaver2003; Wenzel Reference Wenzel2006).
What is missing from the literature on immigrant-government interaction is a more nuanced understanding of how racialized experiences with the state may be experienced by migrants versus native-born citizens. At the same time, we may miss important nuances in racialized experiences with the welfare state if we do not take into consideration that the experiences of more recent immigrants may differ from long-term citizens. The clear distinction between the US-based race literature and the immigrant-focused European literature suggests that a comprehensive understanding of citizen-government interactions among all clients may require combining the two. There may be important intersections between race, citizenship status, and immigrant background that have yet to be identified. Previous research suggests this is likely. For example, among Muslim immigrants, there are generational differences in the ability to recognize a problem within the country, with second-generation citizens more likely to perceive contexts as racialized compared to the first-generation (Yazdiha 2019; Reference Yazdiha2021). This echoes findings about the complex ways in which second-generation citizens of color navigate their identities (Waters Reference Waters1990; Reference Waters2019).
To address these deficiencies in the existing literature, in this paper we investigate how racial identity, citizenship status, and immigration backgrounds intersect to affect who is having government contact and what their experience is like. To the best of our knowledge, this is a new area of study that has received little study. Further, it diverges significantly from research on migrant categories (also entry categories), where legal rights vary considerably across these categories (Söhn Reference Söhn2013; Sainsbury Reference Sainsbury2006; Breidahl et al., Reference Breidahl, Fage Hedegaard and Seibel2022).
Beyond individual-level measures of need and capability, group-level differences related to racial, citizenship, and immigration backgrounds capture the levels of need and capability linked to one’s stage in the immigrant integration process itself. Independent of factors such as age, gender, or employment status, non-citizens may have greater needs for government services, while those with a longer family history may be more integrated into society and better equipped to navigate government programs.
We also expect that when contact occurs, the experience will vary by both immigration and racial backgrounds. Non-citizens may evaluate their experiences more positively, driven by their immediate needs and circumstances, and perhaps their (over-)confidence in their new country of residence. In contrast, individuals with longer immigration histories, especially those born in the country, may have higher expectations and be more critical in their evaluations of the process of government contact. Particularly, among racialized populations, newcomers may experience difficulties but not necessarily recognize more subtle forms of racism they encounter. However, racialized native-born citizens may expect better treatment from the state as well as be more attuned to “seeing” unfair or prejudicial practices as they occur. Given the limited data available on individuals’ past experiences, perceptions, or expectations, this paper explores their evaluation of recent contact to contextualize the experiences of each group.
Data and Measurement
To explore the questions raised in the paper, we analyze data from four Democracy Checkup (DC) surveys, which gather information about Canadians’ opinions on democratic values, public policies, and current issues. The surveys sample the general population of Canada, aged 18 or older, who were either Canadian citizens or permanent residents. The surveys were conducted in 2020, 2021, 2022, and 2023 (Harell et al. Reference Harell, Stephenson, Rubenson and John Loewen2022a; Reference Harell, Stephenson, Rubenson and John Loewen2022b; Reference Harell, Stephenson, Rubenson and John Loewen2023; Reference Harell, Stephenson, Rubenson and John Loewen2024) and include questions about individuals’ contact with the government and their evaluation of that interaction—our two key dependent variables. These specific questions were presented to half of the sample in 2020 and 2021 and to the entire sample in 2022 and 2023. As a result, the total sample size is 23,596, providing a unique opportunity to examine subsamples of white and racial minority respondents across citizenship status and immigration generation.
Need and Capability
We use five variables as indicators of need—age, gender, employment, income, and parental status—and two variables as indicators of capability—education and native speaker status. We code these variables into either binary or categorical formats, acknowledging the potential nonlinearity in their relationship with contact. Forcing linearity could lead to an inaccurate representation of this relationship. Additionally, this approach enhances our ability to grasp individuals’ needs and capabilities more directly and intuitively.
