Introduction
Regionalism, the presence of systematic variations in regional political cultures and the existence of regional cleavages, is a central feature of Canadian politics. At the heart of regionalism lie tensions between popular loyalties to federal and provincial governments. As Clarke and colleagues explain (1980: 35), loyalty to the regions and not to the country as a whole was a basic reason why the Fathers of Canada crafted a confederation. Since then, Canadians exhibit a “will to live together” and a “will to live apart” (LaSelva, Reference LaSelva1996) and consistently balance those distinct and potentially conflicting loyalties to their province and to the country. This is a fundamental part of the basic dynamics of Canadian regionalism and hence of Canadian politics.
Whether regional political cultures and cleavages between dual loyalties are attributable to different settlement patterns (Elkins and Simeon, Reference Elkins and Simeon1980; Schwartz, Reference Schwartz1974; Wiseman, Reference Wiseman and Dunn1996), fundamental regional economic differences (Brodie, Reference Brodie1990; Brym, Reference Brym1986; Wilson, Reference Wilson1974) or Canada's federal political institutions (Simeon and Elkins, Reference Simeon and Elkins1974) remains a focus of vigorous debate.Footnote 1 That said, most acknowledge that variations in the origins, timing and settlement patterns of large flows of immigrants to Canada have had a marked impact on the development of Canadian political culture and, more specifically, regional subcultures (Blake, Reference Blake1972; Simeon and Elkins, Reference Simeon and Elkins1974: 433; Elkins and Simeon, Reference Elkins and Simeon1980; Lipset, Reference Lipset1990; Wiseman, Reference Wiseman and Dunn1996, Reference Wiseman2007). Such immigrant waves as the early settlement of New France, the migration of loyalists from the American colonies to Ontario, New Brunswick and Nova Scotia in the 1780s, and the movement of Americans and Central and Eastern Europeans to Ontario and Western Canada in the late nineteenth and early twentieth century are all thought to have a profound impact on the dominant political outlooks of these regions.
Substantial attention has been paid to understanding the origins of cleavages between these dual loyalties, but relatively few efforts have been directed at exploring the contemporary dynamics of regionalism in Canada. This analysis directly explores the interplay between these two central features of Canada's political dynamics, namely regionalism and immigration. More specifically, the focus is on how contemporary immigrants navigate the dynamics of orientations towards the federal and provincial governments. Previous research examining whether dual loyalties induce identity conflict found that Canadians accommodate these dual loyalties reasonably well (Clarke et al., Reference Clarke, Jenson, LeDuc and Pammett1980: 68). But does the same hold for new Canadians? Do immigrants develop political loyalties that are centripetal or centrifugal? Are these newcomers more federally or provincially oriented than their native-born counterparts living in the same province? And what impact do these new Canadians have on regional dynamics of political loyalties?
There are good reasons to suppose that contemporary immigration patterns have the potential to significantly transform these regional dynamics. Both the magnitude of immigration flows and the settlement patterns of new immigrants have changed considerably in recent decades. First, immigrants made up some 19.8 per cent of the Canadian population in 2006, and as the data in Figure 1 indicate, immigration has contributed to more than half of Canada's population growth since the 1960s. The proportion of population growth due to immigration peaked at just over 85 per cent between 1996 and 2001, a level not seen since immigration waves in the early part of the twentieth century with the settlement of the Western provinces. Second, the settlement patterns of immigrants to Canada have shifted significantly in recent years. The longstanding pattern was that Ontario has been the overwhelming destination of choice for the majority of immigrants. Ontario remains the province of choice for most immigrants, but Quebec, British Columbia and Alberta have experienced recent rapid growth in their foreign-born populations. Third, these shifts have also been accompanied by significant changes in the composition of Canada's immigrant population over the last 40 years. The founding waves of immigrants came from traditional source countries, Europe and the United States. The contemporary record is different. The vast majority of new immigrants now come from non-traditional source countries in Asia, the Middle East and Africa (Chui et al., Reference Chui, Tran and Maheux2007). Given the volume, the sharp shift in country of origin and the significant changes in settlement patterns, it is reasonable to suppose that these new waves of immigrants have the potential to reshape regional political dynamics. As with previous waves of immigration, the inflow of new Canadians potentially brings an influx of distinct values and ways of relating to political institutions. The values of today's immigrants are likely to become an important part of Canada's future political culture. What impact, then, do these new patterns of immigration have on Canadian political culture and more specifically on regional dynamics in the country?
