Larry S. Bourne 

May 1999

 

Questions relating to the social sustainability of cities, however the term sustainability is defined, are invariably linked to changes in population. In most developed countries, fertility levels (i.e. rates of natural population increase) have declined to post-war lows, and at the same time have become more uniform over geographical space. As a result, the other two components of population growth and redistribution - internal (or domestic) migration and immigration - have assumed much greater importance as determinants of urban and regional growth, as sources of change in population composition, and as subjects of public policy concern. They are also the most uneven and variable components, and thus the most difficult to predict.

Despite a very extensive literature on migration and immigration, we know surprisingly little about the processes of population redistribution in general, and less about the geographical distribution of immigrants. We know even less about their movements after initial settlement and the reasons for their choice of destinations. Nor do we know very much about the relationships between immigration and internal migration, or about the intersection between these two flows and the maintenance of social stability and viability in those urban areas receiving large numbers of domestic and overseas migrants.

This paper examines recent trends in domestic migration and immigration in Canada, with an emphasis on the larger metropolitan areas that serve as the primary Agateways@ for immigrants, and with detailed examples drawn from the recent - and often dramatic - Toronto migration experience. The paper then explores the determinants of differential flows of migrants and immigrants, and develops a simple typology to illustrate the very different demography-migration-immigration regimes that characterize Canadian metropolitan areas. Third, the paper examines the relationship between immigration and migration and offers suggestions on the implications of these relationships, again with special reference to the experience of the Toronto metropolitan region. The paper concludes with examples of issues of public policy concern and offers suggestions for further research.

Why examine migration and immigration together? And why begin the analysis of particular cities, such as Toronto, with the broader framework of national trends and population shifts within the larger urban system? The rationale in both cases is the need for context. In the latter case the argument is that cities do not exist in isolation. They are, instead, part of a national system of urban places and their rates of population and employment growth, and changing social composition, reflect their position within that larger system (Bourne and Olvet 1995). In the former case, it is argued that immigration cannot be examined in isolation of other components of population growth - notably domestic migration. This is emphatically true if the subject of interest is social change and sustainability. Immigration, like migration, can be represented both as an attribute of a place, and as a flow. As flows both contribute to population growth and to the redistribution of labour supply, skills, capital income, market demand and voters across the country. But are immigration and migration flows related? And if so, how, where and why?

These questions are of fundamental importance for public-policy makers and social service providers, as well as for scholars of both immigration and migration. For example, if the assumption is made that new immigrants will subsequently spread out over the country, or at least among metropolitan areas, in a pattern roughly similar to that of domestic migrants, then it can be argued that policies directed at assisting their adaptation to (and integration with) Canadian society, and support for local governments, can be both modest in scope and short-term (i.e. transitional). If, on the other hand, newer immigrants do not disperse from the original Agateway@ centres in the years following their arrival in Canada, then a different set of policies are necessary to foster their adjustment experience (or coping strategies). Similarly, sustained policy initiatives are necessary to assist the receiving cities in meeting the social needs of these immigrants.

An equally important set of questions can be raised with respect to any linkages or relationships between immigration and flows of domestic migrants. If the relationship between domestic migration patterns and the destinations of immigrants is positive then it implies that immigration is adding to the already existing growth pressures facing the receiving cities. These pressures, in turn, increase the demands for housing and social services, and thus add to the costs of both. Moreover, immigration would be contributing to creating even wider differences in terms of employment opportunities among the winners and losers in the Canadian urban system. If, on the other hand, the relationship between the two flows is negative, then high levels of immigration would be assisting in reducing disparities in growth rates and in economic and social well-being among urban areas in Canada.

An examination of population redistribution then is part of the essential background for understanding the place-specific impacts and social consequences of immigration. It should be stressed, however, that this paper does not address immigration policy, nor does it debate whether a particular population policy or immigration target is appropriate or inappropriate, efficient or inefficient. Numerous other observers in Canada are engaged in this ongoing debate (Foot 1994, DeVoretz 1995; Beaujot 1998).

