Ghettos in Canada's Cities? Racial Segregation, Ethnic Enclaves and Poverty Concentration in Canadian Urban Areas

Walks, R. Alan, Bourne, Larry S., The Canadian Geographer

Introduction

Are there urban ghettos in Canada? There has been increasing interest in analyzing the impacts of growing visible minority populations in Canadian cities, particularly in terms of their links to levels of concentrated poverty and neighbourhood distress. Recent reports commissioned by municipal governments and service agencies (Federation of Canadian Municipalities 2003; United Way of Greater Toronto 2004) as well as by Statistics Canada (Heisz and McLeod 2004) suggest that poverty (or more accurately, low income) is not only growing in Canadian cities but is becoming increasingly concentrated in poor neighbourhoods. Not unsurprisingly, the spatial concentration of visible minorities, Aboriginals and recent immigrants is cited as one of a number of potential factors underpinning the growth of concentrated urban poverty. Information contained in the United Way report, 'Poverty by Postal Code' (United Way of Greater Toronto 2004, 49-50), for instance, suggests that the growth in visible minority families may explain all of the growth in family poverty within the City of Toronto between 1981 and 2001, since the level of low income rose both in the city at large and among visible minority families, but declined for non-visible minority families. According to this report, visible minority families made up 77.5 percent of the poor families residing in high poverty neighbourhoods in 2001, double the level in 1981.

This raises the spectre of ghettoization emerging within Canadian cities along the lines witnessed in the United States, a spectre fuelled by media reporting of violent crimes potentially linked to mmorities and to gangs, particularly in Toronto. (1) The relationship between visible minority concentration and high-poverty neighbourhoods in Canada, however, remains under-examined, with most studies concerned with segregation conducted by a small number of devoted sociologists and geographers (for example, Darroch and Marston 1971; Balakrishnan 1976, 1982; Clarke et al. 1984; Ray and Moore 1991; Murdie 1994a; Fong 1996; Bauder and Sharpe 2002; Fong and Wilkes 2003; White et al. 2003, 2005; see also Walks 2001). Much of the literature, and the dominant discourse, concerning urban ghetto formation has emanated from the United States where racial segregation, particularly of the black population, has been an overriding concern. studies undertaken in the United States have found not only very high levels of spatial segregation for blacks and Hispanics, but strong neighbourhood effects that grow with the level of racial concentration. In the United States living in a highly segregated neighbourhood not only increases the chance that one is already pour, but also limits the ability of residents to escape poverty due to, among other things, a lack of social networks, locally based resources, and access to employment (Wilson 1987; Massey and Denton 1993; Jargowsky 1997; Ihlanfeldt 1999). Although the most recent U.S. census shows a decline in both residential segregation and neighbourhood poverty, this change has mostly occurred in cities with few blacks or Hispanics to begin with: in cities with larger black populations there has been much less change (Jargowsky 2003; Kingsley and Pettit 2003).

Hajnal (1995) was one of the first to raise the alarm about the growth of neighbourhood poverty in Canada. He showed that a higher proportion of Canadians lived in high-poverty neighbourhoods in 1986 than did residents of the United States in 1981, although he rightly pointed out that the low proportion of visible minorities in such neighbourhoods at the time meant that the main source of this difference lay elsewhere. Fong and Shibuya's (2000, 2003) research using the 1986 and 1991 census data continued to uncover a relationship between visible minority concentration and neighbourhood poverty. Yet, it is Kazemipur and Halli's (2000) work that has perhaps been most influential in bringing the discourse of ghettoization to Canada: their provocative use of the subtitle 'Ethnic Groups and Ghetto Neighbourhoods' implied that Canada was witnessing the birth of urban underclass ghettos directly linked to growing ethnic communities. Analyzing census data to 1991 for Census Metropolitan Areas (CMAs) they suggest that Aboriginals and recent immigrants from Africa, Asia, Latin America and the Caribbean in particular are increasingly likely to live in 'ghetto' or 'underclass' neighbourhoods. This result stands in sharp contrast to other studies of ethnic and racial segregation trends that suggest that they du not mimic the pattern of ghettoization found in the United States (Fong 1996; Balakrishnan 2001; Bauder and Sharpe 2002; Balakrishnan and Gyimah 2003; Myles and Hou 2004; Peters 2005).