Figure 1 Population Growth Due to Immigration (% Change from Previous Census)
Sources: Statistics Canada 2009; Chui et al. Reference Chui, Tran and Maheux2007
Note: Data from 1901 to 1951 are in 10-year increments; Data after 1951 are in 5-year increments
Regionalism and immigration are two central features of Canada's political system, but systematic empirical explorations of the relationship between the two remain relatively rare. One exception is Elkins' pioneering investigation (Reference Elkins, Elkins and Simeon1980) into whether immigrants develop attitudes similar to those of the people already living in the host province. Elkins' central finding was that immigrants generally did adjust to provincial political norms, but his data also showed that provincial patterns were less pronounced among immigrants. According to Elkins, “regional and provincial variations have, if anything, been muted by the vast numbers of immigrants to Canada” (122). Elkins' insights focused on such core aspects of political culture as political efficacy and trust, and they relied entirely on data collected in the 1960s and 1970s and thus on immigrants who settled in Canada during the 1940s and 1950s. Given the recent transformations in immigration patterns it is not at all clear that these original findings provide a firm foundation for generalizing about more contemporary dynamics. The present analysis revisits the question and explores immigrants' federal and provincial loyalties from two vantage points. The first focuses on immigrants from traditional source countries, the group that most closely approximates the subjects of Elkins' study. The second focus is on immigrants from non-traditional source countries. Not only is this group increasingly prominent in contemporary Canadian immigration, it is also more culturally distinct from the country's native-born population.
Immigrants' Political Integration: Provincial versus Federal Poles of Loyalties
The process of immigrants' political integration is complex and it involves many factors that may or may not entail acculturation (Berry, Reference Berry2001). Through acculturation immigrants undergo both “culture shedding” and “culture learning” (Berry, Reference Berry1997); immigrants “unlearn” some orientations acquired in their previous cultural context, while “learning” and adopting the new attitudes and orientations that reflect their new cultural context. Empirical evidence from Canada (Black, Reference Black1987; Black et al., Reference Black, Niemi and Powell1987; White et al., Reference White, Nevitte, Blais, Gidengil and Fournier2008) and elsewhere (Finifter and Finifter, Reference Finifter and Finifter1989; McAllister and Makkai, Reference McAllister and Makkai1991, Reference McAllister and Makkai1992; Bilodeau, Reference Bilodeau2008) supports this perspective. On balance, the data indicate that although immigrants' pre-migration backgrounds tend to influence their political outlooks in the new host country, immigrants are nonetheless adept at political learning in the host society. Thus, the acquisition of political loyalties should be part of the achievable culture learning menu available to immigrants in the new host society. But which political loyalties are absorbed and which ones are not?
One possibility is that immigrants' cultural learning in the Canadian setting entails internalizing provincial-level political norms and values. A large body of empirical research demonstrates that local interpersonal communication networks are vital to the formation of political attitudes and behaviours. People tend to develop political attitudes that are consistent with the local majority opinion that surrounds them daily (Huckfeldt et al., Reference Huckfeldt, Beck, Dalton, Levine and Morgan1998); and people still favour interpersonal communication over other means of acquiring political information (Beck et al., Reference Beck, Dalton, Greene and Huckfeldt2002; McClung, Reference McClung2003). Evidence from internal migrants in the United States, for example, indicates that, when it comes to racial attitudes and partisanship, people who move to new environments (states or neighbourhoods) tend to adopt attitudes that resemble those of the local surrounding population (Glaser and Gilens, Reference Glaser and Gilens1997; McBurnett, Reference McBurnett1991; MacKuen and Brown, Reference MacKuen and Brown1987; Brown, Reference Brown1981; Markus, Reference Markus1979). The precise dynamics of this acculturation process remain somewhat unclear, but one possibility is that immigrants take on the norms and attitudes of their new local environments to lower the costs of “fitting in” to their new contexts (MacKuen and Brown, Reference MacKuen and Brown1987; Huckfeldt and Sprague Reference Huckfeldt and Sprague1987; Huckfeldt et al., Reference Huckfeldt, Beck, Dalton and Levine1995). Applied to the Canadian case, this line of reasoning implies that immigrants will adopt provincial and federal loyalties that mirror those of the local population within their respective provinces and that interprovincial differences in political loyalties among Canadian-born citizens are consequently reproduced among immigrants.
The dynamics of immigrants' acculturation, however, are quite different from those of political socialization among non-immigrant Canadians, and this encourages the expectation that immigrants might hold political views that are distinct from those of the local population within their respective provinces.
The inculcation of core political norms and values through such socialization agents as schools and the family typically occurs during childhood and adolescence, the early stages of the life cycle (Greenstein, Reference Greenstein1965; Hess and Torney, Reference Hess and Torney1967; Easton and Dennis, Reference Easton and Dennis1969). But immigrants typically arrive during later stages in the life cycle. Thus immigrants have not necessarily been inculcated with the same core values and norms as those found within the local populations of their new host settings. Indeed, when it comes to new immigrants who originate from non-traditional source countries, it is more likely that the discrepancies between the value sets of immigrants and native-born Canadians will be greater than the differences between immigrants from traditional source countries and native-born Canadians.
Furthermore, immigrants often have different preoccupations than the local population both upon arrival (for example. employment and housing) and subsequently when faced with the challenges of adapting to the new host society (such as discrimination and marginalization). It is these distinct sets of concerns that frequently encourage immigrants to join local networks and associations for mutual support. Thus, studies of immigrants' integration show that ethnic networks and associations play a powerful role for newcomers (Fennema and Tillie, Reference Fennema and Tillie1999).