Canada has always been a nation of migrants and immigrants. Since 1820 Canada has accepted over 14 million immigrants, over 5.3 million since 1961 and over 2.4 million since 1981. The annual intakes tend to vary widely, largely as a result of public policy shifts and changing economic and political conditions in both Canada and the countries of origin. Since 1948 annual in-flows have ranged from a low of 72,000 (in 1961) to over 282,000 (in 1957), with a long term annual average of 150,000. After the mid-1970s the flow declined throughout the 1980s, then increased again to an annual average of over 220,000 during the 1990s. Whether this high level is itself sustainable remains to be seen.

Given declining levels of natural population growth, it is not surprising that the relative contribution of net immigration to Canada=s population growth has increased. Even during the height of the baby-boom period (1948-1963) when fertility levels (and marriage rates) were among the highest in the western world, and mortality rates declined, immigration represented at least 20-25% of national population growth. This proportion was roughly maintained during the 1970s and 1980s as fertility levels declined sharply (the so-called baby-bust period) while immigration was curtailed, in line with deteriorating economic conditions and increasing unemployment. During the 1990s, however, immigration levels increased substantially, and appear to have become detached from prevailing economic conditions. For the 1991-96 period immigration represented more than half (50.9%) of Canada=s population growth.

Aside from this variability over time, three other attributes of immigration are relevant here. One, immigration is highly concentrated geographically, telescoped on a few metropolitan areas that serve as initial destinations or Agateways@, and the degree of concentration has been increasing over time. As of 1996, 52.4% of all immigrants were resident in the three largest metropolitan areas of Toronto, Montreal and Vancouver, and over 60% in the five largest recipient cities (adding Ottawa-Hull and Calgary), compared to 33% and 39% of the country=s total population, respectively. Among recent immigrants (those arriving since 1991), 74% are resident in these three metropolises (43% in Toronto, 18% in Vancouver and 13% in Montreal), while over 80% are in the five largest recipient cities. In contrast, most areas of the country -particularly small cities and almost all areas east of Montreal - receive few immigrants, or none at all. There is little or no evidence to suggest that this pattern will change in the future, regardless of changes in immigration policy.

One obvious result of this spatial concentration of immigrants is that the level of social and cultural differentiation among Canadian urban areas has increased, and in some instances dramatically so. The implications of this increasing diversity and differentiation are not as yet clear, but they are likely to include a further fragmentation of urban Canada, at both macro- and micro- scales, into markedly heterogeneous or socially homogeneous environments for both living and working.

The relative contribution of immigration and migration to the growth of individual urban areas thus varies even more widely. Table 1 summarizes the contributions of the three components of population growth - natural increase, net internal migration and net immigration (excluding emigration)- to the growth of Canada=s major metropolitan areas for the two five-year census periods from 1986 to 1996. Note that for each component the figures do not represent actual flows but the relative (proportional) contribution of each to total growth. Thus, a place that is growing slowly or declining because of negative domestic migration could have all of its growth accounted for by immigration, even when the level of immigration is small. Note also that for all five of the major metropolitan areas, immigration is the largest single component of urban population growth, and that contribution increased between 1986-91 and 1991-96. For Montreal, net domestic out-migration exceeded natural increase in the early 1990s, and thus immigration represented over 100% of metropolitan population growth. For Toronto, foreign immigration accounted for over 92% of total growth, and in Vancouver over 79%. At the same time the Toronto region lost heavily in the national exchange of domestic migrants, while Vancouver gained from both sources.

Second, the origins of immigrants have shifted dramatically over the last two decades to countries that are non-traditional sources, and thus to more culturally and ethnically distinctive populations. Prior to 1961 over 90% of all immigrants came from Europe, with over 22% from the UK alone; since 1981 only 15% have arrived from Europe. Over 70% of immigrants to Canada in the last decade have come from countries in Asia, Africa, the Caribbean and Latin America (Table 2). The largest source countries for recent immigrants are now China, Hong Kong, Sri Lanka, the Philippines, and India.