Age was coded into three categories: 18–30, 30–65, and 65 and above, reflecting the life cycle that involves changes in occupation and wealth. In terms of contact, we may expect more need in the 18–30 and 65+ age groups compared to the 30-65 group. Gender was initially measured in multiple categories (Q: How do you describe yourself? A: A man; A woman, Other (e.g., Trans, non-binary, two-spirit, genderqueer, gender-fluid)). In 2022 and 2023, the response options changed slightly to A man; A woman; Non-binary; Another gender, please specify. For analysis, we transformed both types of questions into a binary variable distinguishing those identifying as women from those identifying as men to reflect some of the disadvantages in terms of socioeconomic status in society. Employment (Q: What is your employment status? Are you currently… A: Working for pay full-time; Working for pay part-time; Self-employed (with or without employees); Retired; Unemployed/ looking for work; Student; Caring for a family; Disabled; Student and working for pay; Caring for family and working for pay; Retired and working for pay; Other (please specify)) was coded into five categories: employed; unemployed; student; retired; and other.
Income level was originally measured on an eight-point scale (Q: What was your total household income, before taxes, for the year 2021? Be sure to include income from all sources. A: No income; $1 to $30,000; $30,001 to $60,000; $60,001 to $90,000; $90,001 to $110,000; $110,001 to $150,000; $150,001 to $200,000; More than $200,000). We recoded it into a binary variable for above-median income and at or below median income (the median income was $60,001 to $90,000). Parental status (Q: How many, if any, children do you have?) was coded to distinguish those who indicated having a child/ren and those who did not.
For our indicators of capability, education was originally measured on an eleven-point scale (Q: What is the highest level of education that you have completed? A: No schooling; Some elementary school; Completed elementary school; Some secondary/ high school; Completed secondary/ high school; Some technical, community college, CEGEP, College Classique; Completed technical, community college, CEGEP, College Classique; Some university; Bachelor’s degree; Master’s degree; Professional degree or doctorate). We recoded it into a binary variable for those who have Bachelor’s degree and above and those who do not.
For language, we measure whether the respondent is a native speaker of the majority language using two questions (“What is the very first language you learned and still understand?” and “Which province or territory are you currently living in?”). The respondent is coded as a native speaker of the majority language if either the very first language they learned and still understand is English and they live outside Quebec or the very first language they learned and still understand is French and they live in Quebec.
Racial Background
Race is a social categorization of biologically based human characteristics. Defining and measuring this social construct poses significant challenges due to its constantly shifting and evolving nature. The boundaries of race are socially and historically fluid (Omi and Winant Reference Omi and Winant1994); individuals once categorized as part of a specific racial group may no longer be so, and racial categorizations often vary between countries. The boundaries are ambiguous and blurred as some individuals transgress racial categories (e.g., Latinos and Arabs), and some identify with multiple categories. Furthermore, subjectivity adds complexity, as there may be a gap between how individuals define their racial category (self-identified) and how others define them (perceived).
Our coding of race reflects these complexities. We primarily rely on a self-report measure in the survey. Respondents self-identified their racial background (“Do you identify as… (Please select all that apply)”) by selecting from the following options: White; Indigenous (e.g., First Nations, Métis, Inuit, etc.); South Asian (e.g., East Indian, Pakistani, Sri Lankan, etc.); Chinese; Black; Filipino; Latin American; Arab; Southeast Asian (e.g., Vietnamese, Cambodian, Laotian, Thai, etc.); West Asian (e.g., Iranian, Afghan, etc.); Korean; Japanese; and Other (please specify). While this accurately reflects the understanding of ethnoracial categories in the Canadian context, we are cognizant of the limitation that it does not entirely capture how respondents are perceived by others.