Given these different points of reference and distinct settlement challenges, immigrants face choices during the course of adapting to new political settings. The evidence seems to be that immigrants select adaptive strategies that strike a balance between the desire to fit into the new host society and ensure a successful settlement and the desire to retain their own distinctive norms and values (Berry et al., Reference Berry, Kim, Minde and Mok1987; Berry, Reference Berry1997). As Berry explains:
In all plural societies, cultural groups and their individual members, in both the dominant and non-dominant situations, must deal with the issue of how to acculturate. Strategies with respect to two major issues are usually worked out by groups and individuals in their daily encounters with each other. These issues are cultural maintenance (to what extent are cultural identity and characteristics considered to be important, and their maintenance strived for); and contact and participation (to what extent should they become involved in other cultural groups, or remain primarily among themselves).
(Reference Berry1997: 9)The development of political loyalties among immigrants is thus likely to be influenced by adaptation strategies and social network dynamics. These strategies and dynamics, we suggest, increase the chances that new immigrants might acquire distinctive federal–provincial outlooks. One possibility is that immigrants might be inclined to internalize political outlooks that favour the federal pole. The Canadian federal government's multiculturalism policy offers immigrants an alternative pathway to social and political integration by legitimizing the retention of aspects of immigrants' pre-migration identity (Kymlicka, Reference Kymlicka1995, Reference Kymlicka1998). The symbolic value of this policy of multiculturalism, which acknowledges and accepts a multiplicity of cultural models, thus presents attractive re-socialization alternatives to newcomers in Canada. The multiculturalism policy, in effect, not only validates immigrants' pre-migration cultural identity and characteristics, but in doing so also lowers the cost of fitting in socially, economically and culturally.
The implication of this alternative line of speculation is that, irrespective of their province of residence, immigrants may be more inclined to develop federal loyalties rather than provincial ones. Furthermore, given the shorter period of residence and the greater cultural and ethnic differences with the Canadian-born population, the multicultural model might be even more attractive to immigrants from non-traditional source countries than their counterparts from traditional source countries. If that were the case then these immigrants might be more likely than others to develop stronger federal loyalties, thus attenuating interprovincial differences in political loyalties.
Research Design and Data
The analysis proceeds in two stages. The first stage focuses on the question of whether immigrants in each province are systematically more likely than their Canadian-born co-residents to gravitate towards the federal pole. More specifically, it examines whether immigrants in each province adopt political loyalties that are more provincially or federally oriented than the local population in the province where they reside. The second stage of the analysis assesses whether provincial cleavages in political loyalties—systematic interprovincial differences in federal–provincial orientations—are more or less similar among Canadian-born citizens and immigrant Canadians. In this instance, the focus is on whether immigrants reproduce, or alter, regional cleavages in Canada. If the cleavages observed among immigrants across all four provinces are similar to those observed among the Canadian-born population, then the conclusion would be that immigrants simply reproduce existing regional cleavages. However, if the cleavages observed among immigrants across all four provinces vary substantially from those observed among the Canadian-born population, then the implication is that contemporary immigration patterns do have the potential to alter regional cleavages.
The analysis focuses on four dimensions of political loyalties. Following Clarke and colleagues (Reference Clarke, Jenson, LeDuc and Pammett1980) the first three dimensions directly concern respondents' relationship to Canada, the federal government and the provincial governments. Western Canadian citizens' alienation from Canada's centre and Quebec citizens' alienation from the federal government are well documented (Clarke et al., Reference Clarke, Jenson, LeDuc and Pammett1980; Gibbins, Reference Gibbins1982; Henry, Reference Henry, Young and Archer2002). There has been longstanding debate about the integration and powers of some regions or provinces within the federation as well as about the real or perceived unequal treatment that provinces received historically from the federal government. Do immigrants in each province reflect the same kinds of “alienation” from the federal government, and possibly weaker attachment to Canada, than their respective Canadian-born provincial counterparts?Footnote 2
The first indicator measures an affective dimension of political loyalties. Following Clarke and colleagues (Reference Clarke, Jenson, LeDuc and Pammett1980) we examine respondents' feelings toward Canada and their province of residence. Here, the focus is on the differences in immigrants' feelings toward Canada and the respondents' province. The first dependent variable is thus captured by a scale (ranging from −1 to +1) indicating whether respondents express a more positive feeling toward Canada than toward their province (>0), a more positive feeling toward their province than toward Canada (<0), or equal feelings toward Canada and their province (=0). The next two indicators measure evaluative dimensions of political loyalties. The second indicator captures the gap in respondents' levels of political confidence between the federal and provincial governments. This scale (ranging from −1 to +1) indicates whether respondents express more confidence in the federal government than in the provincial government (>0), more confidence in the provincial government than in the federal government (<0), or equal confidence in the federal and provincial governments (=0). The third indicator measures respondents' perceptions of whether their province is treated better (+1), worse (−1) or about the same (0) as other provinces by the federal government.