The composition of the immigrant population has also shifted in other important ways, for example in terms of skill levels and resources, and among the three major Aclasses@ of immigrant - family, independent and refugee - but the latter shifts appear to be less significant than the increasing ethno-cultural and linguistic diversity. These new groups, in turn, are also not evenly spaced across the urban system with individual groups more concentrated in certain cities than in others. The combined result of the changing social and ethno-cultural characteristics of the immigrant population, and the extreme degree of geographical concentration, often in conjunction with the net out-migration of existing (native-born) residents, is that the large gateway cities are beginning to look and feel very different from cities and small towns in the rest of the country.

The third locational trend is the shift in immigrant populations toward the suburbs, notably the older inner suburbs but most recently to the new suburbs. These are areas where jobs and less expensive housing are more readily available. The traditional immigrant reception areas in the inner cities of urban Canada - although still evident in the landscape - are no longer the primary focus of immigrant destinations. Interestingly, the degree of suburbanization also varies widely among particular immigrant groups. In the Toronto case, the least suburbanized among recent immigrant groups are those from Somalia, Sri Lanka, Vietnam, the Philippines and Iran; the most suburbanized are those from India and Hong Kong. These diverse patterns are partly a reflection of differences in the income and educational levels of the immigrants on arrival, in part a reflection of the localizing effects of high proportions of linked or Achain@ migrations, and in part a reflection of timing. Late arrivals are, to some extent, deflected to the new suburbs by the relative absence of housing opportunities in the central city where vacancy rates are very low, at least in Toronto (and Vancouver). In many other regards, however, immigrants are behaving very much like intra-urban domestic migrants, adding to the rate at which the suburbs are growing while also being transformed - from the classical image of social homogeneity to something as yet uncertain, but clearly much more diverse.

The overall rate of internal migration in Canada, as in other developed countries, has been remarkably constant over time. Every five years almost 45% of all Canadians (or 11.5 million people) are reported as living at a different address, and about half of those (20.3%) move across municipal boundaries (i.e. become migrants). The overall rate (as a % of population) at which residents move is driven primarily by differences in demographic structure and seems to be largely independent of national economic conditions. There is evidence that the rate of inter-provincial migration has declined, while the rate of within province migration and mobility have increased.

On the other hand, the actual destinations selected (i.e. the geography) do vary widely from one period to another depending on regional economic performance and on which sectors and urban areas are growing or declining. Most domestic migrants move for the usual reasons: for job opportunities, schooling, family formation (or dissolution), housing, environmental amenities, or for life style or life cycle reasons such as retirement. Some move by choice; others involuntarily. Migration flows within any urban system also tend to be symmetrical, that is in-flows and out-flows are usually of similar magnitude. Net migration flows (in-migration less out-migration), thus, are typically small, but nonetheless significant. Most such net relocations involve movements upward in the urban size hierarchy and outward from the larger metropolitan areas to smaller communities in the periphery of those regions.

The contrast with immigration is obvious: the rate of foreign immigration has varied considerably over time, but the destinations have remained limited in number and become increasingly concentrated. For internal migrants the overall rate of movement has remained more-or-less constant over time, while the destinations have become more dispersed and more variable from one period to another. Immigration, unlike most types of domestic migrations, is a more-or-less Amanaged@ flow in the sense that it is subjected to political whims, frequent changes in policy directions, interest-group pressures, and unwritten administrative constraints. Moreover, the choices of immigrant destinations are typically not based on detailed and current knowledge of the changing economic fortunes of individual places in the urban system, but rather are based on the locations selected by earlier immigrants of similar ethnic or linguistic background from the same source region or country.

It is also important to re-iterate that cities are themselves dynamic demographic systems; that are continually being redefined and re-constituted as social entities, working and living environments and consumption spaces. Consider the Toronto example once again. During the single decade 1986-96, over 390,000 people migrated to the Toronto metropolitan area from elsewhere in the country, while 592,000 left ( a net migration loss of 202,000). In total, in-and-out movements involved almost one million people. Natural increase (the difference between births and deaths) added roughly 255,000 people over the same time period, while net foreign immigrants totaled 573,000. In other words, out of a 1996 metropolitan area population of 4.6 million, over 1.5 million are Anew@ residents. To further complicate any analysis, over 35% of continuing residents actually changed their place of residence within the metropolitan area during the decade. This is an example of the highly fluid local demographic situation into which overseas migrants are introduced. With such high mobility rates, the potential for the rapid social (and ethno-cultural) transformation of neighbourhoods, indeed of entire urban regions, is very high. Clearly, a careful monitoring of these trends, and the relationships that underlie them, are important elements in constructing viable and equitable social policies.