We created a binary variable categorizing individuals as either “White” or “Non-White.” Our focus in this study is to explore how the “status” of racial majority and minority reflects levels of socialization, privileges, and discrimination. This inquiry holds particular significance in the present context, given the rapid growth of the racial minority population in Canada, historically defined as “visible minorities” by Statistics Canada. While our primary analysis utilizes this binary variable, we also offer additional breakdowns in the appendix when possible.
To address the issue of blurred boundaries between the two categories, we devised three alternative codings (see Table 1A in the Appendix). Each version aims to illuminate differences in the levels of access, treatment, and expectations discussed earlier. The first version adopts a broad definition of White, coding anyone who identifies that way as White. This includes multiracial individuals who identify as White alongside another ethnoracial category. This version addresses the higher level of capability and privilege of the racial majority compared to the racial minority, on average.
The second version centers more on phenotypical perceptions of race. Skin color is one of the prominent and important facets of the construction of race, and has not only contributed to the stratification of experiences between racial groups but also shaped the experiences within these categories (Yadon and Ostfeld Reference Yadon and Ostfeld2020; Ostfeld and Yadon Reference Ostfeld and Yadon2022). Our second alternative coding reflects a focus on skin tones, including those who identify as both White and also either Latin American, Arab, or West Asian, as “White.” This version may shed light on the differential treatment that individuals receive based on their skin color during their contact with the government. The focus here is—to the extent our rough measures can capture it—likely perceptions of racial minority status.
The third version offers a narrow definition of White, coding only individuals who exclusively identify as White in the category. This version recognizes that multiracial White individuals may perceive themselves differently from those who exclusively identify as White. This coding may highlight distinctions in expectations.
Immigration status and history
We explore immigration based on their current citizenship status as well as their birth country of and that of their parents. Respondents indicated if they were Canadian citizens (Q: Are you a… A: “Canadian citizen” or “Permanent resident”) and if they were born in Canada (Q: Were you born in Canada? A: “Yes” or “No”). In addition, respondents were asked whether their parents were born in Canada (Q: Were your parents born in Canada? A: “Both parents born in Canada,” “One parent born in Canada,” or “Neither parent born in Canada”)Footnote 3 . Combining these three questions, we can create four categories: Canada-Born Citizens (3+ Gen); Canada-Born Citizen (2nd Gen); Foreign-Born Citizen (1st Gen); and Foreign-Born Non-Citizen.
Citizenship Categories: Race x Immigration Status
By intersecting race and immigration status, we create an eight-category variable called “Citizenship Categories.” This variable consists of (1) White Canada-Born Citizen (3+ Gen), (2) White Canada-Born Citizen (2nd Gen), (3) White Foreign-Born Citizen, (4) White Foreign-Born Non-Citizen, (5) Non-White Canada-Born Citizen (3+ Gen), (6) Non-White Canada-Born Citizen (2nd Gen), (7) Non-White Foreign-Born Citizen, and (8) Non-White Foreign-Born Non-Citizen.
Across these eight categories, we observe some notable differences in demographic and socioeconomic characteristics that represent need and capability (see Table 2-1A, 2-2A, and 2-3A in the Appendix). In general, White respondents, on average, were younger than Non-White respondents, and the proportion of respondents aged 65 and above was higher among Non-White 2nd generation individuals and non-citizens. The proportion of women was higher among White participants than Non-White. Regarding employment, unemployment rates are higher among Non-White groups than among White people. The proportion of retired individuals was higher among White groups, while the proportion of students was higher among Non-White groups. Some patterns in income level were also observed, where compared to those in the 3+ generation, 2nd- and 1st-generation individuals had higher income, and non-citizens had lower income. Regarding parental status, on average, the proportion of parents was higher among White than Non-White individuals.
The proportion of individuals with a university degree was lower for White than Non-White individuals, lower for those in the 3+ generation than other immigration statuses, and lowest for the White 3+ generation category compared to the other seven categories. Regarding native speaker status, no difference was observed between White and Non-White 3+ generation individuals, but all other categories had a lower proportion of native speakers compared to 3+ generations.