Following conventional practice (Clarke et al., Reference Clarke, Jenson, LeDuc and Pammett1980), we also include a behavioural dimension of political loyalties and examine which party immigrants tend to support. The conventional wisdom is that immigrants typically support the Liberal party of Canada (Blais, Reference Blais2005; Bilodeau and Kanji, Reference Bilodeau, Kanji, Stephenson and Anderson2010), but what is far less clear is whether that support is uniformly distributed or varies substantially across provinces. Given that recent elections have produced major regional variations in partisan preferences (Gidengil et al., Reference Gidengil, Blais, Nevitte and Nadeau1999; Blais et al., Reference Blais, Gidengil, Nadeau and Nevitte2002; Nevitte et al., Reference Nevitte, Blais, Gidengil and Nadeau2000) the relevant question to ask is whether partisan support among immigrants follows those regional lines or whether immigrants are more likely to vote Liberal regardless of region of residence. Precisely why immigrants have tended in the past to rally behind the Liberal party is not entirely clear but some scholars speculate that the Liberal party might attract immigrant voters because it is perceived as the party most committed to promoting the multiculturalism agenda in Canada, an issue potentially important to immigrants when making their vote choice (Bilodeau and Kanji, Reference Bilodeau, Kanji, Stephenson and Anderson2010).Footnote 3 Consequently, our focus is specifically on provincial variation in immigrants' voting for the Liberal party of Canada.Footnote 4
These questions are empirically explored with pooled data from the 1993, 1997, 2000, 2004 and 2006 Canadian Election Studies (CES).Footnote 5 For reasons of sample size and population distribution, the analyses are limited to the immigrant-rich provinces of Ontario, Quebec, Alberta and British Columbia. Table 1 presents the sample composition for each group of respondents for each of the four provinces. The analyses control for such socio-demographic variables as sex, age, education, income and employment to take into account the possibility that potential differences between immigrants and Canadian-born respondents might be attributable to socio-structural variation between these subpopulations. Furthermore, because it is possible that feelings toward Canada and the provinces, levels of confidence in provincial and federal governments, perception of the province treatment by the federal government and the propensity of voting for the Liberal party of Canada varied across elections between 1993 and 2006, the analyses also control for the election year in which respondents were interviewed. (For detailed information about variable construction, see appendix A.)
Table 1 Sample Distribution and Size by Province
Source: 1993, 1997, 2000, 2004 and 2006 Canadian Election Studies
Immigrants' Federal and Provincial Loyalties
The first step entails determining whether immigrants in each of the four provinces exhibit political loyalties that are more provincially or federally oriented than the Canadian-born population in the province where they reside. The descriptive findings are presented in Table 2 and the summary results from the multivariate analyses, which compare for each province the political loyalties of immigrants from traditional and non-traditional source countries with those of Canadian-born citizens, are presented in Table 3. The full multivariate results are reported in appendix B.
Table 2 Outlooks of Immigrants and the Canadian-born Respondents
1 Mail-back component of the Canadian Election Studies.
Table 3 Differences between Immigrants and Canadian-born Respondents by Province
Source: 1993, 1997, 2000, 2004 and 2006 Canadian Election Studies
Entries report unstandardized B coefficients based on OLS regressions for Canada and province feeling, gap in confidence between federal and provincial governments and province treatment received by federal government and B coefficient based on logistic regression for Liberal voting. Analyses control for age, sex, education, income, employment status and year of interview. See appendix B for full tables.
***: B significant at p<.01;
**: B significant at p<.05;
*: B significant at p<.10.
Data reported in Table 2 indicate that there are significant differences in respondents' feelings towards Canada and their province between the immigrant and Canadian-born population in Quebec. About 23 per cent of the Canadian-born population in Quebec express more positive feelings toward Canada than Quebec, but those proportions climb to 43 per cent and 39 per cent among immigrants from traditional and non-traditional source countries respectively in that province. The multivariate analysis reported in Table 3 support these descriptive results. Everything else being equal, immigrants from traditional source countries score, on average, more than .10 point higher than Canadian-born residents of Quebec on the −1 to 1 scale. And immigrants from non-traditional source countries score about .08 point higher on average than the Canadian-born Quebec population.
That pattern, however, is not replicated in the other provinces under consideration. For example, there is no discernable difference between immigrants from traditional source countries and the Canadian-born populations in Ontario, Alberta or British Columbia, or between immigrants from non-traditional source countries and the Canadian-born population in British Columbia. The only other significant difference to emerge is in Ontario, where according to the multivariate analysis, immigrants from non-traditional source countries express less positive feelings for Canada than for their province in comparison to the local population (.03 point lower than their Canadian-born provincial counterpart). In Alberta, immigrants from non-traditional source countries also appear more likely to express positive feeling toward Canada than the local Albertan population (see Table 2) but this finding is not replicated in the multivariate analysis (see Table 3).