Most research on immigration makes little or no reference to domestic migration, and vice versa. But there is a legitimate question as to whether these two flows are related or not, and if so, how? Indeed, in the U.S. there is considerable interest recently in the apparent divergence of immigrant destinations from those of internal migrants, and the consequences of those trends (Frey 1995; Walker, Ellis and Barff 1992; Frey and Liaw 1998). Suggestions that the divergence of these two flows is leading to a ABalkanization@ of U.S. metropolitan areas has generated an intense debate (Ellis and Wright 1998). There is also concern over the increasing degree of immigrant concentration, and with the possible emergence of an urban Aunderclass@ of very poor immigrants (Clark 1998).

In the U.S. the five largest destinations or gateways for foreign immigrants - New York, Los Angeles, San Francisco, Chicago and Miami - received over 58% of all immigrants; New York and Los Angeles alone received over 41%. At the same time, all five of these metropolises had substantial net losses in terms of domestic migration (Frey 1995; Ellis and Wright 1998). Those U.S. metropolitan areas that have attracted the largest proportions of domestic in-migrants over the 1980s decade, such as Atlanta, Dallas, Phoenix, Houston, Las Vegas, Orlando and Tampa-St. Petersburg, in contrast, have as yet attracted few overseas immigrants.

In Canada, as noted, international migration is even more telescoped on a few metropolitan centres. Over 82% of immigrants now go to just five metropolitan centres, and over 61% to Toronto and Vancouver alone. It would appear that domestic migrants in both countries find different places attractive in comparison to international migrants; and it would seem reasonable to conclude that this concentration is not transitional. Obviously, for overseas migrants chain migrations play a major role in destination choices; but are these a matter of choice?; are domestic migrants being pushed out as a result? Do immigrants select large metropolitan areas because they provide more opportunities and a larger pool of previous immigrants from similar cultures, and perhaps offer more accommodating social, commercial and institutional environments. Or do they do so because that is where personal connections and circumstances, and public policy constraints, channel them?

There are at least three possible ways to conceptualize the relationships between immigration and migration:

1) the two flows are independent. Clearly, the two flows are different in their origins, attributes, constraints and logistics. And, given that immigration constitutes a relatively small proportion of total population movements nationally, it is possible to argue that the two flows - in terms of both their aggregate rates and the specific patterns of their destinations - are essentially unrelated, in both time and space.

2) the two flows are complementary. The hypothesis here is that immigrants find the same places attractive as do domestic migrants, for essentially the same reasons. These reasons, typically, are the presence of job opportunities, higher incomes, housing, local amenities, friends and kinship networks. Thus, higher levels of net (domestic) migration should be positively correlated with levels of immigration. In fact, the relationship could be mutually self-reinforcing.

A complementary relationship between the two flows, however, would tend to accentuate differences in growth rates and living conditions between those places that are winners and those that are losers in the migration exchange.

3) the flows are related but in an inverse fashion. There are two variations to this line of argument. One is the push or Adisplacement@ effect, in which new immigrants indirectly push out domestic residents (thus creating domestic out-migrants). They might do so through the increased competition they represent for jobs, living space and social services, and in some instances because of an aversion on the part of existing residents to increasing social diversity. We have, of course, frequently heard the lament that Aimmigrants take away jobs@, and that they reduce incomes by undercutting wage levels, and in so doing encourage local residents to leave the recipient cities. This Alabour-market displacement@ thesis is the principal focus of interest in the debate on the migration-immigration linkage in the U.S. In that research the dominant group of domestic out-migrants from high-immigration metropolitan areas tend to be lower-skilled, working class families (Frey and Liaw 1998).