Each of these studies, however, uses different definitions (of ghettos, concentrated poverty, etc.) and a different methodology to arrive at their conclusions, and the methods used often du not permit direct comparison with existing U.S. and international studies. Furthermore, the effect of changes witnessed in the most recent census period is not yet known. While a positive torn in the business cycle has led to a decline in concentrated low income in most CMAs in Canada between 1996 and 2001, the benefits are shared unequally with some cities worse off than a decade earlier (Heisz and McLeod 2004). It remains unclear whether the growth of the visible minority population (that mainly results from changes in the source countries of immigrants to Canada as well as higher birth rates among certain minority groups), has led to their spatial integration or segregation, and whether such spatial changes are linked to the patterning of high-poverty neighbourhoods.

This paper seeks to answer these questions. It explores the relationship between the spatial concentration of visible minorities and the growth of neighbourhood poverty, using information from the 1991 and 2001 censuses at the level of census tracts. In so doing it simultaneously updates and expands upon the existing Canadian literature on the subject in two directions (see also Hiebert 2000; Bauder and Sharpe 2002). First, to examine the question of ghetto formation, a neighbourhood classification system is adopted that has been recently developed precisely for the task of comparing levels of neighbourhood isolation, ghettoization and integration across national contexts (Poulsen et al. 2001; Johnston et al. 2003). This classification scheme allows for a comparative test of whether ghetto formation is also occurring in Canada, and whether neighbourhood poverty rises as visible minority concentration increases. The relative importance of minority concentrations for predicting the spatial patterning of high-poverty neighbourhoods in comparison with other factors is then ascertained using regression analysis undertaken for the most highly segregated metropolitan areas.

We are aware that this analysis sits at the intersection of parallel research interests in the study of immigrants, ethnic communities and racialized groups. These groups are not the same, although there is considerable overlap as most recent immigrants also qualify as members of distinct ethnic and racial communities. It should be noted, however, that the concept and meaning, and the formal categorization, of visible minorities are themselves problematic, and such labels tend to obscure a considerable diversity in the character and living conditions of the population involved. Despite these misgivings, our focus is on the visible minority population since it is the increasingly widely used currency in the census data and in social science research. (2)

The Geography of Assimilation and Exclusion? Enclaves, Communities and Ghettos

A situation of increasing neighbourhood concentration of visible minority groups violates traditional ecological models that see immigrants and ethnic groups integrating geographically as they assimilate culturally (Park et al. 1925). While congregation in a particular district may have temporary benefits for a particular group, continued or increasing concentration implies a breakdown of the assimilation process and/or social exclusion on the part of the 'host' society (Philpott 1978). However, a more nuanced approach characterizes recent thinking, particularly concerning the growth of 'EthniCities' and multiculturalist/ pluralist policy (Clarke et al. 1984; Roseman et al. 1996). Long-term ethnic concentration may help promote cultural goals and group identity (Peach 1996; Boal 2005), and may be strategic in light of increasing transnationalism and/or the global marketing of ethnic spaces (Lin 1998; Ong 1999; Walton-Roberts 2003), and thus should not be viewed as necessarily negative (Qadeer 2005). It is important to differentiate between induced, involuntary and strategic forms of spatial concentration--between ghettoization, which the literature invariably views negatively, and other forms including the growth of traditional enclaves dominated by a single ethnic group, which is now often looked upon as potentially beneficial.

The traditional definition of the ghetto in the U.S. context is of a residential district that both concentrates a particular racial or ethnic group and at the same time contains it, in that a majority of its members are forced to live there due to discrimination on behalf of the host community (Philpott 1978; Massey and Denton 1993; Peach 1996; Jargowsky 1997; Marcuse 1997; Logan et al. 2002; Johnston et al. 2003; Pattillo 2003). Such ghettos are produced through race-based discrimination in the housing and labour markets. This is a different definition of the ghetto than that employed by Wilson (1987) who examines the increase in high poverty levels among black and Hispanic neighbourhoods, with residents often referred to as the 'ghetto poor'. Bridging the two, Marcuse (1997) suggests that as a result of industrial decentralization and globalization, a new form of 'outcast ghetto' (distinguished from the ghetto of old) may be emerging in U.S. cities, composed only of the poorest segments of subjugated racialized groups (mostly blacks and Hispanics) who are marginal to current production needs.