Thus, the Quebec results diverge significantly from others when it comes to the affective loyalty of immigrants. Immigrants from both traditional and non-traditional source countries, express more federally oriented loyalties than the Canadian-born population of their respective province. This result may not be so surprising given that it is in Quebec, according to the descriptive results, that we find the largest proportion of the Canadian-born population expressing more positive feelings for their province than for Canada: 42 per cent of Canadian born Quebecers view their province more positively than Canada, compared to 21 per cent of Canadian-born Albertans or British Columbians and 9 per cent of Ontarians.Footnote 6 It is nevertheless striking to see that this more positive feeling toward Quebec than Canada is not transferred very efficiently to immigrants; only 16 per cent of immigrants from traditional and 24 per cent from non-traditional source countries respectively express a more positive feeling for Quebec than for Canada.Footnote 7
When it comes to expressions of confidence in governments, there are no discernable differences between immigrants and local populations in each of the four provinces. Immigrants in Quebec and Ontario are somewhat more likely to express greater confidence in the federal than in their provincial government. But as the multivariate results reported in Table 3 indicate, these differences are not statistically significant. The only statistically significant difference observed in Table 3 concerns immigrants from traditional sources countries in Alberta; this group seems less federally oriented than the local population. When it comes to confidence in these institutions, immigrants are not, on balance, more federally oriented than their respective Canadian-born counterparts, even in Quebec.Footnote 8
The differences between immigrant and Canadian-born populations, however, are larger and more numerous when it comes to evaluations of how the federal government treats provinces. Immigrants from non-traditional source countries in Alberta and British Columbia supply significantly more favourable evaluations of the role played by the federal government than do Canadian-born respondents. Some 48 per cent of native-born Alberta residents and 51 per cent of immigrants from traditional source countries believe that Alberta receives worse treatment than other provinces from the federal government. That proportion drops to just 31 per cent among immigrants from non-traditional source countries. The distribution of responses in British Columbia, 58 per cent, 60 per cent and 47 per cent, respectively, replicate the same pattern and these findings are supported by the multivariate analyses in Table 3. Immigrants from traditional source countries in Alberta and British Columbia express similar evaluations to those of the local populations of the treatment received by their respective province from the federal government. Immigrants from non-traditional source countries in these two provinces, however, are significantly more inclined to express more positive evaluations than the local populations.
Once again, it is the Quebec findings that turn out to be most striking. In that province, immigrants from both traditional and non-traditional source countries are more likely than their Canadian-born co-residents to think that Quebec receives fair treatment from the federal government. Thirty-three per cent of non-immigrant Quebecers think that their province receives worse treatment from the federal government than other provinces, while 20 per cent of immigrants from traditional and 25 per cent of those from non-traditional source countries share that view. Only 10 per cent of non-immigrant Quebecers believe that their province receives better treatment from the federal government than other provinces while more than twice as many immigrants from traditional (22 per cent) and non-traditional (24 per cent) source countries hold that view.
Given the previous findings it comes as no surprise to discover that it is Ontarians who express the most positive evaluations of how the federal government treats their province. In that setting, 29 per cent of non-immigrant respondents think that the federal government treats their province better than other provinces. That finding contrasts sharply with the views held by comparable groups in Quebec (10 per cent), Alberta (5 per cent), and British Columbia (4 per cent). And it is only in Ontario that both groups of immigrants and the Canadian-born population share the same evaluations.Footnote 9
A similar pattern of differences between the immigrant and Canadian-born populations in each of the provinces emerges with respect to partisan support. Blais (Reference Blais2005) and Bilodeau and Kanji (Reference Bilodeau, Kanji, Stephenson and Anderson2010) demonstrate that Canadians of non-European origins are more likely than other Canadians to support the Liberal party of Canada. Our analyses indicate that this holds for the most part in the four provinces examined. Descriptive data reported in Table 2 indicate that the propensity to vote Liberal among immigrants from non-traditional source countries is 37 points higher than among the local population in Quebec, 18 points higher in British Columbia, 16 points higher in Ontario and 9 points higher in Alberta. Quebec emerges yet again as an outlier. Furthermore, Quebec also distinguishes itself from other provinces when it comes to the case of immigrants from traditional source countries. Immigrants from traditional source countries are almost as likely as Canadian-born respondents to express support for the Liberal party in British Columbia and Ontario and they are even less likely to do so in Alberta. Immigrants from traditional source countries in Quebec, however, are significantly more likely than the local Quebec population to support the Liberal party (by 24 percentage points). All of these findings are confirmed by the multivariate analyses.
These initial findings indicate that new Canadians, and more particularly newer waves of immigrants from countries with social and political systems that are vastly different from Canada's, tend to exhibit political loyalties that are more federally oriented than those of Canadian-born populations. They evaluate more positively the federal governments' treatment of their province and they are also more likely to support the Liberal party of Canada. The discrepancies between immigrant and Canadian-born populations are most pronounced in Quebec and generally least pronounced in Ontario than elsewhere. The most striking finding, perhaps, concerns the extent to which both groups of immigrants in Quebec differ from their Canadian-born counterparts in that province. Elsewhere, it is mostly only immigrants from non-traditional source countries that differ from the Canadian-born population. These initial results suggest that immigrant populations do have the potential to alter regional cleavages in loyalties to the federal and provincial governments.