The other variation on this line of argument introduces a more positive or Afacilitating@ effect. In this case, the presence of an expanding immigrant population creates the potential for more Aequity@ out-migrants; that is, local residents who take advantage of increasing house prices through the demands of immigrants to sell (or rent) their property and move to other locations. These are usually less costly and often less socially-diverse places. In any case, it appears that some residents of metropolitan areas are only too happy to be given the opportunity to move out. In this case both groups benefit. In contrast, in slow-growth urban areas, such as those in the Atlantic region, Quebec (outside Montreal) and the Prairies, where domestic in-migration is low and foreign immigrants are few, there is much less opportunity for residents to sell and leave town should they wish to do so. Both processes can of course occur at the same time in the same places.

It is not possible, given available data sources, to formally test these hypotheses, but we can offer a number of generalizations based on scattered information and highly aggregated statistical results. At the national level, there is no correlation between the rate of immigration and the rate of internal migration over time. When all 140 urban areas in Canada are examined we find that there has been (at least until 1991) a modest, but significant and positive correlation coefficient (r = +0.342) between net migration and net international migration (measured as the % foreign born). This result would suggest that the two flows are complementary; that is both groups are attracted to the same places. Yet, during the early 1990s the correlation has been reversed and is now - at least for metropolitan areas - marginally negative (r = -0.40). This, in contrast, indicates a divergence in the two flows and thus the possibility of a displacement or facilitating effect.

At the same time, it is clear from correlation and regression analyses that recent immigrants to Canada are attracted to urban places that are larger (r = +0.728 with population size), have higher incomes (r = +0.513), service-based economies (r = +0.578), higher educational levels (r = +0.382) and higher average house prices (r = +0.671). They also tend to avoid places that have high unemployment (r = -0.307), low education levels (r = - 0.421), relatively high levels of social dependence and government transfer payments (r = -0.474), and predominantly French-speaking populations (r = - 0.398). These relationships are not unlike those for internal migrants, but with some interesting variations.

Closer examination of the migration-immigration experience for individual metropolitan areas suggests that there is not one model or type of relationship. Each urban area has a somewhat different migration regime, that is, a different combination of migration and immigration flows, the balances among which can and do shift over time. To illustrate the point we undertook to construct a typology of metropolitan areas combining relative rates of natural population increase (above or below the average for the urban system), net internal migration (positive or negative) and net immigration for the 1991-96 census period (Table 3). The result is an eight-group classification, in which only one group of three places (Vancouver, Kitchener, Calgary) has above average levels for all three components.

Among the three largest metropolitan areas the migration regimes are also very different, as are the factors required to explain those regimes. Toronto and Montreal both registered substantial net out-migration (domestic) flows while receiving large immigration flows. Vancouver, in contrast, has recorded positive in-flows in terms of both domestic and foreign migrants throughout most of the post-war period. Yet Toronto and Montreal also have very distinctive domestic migration regimes. Montreal, which might surprise some readers given its relatively slow growth rate, has had smaller net migration losses than Toronto for the last two decades, despite the very rapid growth of the latter. At the same time, Montreal also has much lower in- and out-migration rates overall, suggestive of the importance of the language barrier in domestic migration.

The size of the Toronto metropolitan area (1996 population 4.6 million), and its dominant position with respect to Canada=s urban system, its economy, and social and cultural institutions, suggest that it will play a crucial role in the process of population redistribution and in the accommodation of immigrants in particular. As such it warrants special attention. The Toronto situation is indeed unique in both regards, and its migration experience has changed dramatically over the last two decades.

As noted above the greater Toronto area (GTA) has been the major recipient of overseas migrants since WWII, and that proportion has been increasing. The GTA, with 16% of Canada=s population in 1996, has 37% of all immigrants and over 43% of recent immigrants (City of Toronto 1998). As everyone knows it is now the most ethnically diverse metropolis on the continent. Almost 40% of the total GTA population (or 1.8 million people), at the 1996 census, was foreign born, compared to 17.4% in the country as a whole. The GTA has also captured an increasing proportion of the country=s recent immigrants, most of whom are from non-traditional source countries in South and East Asia, and the Caribbean (Table 4). Over 100 different countries have each contributed at least 1,000 immigrants. The face of the city and its suburbs is no longer recognizable, even to frequent visitors from most other parts of the country.