Government Contact
In 2020 and 2021, respondents were asked “Have you contacted a government office in the past 12 months?” and given response options of “Yes” and “No.”Footnote 4 In 2022 and 2023, the question was updated to “Have you contacted a government office about any of the following programs in the past 12 months? Please select all that apply.” Response options included “Employment insurance,” “Maternity or parental benefits,” “Student aid,” “Income assistance,” “Other (please specify),” and “I have not contacted a government office.” We created a binary variable using the whole sample to indicate whether contact occurred. For the 2022 and 2023 data, we have a multinomial variable indicating the type of programs and an ordinal variable representing the number of programs, ranging from 0 to 5.
Perceived Quality of Contact
For those respondents who said they had contacted a government office, they had a follow-up question asking, “Thinking about your most recent contact, how would you rate the treatment that you received on the following scales?” The four seven-point scales were “Unhelpful–Helpful,” “Disrespectful–Respectful,” “Slow–Fast,” and “Made me feel badly about myself—Made me feel good about myself.”
Results
Government Contact
We begin with a description of the amount and type of contact people reported with the government during the previous 12 months. In the combined dataset, about one-quarter of the respondents indicated having some interaction with the government in the past 12 months. Yet, as is clear in Table 1, the proportion of contact varies by background: there is more significant variation among Non-White individuals than White individualsFootnote 5 across immigration categories. Among White respondents, there is minimal variation in government contact proportions: compared to White Canada-born citizens (3+ Gen), both White Canada-born citizens (2nd Gen) and White foreign-born non-citizens had similar proportions (p = 0.7 and 0.6, respectively), and there was slightly lower proportion for White 1st generation citizens (p = 0.01). Conversely, among Non-White respondents, there are statistically significant differences in contact proportions: compared to Non-White Canada-born citizens (3+ Gen), both Non-White Canada-born citizens (2nd Gen) and Non-White foreign-born citizens (1st Gen) (both p < 0.001) had lower proportions, but there was little difference with Non-White foreign-born non-citizens (p = 0.2), though that category has substantially higher contact compared to White foreign-born non-citizens (p < 0.001).
Proportion of contacted across citizenship categories
Table 1 Long description
The table presents data on the proportion of contacted across different citizenship categories and racial groups. It has four rows and four columns. The row labels are White and Non-White. The column headers are 3+ Gen, 2nd Gen, 1st Gen, and Non-Citizen. Row 1: White, 3+ Gen, 24.92; 2nd Gen, 24.62; 1st Gen, 22.11; Non-Citizen, 26.39. Row 2: Non-White, 3+ Gen, 36.65; 2nd Gen, 25.02; 1st Gen, 22.87; Non-Citizen, 40.38. The table shows that Non-White individuals have more variation in government contact proportions across immigration categories compared to White individuals. Among White respondents, the proportions are relatively similar across different citizenship categories. In contrast, among Non-White respondents, there are significant differences in contact proportions, with Non-White foreign-born non-citizens having the highest contact proportion.
The 2022 and 2023 datasets provide additional information about the amount and types of contact. Figure 1 presents the type of programs contacted. Recall that in these surveys, respondents were asked what government programs they had accessed and could select as many as applied from a list that included employment insurance, maternity or paternity benefits, student aid, income assistance, and an open-ended “other” option. The most common program accessed was employment insurance, followed by the “other” category, and then income assistance. When we look at the use of these programs by citizenship categories, we see some differences in the patterns that emerge. Employment insurance, the most common category, has the greatest proportion of use among White 3rd generation citizens. Student aid appears to be particularly important among racial minority respondents, especially those who are non-citizens or 1st or 2nd generation.
Type of Programs Contacted. Source: DC 2022, 2023.