Do Immigrants Reproduce Regional Cleavages in Political Loyalties?
The preceding analysis sheds some light on the question of immigrants' integration in regional dynamics but it provides only a partial view of how newcomers shape the dynamics of regionalism. Do immigrants reproduce regional cleavages in political loyalties? That question is explored by analyzing the data from each of our three subgroups of citizens (Canadian-born, immigrants from traditional and non-traditional source countries) in a multivariate setup. At issue is the direction and size of the differences in political loyalties between Canadian-born residents in Quebec, Ontario, Alberta and British Columbia and the question of whether these differences are replicated when immigrants of all four provinces are compared. For instance, if non-immigrant Albertans evaluate their provincial government more positively than their counterparts in Ontario, then the expectation is that immigrants in Alberta would also evaluate their provincial government more positively than their counterparts in Ontario. As before, these comparisons are undertaken for both groups of immigrants separately. The core findings are summarized in Table 4. The full specification is presented in appendix B. In these tables and analyses, respondents from Quebec, Alberta and British Columbia are each compared to those from Ontario.
Table 4 Differences between Provinces (by subgroup)
Source: 1993, 1997, 2000, 2004 and 2006 Canadian Election Studies
Entries report unstandardized B coefficients based on OLS regressions (and logit regression in the case of Liberal Voting) controlling for age, sex, education, income, employment status and year of interview. See appendix B for full tables. Coefficients correspond to predicted differences between each group of respondents (Canadian-born respondents, immigrants from traditional source countries, and immigrants from non-traditional source countries) with its counterpart in Ontario.
***: B significant at p<.01;
**: B significant at p<.05;
*: B significant at p<.10.
For the most part regional outlooks are reproduced among immigrants, notwithstanding the differences between immigrant and Canadian-born populations within provinces observed in the previous section of the analyses. Consider first the data concerning interprovincial differences in confidence in federal and provincial governments. The differences observed between non-immigrant Quebecers, Albertans and British Columbians with Ontarians are almost exactly replicated among both groups of immigrants. For instance, the gap between non-immigrant respondents of Quebec, Alberta and British Columbia with Ontario are respectively −.09, −.13 and .03 and those observed among immigrants from non-traditional source countries are respectively −.05, −.14 and .00. The structure of regional cleavages, then, appears to be replicated among newcomers.Footnote 10
The same broad findings emerge from the analysis of the three other indicators for Ontario, Alberta and British Columbia. First, there are only small differences in feelings toward Canada and the provinces among the non-immigrant populations and a similar pattern is observed among immigrants. Second, native-born Albertans and British Columbians tend to evaluate more negatively than local Ontarians the treatment their province receives from the federal government. The same holds among both types of immigrants in these respective provinces. And third, non-immigrant Albertans and British Columbians are less likely to vote for the Liberal party than their counterparts in Ontario. That same pattern is also reflected among immigrants of these respective provinces. The regional cleavages observed among immigrants in Ontario, Alberta and British Columbia on these three dimensions, however, do not precisely correspond to those of their native-born counterparts. Rather, the regional cleavages observed among immigrants tend to be more moderate than among the local populations when it comes to feelings toward Canada and the province, and when it comes to the perceptions of how the province is treated by the federal government. Moreover, the regional differences tend to be even less pronounced among immigrants from non-traditional source countries particularly when it comes to evaluations of province treatment by the federal government. Thus, for instance, while the predicted differences between local Albertans and Ontarians in perception of the treatment received by the province from the federal government is −.54, the predicted difference is −.47 for immigrants from traditional source countries and −.30 for immigrants from non-traditional source countries.
The final core finding that emerges from the data presented in Table 4 concerns Quebec. As with the initial findings, it appears that provincial loyalties are least efficiently transmitted from the Canadian-born population to both traditional and non-traditional immigrants in Quebec. Canadian-born Quebecers score approximately .12 point lower than their Ontario counterparts on the Canada/province thermometer. The corresponding gaps for immigrants from traditional and non-traditional source countries, however, are just .02 and −.01 points respectively, and they are not statistically significant. Similarly, Canadian-born Quebecers are significantly less likely than their Ontario counterparts to view the federal government's treatment of their province favourably (−.35). But there is no corresponding chasm between immigrants from traditional (.02) and non-traditional source countries (−.14). Finally, neither group of immigrants replicate the regional cleavages observed between local Quebecers and Ontarians when it comes to the matter of support for the Liberal party. Certainly, local Quebecers are significantly less likely to vote for the Liberal party than their Ontario counterparts (−.70) but the same cannot be said for immigrants from traditional (.27) and non-traditional source countries (−.07).