In parallel with increasing immigration, the Toronto metropolitan area has witnessed a surprisingly sharp reversal of internal migration flows. Prior to 1986 Toronto had a relatively balanced migration exchange with other parts of its regional hinterland and with the rest of the country. Indeed, during the 1981-86 period the metropolitan area recorded a positive (net) migration balance of over 80,000. Recall that this was also a period of relatively low foreign immigration. From 1986 on, in contrast, the situation has reversed. The net migration balance became negative, and dramatically so, with a net loss of 115,000 persons for 1986-91 and over 87,000 for 1991-96. At the same time, foreign immigration accelerated to an average of over 60,000 per year.

Why did this shift happen? And would we be correct in concluding that immigration played a role in this reversal of fortunes in domestic migration for Toronto? Of course we cannot answer this question without access to micro-level data, preferably in a longitudinal format, but the aggregate data available indicate that the two flows are not statistically independent. As immigration levels increased in the late 1980s and early 1990s net internal migration became negative. There is no single source for this shift: changes in both in- and out-migration flows contributed to the reversal. In-migration rates for Toronto dropped, from 8.5 to 4.5 (as a % of base population), while out-migration rates increased from 5.9 to 6.8%. Since in- and out-migrations respond to somewhat different sets of factors, the reasons for this reversal are likely to be many, and far too complex to be examined in detail here.

Who are these people and where are they going? The answer to the first question is that we do not know; the answer to the second question is almost everywhere in the country, but especially to the far west (Alberta and B.C.), to nearby Oshawa and Hamilton, and to small towns and rural areas in the region around Toronto. These represent different migration flows: those to the west are largely job related, those to Oshawa and Hamilton are primarily overspill suburbanization, while those to small towns located immediately outside the census metropolitan area likely represent long-distance commuters as well as retirement and recreational migrations.

Whatever the reason, the implication is that these flows represent the combined outcomes of housing and labour market dynamics internal to the region, as well as the push (displacement) and pull (facilitating) effects of higher immigration levels. Higher house prices in metropolitan Toronto in the late 1980s, for example, tended to discourage domestic in-migration while also permitting more equity out-migrants. Higher unemployment levels in the early 1990s also discouraged potential in-migrants, while at the same time encouraging an increase in job-seeking out-migrants.

The migration-immigration relationship, however, is not that simple; public policy, differences in demographic structure, and the business cycle, also have had an impact. For example, the national recession that began in 1989-90 hit the Toronto regional economy, and its large manufacturing sector, particularly hard. Unemployment rates, formerly among the lowest in the country, reached and then exceeded the national average. Domestic migrants obviously responded to these cues, either by moving out of the metropolitan area, or more often, by not moving in. The flow of immigrants from overseas, in contrast, continued to rise, as permitted by federal government policy. Immigrants, with less explicit information on current economic conditions on which to base their destination choices, came to places in which their kin or countrymen had previously settled. What then appears as a strong inverse time-series correlation between foreign in-migrants and domestic out-migrants in the Toronto case may in fact be, at least in part, an accident of timing with respect to the effects of regional business cycles and public policy decisions.

This paper has attempted to demonstrate the increasingly significant role played by migration and immigration - in other words the processes of population redistribution - in shaping Canada=s population growth and in driving the rapid and uneven social transformation of Canadian urban areas. Both of these flows are highly volatile and unpredictable, but in reverse: for domestic migration it is the destinations that vary not the overall rate, for immigration it is the rate less than the geography of immigrant destinations that varies.

With declining fertility levels, immigration alone now accounts for over 50% of national growth, but the impacts of that flow are very unevenly distributed across the country. In some metropolitan areas immigration is responsible for all recent population growth; in others immigration has simply compounded the effects of high levels of internal migration; in still others, immigrants seem to have replaced (or displaced) domestic migrants.