A number of authors contrast the ghetto with the ethnic enclave in which residency appears voluntary and members have the option of leaving. Enclaves, therefore, crystallize because they conform at least partly to the needs of a minority group, while ghettos are formed through exclusion on behalf of the host society against the interests of its residents and from which they cannot easily escape (Marcuse 1997). Logan et al. (2002) distinguish further between the immigrant enclave of old, typically seen as a temporary neighbourhood of convenience containing ethnic resources to be drawn upon until immigrants assimilate into the host society and relocate, and an emerging new ideal type, the 'ethnic community'. Unlike the immigrant enclave, many (but certainly not all) ethnic and cultural groups view the ethnic community as the desired residential endpoint, typically a neighbourhood with a single group dominant and at the same time relatively prosperous (Logan et al. 2002). Finally, there are the citadels, the isolated exclusive neighbourhoods formed by the elite class of the host community for their own benefit (Marcuse 1997).

Confusion over these definitions is one reason for the lack of comparability among Canadian research on the topic. For example, many studies that conclude that there is little evidence of ghetto formation in Canadian cities are based on an examination of indices of dissimilarity or exposure to own-group members, a method that cannot distinguish between ghettos and enclaves (Balakrishnan 2001; Bauder and Sharpe 2002; Balakrishnan and Gyimah 2003; Myles and Hou 2004). On the other hand, Kazemipur and Halli (2000) parrot Wilson (1987) in employing the terms 'ghetto', 'underclass' and 'high poverty neighbourhood' interchangeably, defining them as any census tract with an incidence of low income above 40 percent. Such a practice, however, impedes clarity when it is the relationship between ethnic segregation and concentrated poverty that is in question. As yet, few Canadian studies have employed strict criteria that would allow one to distinguish between the growth of ethnic enclaves and contained ghettos.

Recently, Johnston et al. (2002) have presented explicit criteria for classifying urban neighbourhoods. They employ an approach of absolute floors and ceilings, rather than relative measures and indices, in order to compare situations across time and space. According to the authors, this classification is 'a robust approach to comparative study' across international contexts, which in torn is 'directly linked to the homogeneity-heterogeneity continuum which underpins all studies of segregation' because it simultaneously identifies three different types of spatial segregation proposed by Massey and Denton (1993): isolation, clustering and concentration (Poulsen et al. 2001, 2071). Such a classification allows for the examination of a number of interesting questions, including the proposition that 'outcast' ghettos or 'ethnic communities' could be forming among particular minorities regardless of the direction of movement in overall segregation indices. This approach facilitates a more complex investigation into the importance of visible minority concentration for our understanding of the growth of concentrated neighbourhood poverty.

This classification has so far been used to compare British cities, and cities in Australia and New Zealand, to the U.S. case (Poulsen et al. 2001; Johnston et al. 2002), and to analyze differences between U.S. cities over time (Johnston et al. 2003). Under this definition, ghettos such as those in New York, Los Angeles, Chicago and Miami, were found only to exist in a few smaller British cities, namely Leicester, Oldham and Bradford, with South Asians being the most concentrated in such areas. There is no evidence of ghettoization in larger British cities such as London, nor in Australia or New Zealand, countries that are said to have adopted a similar policy of multicultoralism to the Canadian model (Smolicz 1995, 592; Johnston et al. 2002). It is not yet known how Canadian cities compare to these others.