Explaining the Case of Quebec: Language Matters
Immigrants in Quebec, evidently, hold patterns of outlooks that are strikingly different from those of their counterparts in Ontario, Alberta and British Columbia. That finding requires closer scrutiny. And one candidate explanation for these variations might concern the role of language. To explore that possibility we compare 1) the political loyalties of immigrants who speak French at home to those of the native-born population in Quebec and 2) the political loyalties of immigrants who speak either English or another language (other than French) at home to the native-born population in Quebec.Footnote 11 In both comparisons we continue to distinguish between immigrants from traditional and non-traditional source countries. Table 5 reports the results of these multivariate analyses. These findings should be interpreted cautiously, given the modest samples of immigrants available for the analyses (N=186 and 214, respectively, for immigrants who speak French and English or another language).
Table 5 Difference between Immigrants and Canadian-born Population in Quebec
Source: 1993, 1997, 2000, 2004 and 2006 Canadian Election Studies.
Entries report unstandardized B coefficients based on OLS regressions for Canada and province feeling, gap in confidence between federal and provincial governments and province treatment received by federal government and B coefficient based on logistic regression for Liberal voting.
a: B significant at p<.01;
b: B significant at p<.05;
c: B significant at p<.10.
The analyses suggest distinct patterns of political integration, depending on which language immigrants speak at home. For three of the four types of orientations examined immigrants who speak English or another language at home exhibit political loyalties significantly different from those of the native-born population in Quebec. That pattern holds for both immigrants from traditional and non-traditional source countries. By contrast, immigrants who speak French at home exhibit political loyalties similar to those of the local population. There are no discernable differences in the orientations of French-speaking immigrants from traditional source countries and the native-born population in Quebec. And although French-speaking immigrants from non-traditional source countries are more likely than the native-born population of Quebec to support the Liberal party, the difference in party support is quite modest when compared to their counterparts who speak English or another language.
These tentative findings suggest that the dynamics of immigrants' integration in Quebec are different from those of immigrants in other provinces and that integration in Quebec appears to follow lines of linguistic integration. Immigrants who speak French at home tend to develop political loyalties that are similar to those of the local population. By contrast, those who speak English or another language at home tend to exhibit orientations that are more federal than those of the local population.Footnote 12 These findings are consistent with other research showing that, when it comes to support for the Parti Québécois or support for Quebec sovereignty, the patterns of support within French-speaking ethnic communities are more similar to those of the native-born Quebec population than those from other ethnic communities who do not speak French (Lavoie and Serré, Reference Lavoie and Serré2002).
Concluding Discussion
Citizen outlooks towards their federal and provincial governments are a primary prism for understanding the dynamics of regionalism in Canada. Most Canadians are socialized to these dualities, but what about new Canadians? The significant changes in the scale, distribution and composition of Canada's immigrant population raise the question of how these changes contribute to the dynamics of regionalism in the country. The analysis began by investigating whether immigrants adopt political outlooks that are more federally or provincially oriented than the Canadian-born population. It then turned to evaluate whether new immigration dynamics have the potential to transform the structure of cleavages in federal–provincial orientations between Canada's provinces. Almost 30 years ago, Elkins reported that immigrants who arrived in Canada in the 1940s and 1950s exhibited relatively weak regional differences in such core orientations as political efficacy and trust. Canadian immigration patterns have changed quite substantially since then. Even so, the conclusions emerging from this analysis corroborate Elkins' findings with respect to other core features of political culture. For the most part, immigrants do reproduce the structure of interprovincial cleavages. But a significant caveat is in order: the cleavages observed among immigrants, especially those among immigrants from non-traditional source countries, are weaker than those found among the Canadian-born population. Contemporary immigration thus appears to have the potential to continue to slowly erode regional cleavages.
We presented at the outset two competing sets of expectations about immigrants' acquisition of federal–provincial orientations. The first possibility explored was that immigrants' federal–provincial loyalties essentially reflect those of their new provincial context. The alternative possibility was that immigrants acquire loyalties that are more federally oriented than those of the local population, regardless of province of residence. On balance, the data support the latter interpretation: immigrants tend to exhibit political loyalties that are somewhat more federally oriented than those of the Canadian-born population. And these federally oriented outlooks are particularly striking among a growing segment of the immigrant population, namely, those coming from non-traditional source countries. However, the fact that immigrants develop somewhat stronger federal loyalties than the Canadian-born population in their respective provinces does not imply that they are completely impervious to local dynamics. The analyses indicate that in spite of the difference between immigrants and their corresponding Canadian-born provincial population, there is clear evidence of a strong reproduction of regional cleavages.
The expectation that immigrants might be inclined to evaluate the federal pole more positively was informed by the special features of immigrant adaptation. If immigrants choose strategies for adjusting to new environments that lower the costs of adaptation, (see Berry et al., Reference Berry, Kim, Minde and Mok1987; Berry, Reference Berry1997), then the federal governments' multiculturalism policy might offer them a promising pole of identification. Immigrants, especially those from non-traditional source countries, plausibly might identify with the policy of multiculturalism of the federal government, a policy that explicitly recognizes cultural specificity and encourages the retention of cultural difference. This attraction would then lead to greater attachment to the federal pole of loyalty. The data, however, do not allow us to pinpoint the specific reasons for why newcomers develop stronger loyalties to the federal government and so the precise origins of this stronger attachment remain open for debate and for future research.