For a more nuanced perspective, we delve into the number of different program contacts reported, utilizing the updated question version from 2022 and 2023 (also see Figure 2A in Appendix). It remains evident that, on average, Non-White individuals have a higher frequency of government contact compared to White individuals and, importantly, that this varies across citizenship categories. Figure 2 is partly in line with our expectations: among citizens, those in older immigration generations tend to have more government interactions, reflecting a higher level of capability. On the other hand, non-citizens are likely to have more need for engagement with the government due to their status, which is reflected in the higher number of average contacts. Yet what is most striking here is the difference across citizenship categories across racial backgrounds. In every instance, racial minority respondents have higher contact than their Non-White citizen group counterparts, and there appears to be greater variance in contact among racial minorities compared to white respondents.
Average Number of Programs Contacted. Source: DC 2022, 2023.
Why might this be? Recall that our expectations about who interacts with the government are based in part on differential needs and capabilities, some of which we can be captured with standard demographic variables. Do these measures explain the observed differences between groups? Table 2 presents OLS estimates of five need variables and two capability variables on contact, both separately and in a multivariate model with our citizenship category variables. In general, and as expected, individuals with higher levels of need are more likely to have contact with the government: they are, on average, older, a student, unemployed, with below-median income, and a parent. This pattern aligns with the nature of many government interactions, often involving financial assistance targeted at these groups. Interestingly, being a woman is insignificant. Regarding the two capability variables, native-language speakers had a higher proportion of contact compared to non-native speakers. In contrast, there was no significant difference in the proportion of contact between those with a university degree and those without.
OLS estimates using contact (1 or more)
Table 2 Long description
The table presents OLS estimates of contact levels across various demographic variables. It includes columns for different contact groups (N1, N2, N3, N4, N5, C1, C2, N&C, Categories, All) and rows for different demographic categories such as Age, Woman, Employment, Income, Parent, University Degree, Native Speaker, and Categories. Each cell contains a value with a standard error in parentheses. Notable trends include higher contact levels for individuals aged 65 and above, students, unemployed individuals, those with below-median income, parents, and native speakers. The table also shows that being a woman is insignificant in terms of contact levels. The R-squared and adjusted R-squared values are provided at the bottom, along with the number of observations and RMSE for each contact group.
***p < 0.001; **p < 0.01; *p < 0.05, Robust standard errors (HC2) in parentheses.
If race and citizenship status are merely indicative of disparities in need and capability, the observed gap in contact across citizenship categories should disappear. However, even in the model incorporating citizenship categories alongside indicators of need and capability, differences in contact persist. The final column of Table 2 shows that while the small differences among White respondents of different citizenship categories disappear, variation among Non-White respondents continues. Compared to White Canada-born citizens (3+ Gen), Non-White Canada-born citizens (3+ Gen) and Non-White foreign-born non-citizens exhibit a higher proportion of contactFootnote 6 , and Non-White Canada-born citizens (2nd Gen) have a lower proportion of contact. Nevertheless, the significance for Non-White Canada-born citizens (2nd Gen) disappears in two other versions of coding. The size of the effect for Non-White foreign-born non-citizens does decrease somewhat when controls are introduced, but the significant effect for three+ generation racial minorities is persistent.
The persistence of this gap across various citizenship categories raises questions. While our controls for need and capability certainly do not capture all forms of need or capability, they do capture a plausible set of controls that are directly related to the concepts of interest and, at a minimum, are likely correlated with other possible forms of need and capability.
One critical implication of our findings is that the socialization of these groups, which would vary based on the combination of their citizenship, immigration history, and racial background, matters. Existing literature suggests that individuals’ interactions with the government shape their expectations. While data on individuals’ past experiences with the government are lacking in our dataset, we can capture patterns that may be emerging based on how they evaluate their most recent contact.
Perceived Quality of Contact
Turning now to the nature of experiences with the government, we again consider variation by race and citizenship status. On average, the respondents who interacted with the government rated the treatment they received as slightly positive (4.5) (on a 1–7 scale). This average does not mask any dramatic heterogeneity across the dimension of the experience evaluated: the highest is “Respectful” (5), followed by “Made me feel good about myself” (4.5), “Helpful” (4.4), and “Fast (3.8)” (see Tables 5-1A, 5-2A, and 5-3A in Appendix). We also checked whether the type of contact mattered for the evaluation, but found no significant difference across types of programs contacted (see Figure 5A in Appendix).