Immigration, Visible Minority Neighbourhoods and Concentrated Poverty in Canada

It is clear that changes in immigration policy implemented in the late 1960s have had an increasingly disproportionate impact on the face of urban Canada. Between 1971 and 2001, the proportion of Canada's population that had been born in Asia, Africa, the West Indies, or Latin America rose from 1.7 to 10.4 percent, while visible minorities grew from 4.7 to 13.4 percent of the population between 1981 and 2001 (Statistics Canada 2003, 10). Changes in the source countries of immigrants have been accompanied by their increased concentration within Canada's census metropolitan areas (CMAs) (Hou 2004; Simmons and Bourne 2003). (3) Recent immigrants, in particular, have increasingly preferred to move to the largest urban regions since 1971, when 50.3 percent of foreign-born residents lived in one of the five largest cities. By 2001, this proportion had risen to 80.4 percent, with a full 43 percent of immigrants arriving in the preceding 10 years moving to the Toronto CMA (Statistics Canada 2003, 7). Accordingly, visible minorities have become highly concentrated in the largest cities and their suburbs. Both the Toronto and Vancouver CMAs went from having just under 14 percent to just under 37 percent of their population classified as visible minorities between 1981 and 2001 (Statistics Canada 2003, 44). (4) Meanwhile, the Aboriginal population, which is treated as a distinct group in the census, is becoming more concentrated in a number of Prairie cities, particularly Saskatoon, Regina and Winnipeg.

While the Canadian media may have constructed immigrant settlement patterns as a national policy issue (mostly in relation to regional development and the distribution of services) (Abu-Laban and Garber 2005), there is no consensus in the literature concerning whether the growth and concentration of visible minorities in Canadian cities is leading to their increased isolation from mainstream society. studies of ethnic and racial segregation have found that while certain groups are highly concentrated in certain neighbourhoods (Jewish, then South Asian groups were the most segregated), the levels of concentration for such groups were not as high as for either blacks or Asians in U.S. cities (Fong 1996; Balakrishnan and Hou 1999; White et al. 2003) and they changed little or declined between 1986 to 1996 (Balakrishnan 2001; Bauder and Sharpe 2002; Balakrishnan and Gyimah 2003).

Such trends have led some researchers to conclude that mobility and assimilation, rather than entrapment, exclusion or cultural separation, more commonly characterize the residential geography of immigrant groups in Canada (Ley and Smith 2000; Hiebert and Ley 2003). However, the results are decidedly uneven (Bauder and Sharpe 2002; Hou and Milan 2003), and recent findings concerning the relationship between low income, immigration and visible minority status raise some new concerns. While the earnings gap between both whites and visible minorities, and whites and Aboriginals, decreased during the 1970s and stabilized during the 1980s, it grew rapidly during the first half of the 1990s (Pendakur and Pendakur 2002). Morissette and Drolet (2000) found that members of visible minorities and immigrants who arrived after 1977 suffered disproportionately from low income during the early 1990s. Picot and Fou (2003) show that both low-income levels and gaps grew among recent immigrants between 1980 and 2000, particularly during the 1990s among immigrants from Asia and Africa.

Deterioration in earnings levels for recent immigrants occurred despite demographic shifts among recent immigrant cohorts (toward higher education and skill levels) that should have left them significantly better off, rather than worse, than earlier cohorts (Aydemir and Skuterud 2004). Part of the blame lies with declining returns to foreign work experience and the devaluation and non-recognition of foreign credentials (Bauder 2003). It is argued that this trend stems not only from an inability to transfer skills, but also from institutionalized forms of occupational exclusion as well as racial discrimination within the labour market (Bloom et al. 1995; Bauder 2003). (5) While recent immigrants arriving in the 1990s were found to be able to catch up to the Canadian-born faster than was the case in the past (Li 2003), this outcome is least true for visible minority immigrants, particularly refugees from Africa, Latin America and the Middle East who rarely ever attain occupational levels they enjoyed in their countries of origin (Krahn et al. 2000).

As a result of such shifts, Moore and Pacey (2003) found that much of the increase in income inequality experienced across Canada during the early 1990s may be accounted for by the growth of inequality and low income within recent immigration cohorts. Furthermore, as a result of the restructuring of Canada's set of taxes and transfers during the 1990s, welfare-state programs and benefits did not offset either the growth of low-income intensity (Picot et al. 2003) or income inequality (Frenette et al. 2004) that followed the severe recession of the early 1990s. This contrasts with the situation during the 1980s when the transfer and …

Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Ghettos in Canada's Cities? Racial Segregation, Ethnic Enclaves and Poverty Concentration in Canadian Urban Areas. Contributors: Walks, R. Alan - Author, Bourne, Larry S. - Author. Journal title: The Canadian Geographer. Volume: 50. Issue: 3 Publication date: Autumn 2006. Page number: 273+. © 2001 Canadian Association of Geographers. COPYRIGHT 2006 Gale Group.

This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.