The Quebec case suggests, clearly, that regardless of the salience of the multiculturalism policy explanation, other factors are at play in immigrants' integration dynamics. Of the four provinces examined, Quebec turns out to exhibit exceptional patterns. As in other provinces, Quebec immigrants from non-traditional source countries exhibit political loyalties more federally oriented than those of the local population. And uniquely, immigrants from traditional source countries in Quebec also exhibit more federally oriented loyalties.
The Quebec case seems to present counterintuitive findings. Of all Canadian provinces it is Quebec that most actively promotes policies to ensure immigrants' integration. Bill 101 requires French education. Quebec was also the first province to sign an agreement in 1991 with the federal government giving the province a greater role in the selection of their immigrants. And through this agreement Quebec has the opportunity to favour francophone immigrants. Most significantly, perhaps, Quebec has a policy of interculturalisme to address its cultural and ethnic diversity, one that is substantively similar to the federal government's multiculturalism policy (Kymlicka, Reference Kymlicka1998: 67–68; Gagnon and Iacovino, Reference Gagnon, Iacovino and Gagnon2004). The collective impact of these policies might facilitate more efficient integration to Quebec society. Yet it is in Quebec that the immigrant population carries political outlooks that are most at odds with those of the native-born provincial population, and in that sense it is in Quebec that immigration has the greatest potential to attenuate Canadian regional cleavages.
Our analyses of the role of language suggest that the absorption of federal–provincial loyalties in Quebec is consistent with patterns of linguistic integration. Immigrants who speak French at home develop political loyalties that are almost indistinguishable from those of the local population while those of immigrants who speak English at home or another language are significantly more federally oriented. A determining factor in Quebec, then, appears to be immigrants' choice of which linguistic community they join.
One question raised by these findings concerns the matter of whether the impact of immigration on regional dynamics is short-lived or long-term. To answer this question definitively required a detailed investigation of whether immigrants' political loyalties evolve or remain more federally oriented the longer they reside in Canada. To this point, the limits of the data make it difficult to address that question directly or in detail.Footnote 13
Canadian political culture has been substantially shaped by the series of immigration waves that settled in the country in the last centuries. Whether they came from France, the British colonies, Central and Eastern Europe and Asia, immigrants have made a significant contribution to determining how Canadians relate to politics and their political institutions. This paper presents evidence suggesting that today's immigrants, like their predecessors, are indeed forging contemporary regional dynamics in Canada. Immigrants, especially newer waves from non-traditional source countries, seem to develop somewhat stronger federal political loyalties, even if they also absorb regional political norms to a significant degree. The specific reasons why this is so and whether these differences are sustained or diminish with time, remain to be demonstrated, but in the meantime it appears that immigration to Canadian provinces has some potential to dilute regional cleavages in federal–provincial political orientations as Elkins (Reference Elkins, Elkins and Simeon1980) had shown close to three decades ago. It strengthens what LaSelva (Reference LaSelva1996) characterized as the “will to live together” within the Canadian confederation.
Appendix A: Construction of Variables
1993–2006 Canadian Election Studies
Appendix B
Table B1 Difference between Immigrants and Canadian-born Populations in Gap in Canada and Province Feelings
Source: 1997 and 2000 Canadian Election Studies. a: P<0.01; b: P<0.05; c: P<0.10.
Entries report OLS unstandardized coefficients.
Table B2 Difference between Immigrants and Canadian-born Population in Gap in Confidence between Federal and Provincial Governments
Source: 1993, 1997, 2000 and 2004 Canadian Election Studies.
a: P<0.01;
b: P<0.05;
c: P<0.10 Entries report OLS unstandardized coefficients.
Table B3 Difference between Immigrants and Canadian-born Population in Evaluations of province Treatment by the Federal Government
Source: 1993, 1997, 2000, 2004 and 2006 Canadian Election Studies.
a: P<0.01;
b: P<0.05;
c: P<0.10 Entries report OLS unstandardized coefficients.
Table B4 Difference between Immigrants and the Canadian-born Population in Liberal Voting
Source: 1993, 1997, 2000, 2004 and 2006 Canadian Election Studies.
a: P<0.01;
b: P<0.05;
c: P<0.10
Entries report logit estimates.
Table B4.1 Reproduction of Regional Cleavages among Immigrants in Canada (Part 1)
Source: 1997 and 2000 Canadian Election Studies for the “Gap in Canada Province Feelings” and the 1993, 1997, 2000 and 2004 Canadian Election Studies for the “Gap in Confidence between Federal and Provincial Governments”.
a P<0.01;
b P<0.05;
c P<0.10 Entries report OLS unstandardized coefficients.
Table B4.2 Reproduction of Regional Cleavages among Immigrants in Canada (Part 2)
Source: 1993, 1997, 2000, 2004 and 2006 Canadian Election Studies.
a: P<0.01;
b: P<0.05;
c: P<0.10
Entries report OLS unstandardized coefficients for “Province Treatment” and Logit estimates for Liberal Voting.