These averages hide notable patterns across different citizenship categories, though. Figure 3 reports the four dimensions of experience evaluations across the eight citizenship categories (also see Figures 6-1A and 6-2A in Appendix). On helpfulness, Non-White Canada-born 2nd and 3+ generation individuals rated their experience more negatively than the baseline (p = 0.03 and p < 0.001, respectively). On speed, only the Non-White Foreign-born Non-citizen group had a more positive evaluation than the baseline (p = 0.02). On feeling good about oneself, Non-White Canada-born 2nd and 3+ generation individuals rated their experience more negatively than the baseline (p = 0.004 and p < 0.001, respectively).
Differences in the Experience across Citizenship Categories.
Special attention should be paid to respectfulness, given that such evaluations are likely to be shaped by expectations influenced by racial background and immigration experience. Unlike the other three dimensions, we see a noticeable difference between White and Non-White individuals in terms of reporting “Respectful” interactions. Compared to the White 3+ generation, all Non-White groups reported lower levels of respectfulness, although the differences with Non-White non-citizens were not statistically significantFootnote 7 .
These findings point to the importance of understanding government interactions through the intersection of citizenship, immigrant generation, and racial background. The pattern we observe in Figure 3 could capture, on the one hand, what we suspect is real differential treatment between White and Non-White groups during government interactions. Yet the fact that this gap grows across generations suggests that in addition to differential treatment, how this treatment is interpreted across the intergenerational integration of migrants is important. Native-born racial minorities report more unequal treatment than non-citizen racial minorities, especially among those from families where the parents are also native-born (3+ generation).
Such a finding, we suspect, may be due to two interrelated processes. On the one hand, as some of the literature on immigrant interactions suggests, newer immigrant communities may be (over-)confident in the institutions of their new country. This could be because their experiences in their countries of origin were relatively worse, or the program was simply not available. The most recent immigrant waves to Canada largely come from countries with less developed welfare states (India and China) (Sivakumar Reference Sivakumar2023). If one comes from a country with fewer government programs, any program contact may be seen in a positive light. Immigrants may also feel more pressure to report the treatment was positive, as an indication of their own integration within the new country. We see some evidence of more positive evaluations among more recent immigrants (especially non-citizen racial minorities). The fact that these appear primarily on the “fast” and “helpful” dimensions may reflect the first explanation: the process was viewed as efficient, while there was less evidence that the contact left the person feeling particularly good or respected compared to others in the sample.
At the same time, the more negative evaluations among 2nd and especially 3rd generation+ raise questions. Is it the case that native-born racial minorities are treated worse than foreign-born non-citizen racial minorities? This seems improbable. What we suspect, rather, is that expectations of fair and equitable treatment by racial minorities may be greater for those born and raised in Canada, where an ideal of equality and fairness plays an important role in the ways in which the country is portrayed in both policies and in the minds of its citizens. As Yazdiha (Reference Yazdiha2019) recently found in her study of Muslims in Europe, more inclusive contexts actually led to greater experiences of discrimination, and this was particularly true of native-born Muslims. As people come to expect better treatment, we suspect that they are better able to identify and evaluate discriminatory treatment when it emerges.
Conclusion
Programs designed to assist citizens in certain situations (unemployment, poverty, schooling, etc.) should, in theory, be reaching those who need the program. Yet, as we have seen in this paper, there are differences that emerge in the amount of contact not only just by program requirements (needs) or by the specific resources available to the person (capability), but also by racial minorities across immigrant categories. In this paper, we document higher contact among some racial minority groups (non-citizens and 3rd generation+ citizens) that require further study. The result leaves us with new questions, such as whether these groups are well-resourced in other ways that help them to navigate the programs available to them. One possibility is that there are resources embedded within communities themselves where knowledge is shared.
What is more striking in our empirical findings is that the evaluations of these contacts vary in systematic ways that do not lead to a simple racial or immigration explanation. Rather, while White individuals’ evaluations of their contact with the state vary little by citizenship status or immigrant generation, we see starker patterns among racial minorities. While both non-citizen racial minorities and 3+ generation racial minorities tend to have more contact with the state, even controlling for needs and capabilities, they have starkly different evaluations of this experience. Non-citizens tend to be most likely to evaluate their contact as fast and helpful. Yet, native-born citizens, especially those with a longer family history in Canada, report the most negative experiences. This insight would have been missed if we had only examined the data through the lense of either native-born versus foreign-born or white versus non-white distinctions, since these negative experiences would have been masked when grouped with White Canada-born citizens or Non-White newcomers who reported more positive experiences. These findings highlight the need to address potential systemic prejudice within the state to ensure racial minorities feel that they are treated equally and fairly.
Our study is not without limitations, but it is the first study we know of in the Canadian context that seeks to understand how immigration history and racial background affect how people interact with the welfare state. Clearly, more needs to be done to understand the nuances of government contact, including more details about the nature of the contact (for example, was it in-person or via an electronic platform? Was it successful or not? And how did this experience relate to prior experience?). While our emphasis here has been on contact associated with social welfare programs, it is essential to acknowledge other forms of contact. Exploring contact experiences with immigration services, the criminal justice system, or universal programs like healthcare may provide further evidence of the patterns observed here, or it may provide nuance about how these patterns play out across sectors. Such sectors vary in the intensity of interactions from mundane (e.g., renewing a driver’s license or passport) to more intense (e.g., interactions with police and courts). Especially in the context of a federal system, certain contacts involve local governments, while others involve the subnational or national levels. Therefore, a more comprehensive examination across a broader spectrum of interactions would enable a more nuanced understanding of the nature of government contact and how this might be structured by the intersection of citizenship status and racial background.
Of course, racial and immigration backgrounds are far from being monolithic categories. The experience of Black Canadians or Indigenous peoples in Canada may vary in important ways from other ethnoracial minority groups. Future research would benefit from data collections designed with oversamples, or that use in-depth work with specific communities, to tease out how one’s group identity structures these interactions beyond the majority white racial minority divide. Furthermore, while we focus only on non-citizens who have already acquired permanent residency, those who are still in the process of applying for permanent residency, hold temporary visas (e.g., work or student permits), may exhibit different patterns. The particularities of the path to migration may also matter deeply. Refugees or asylum-seekers may experience the state very differently from those who came in as economic immigrants. This also is a potential direction for future research.
Despite its limitations, this work provides an important exploration of how government interaction is structured along citizenship status, immigrant generation, and racial background. Our research points out that different levels of contact cannot be easily explained by needs or capabilities. We also document experiences of worse treatment, which becomes more prominent among native-born citizens with a racial minority background. These differences are important because there is substantial evidence in the policy feedback literature that demonstrates how experiences with the state have spill-over effects on citizens’ attitudes about the government and their own ability and capacity to influence it. This study provides a starting point for investigating these experiences through an intersectional lens that seeks to understand how such interactions may be shaped by immigration experience and persistent categorical inequalities based on racial background.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/rep.2024.31
Acknowledgment
The authors would like to thank Andrea Lawlor and three anonymous reviewers for their helpful comments on the draft.
Funding statement
This research was supported by the Consortium on Electoral Democracy (C-Dem) with funding from the SSHRC Partnership Grant (#895-2019-1022) and the SSHRC Insight Grant (#435-2018-1208). Ethics was reviewed by the University of Western Ontario (#115293) and the Université du Québec à Montréal (#2021-3239, 4254).
Open access