The composition of those flows has also shifted dramatically toward countries that have not traditionally been sources of immigrants and that provide more culturally distinctive populations. Such flows, moreover, are also highly telescoped on a few large metropolitan areas, most notably Toronto and Vancouver. The result, as noted, has been a rapid transformation of the social landscapes and lifestyles of the population of the entire urban region, in Toronto and a few other high-immigration metropolitan areas, and the increasing differentiation of these places from the rest of the country. What effects this divergence in social characteristics, between and within Canadian metropolitan areas, will have on national and regional markets, political agendas, and local public policies, has yet to be determined. Diversity obviously creates tensions and conflicts, it also offers numerous opportunities for social progress.

The effort to define the relationship between migration flows and immigration has only been partially successful, which is perhaps not surprising given the paucity of suitable micro-level data. The analysis to date, however, has at least been able to confirm that no single model or explanation will suffice. There are many different migration regimes that characterize Canadian metropolitan areas. The Toronto experience is only one, but perhaps the most unique, of those regimes. There is some empirical evidence, albeit mostly in aggregate form, that all three hypothesized relationships between domestic migration and immigration apply in the Canadian case, for different metropolitan areas and at different times. That is, the two flows are, in some periods and at different spatial scales, essentially unrelated to or independent of each other, while at other scales and in other places, there is evidence of complementary linkages, as well as both displacement and facilitating effects.

What can we say about the mobility rates and patterns of immigrants after their initial settlement? The answer, again in the absence of longitudinal data, is very little. What we can do as an approximation is to trace the changing locations of the foreign-born population at different census points according to their initial period of immigration (e.g. before 1960, 1961-1981 and after 1981) . Here the empirical results are mixed. There has indeed been, on balance, some dispersion of earlier immigrant population across the country and away from the original gateway centres, but not very much - at least for the first generation of recent migrants. After that generation we lose them - they become domestic migrants. Indeed, a significant proportion of the dispersion that we do observe within the Canadian urban system is from the Montreal and Quebec City metropolitan areas to urban areas in Ontario and in the provinces of Alberta and B.C. Within the two major receiving provinces, Ontario and B.C., a modest degree of within-province geographical dispersion can be detected, largely to smaller cities not far removed from the larger metropolitan areas. Whether this is due to retirement, or to the search for less expensive housing, or other reasons, we do not know.

On the other hand, as the immigrant population becomes more Adistinctive,@ in terms of ethno-cultural background, language and race, it is reasonable to hypothesize that the spatial dispersion process may well be even slower in the future. Newer immigrants may demonstrate a reluctance, and perhaps understandably so, to move into culturally uncharted - and perhaps much less friendly - territory. Kinship ties, and friendly environments, are crucial variables in the location decisions of immigrants. Clearly, further research is needed on the subsequent migration decisions of immigrants if we are to sort out these relationships and identify the real costs of social adjustment policies.

It has been argued that policy initiatives designed to foster social sustainability and social cohesion must be cognizant of and responsive to social change and the processes of population growth and redistribution. Immigration is one obvious manifestation of globalization, and a major source of social change - as well as a factor in increasing social and ethnic diversity - in many developed countries. But immigration does not take place in a social vacuum. It must be set in the larger context provided by an understanding of population dynamics, and specifically of demographic change (e.g. an aging society) and internal migration within the host country and particularly within the receiving metropolitan areas.

At the same time, Toronto provides an excellent laboratory in which to study the dynamic linkages between domestic migration flows and immigration from abroad. In that setting we can observe the full range and complexity of possible locality-specific responses between a large regional demographic system (a rapidly growing the metropolitan area) and a process of increasing social diversification through one form of globalization (immigration). The reversal of Toronto's domestic migration exchange over the last decade, in which net internal migration is now markedly negative, coincident with a sharp increase in immigration levels, suggests that the two processes are related. If they are not related directly then they certainly are indirectly, through the market and through the competition for work, living space and social services.

Based on this example, it is clearly not possible to study either migration or immigration, as processes of social change, or in terms of their specific social consequences, in isolation. Social sustainability, and the quality of urban life generally, in the greater Toronto region and likely everywhere else, very much depends on how well society responds to the challenges posed by this combination of demographic and migration transitions, and to the resulting increase in the social and ethno-cultural diversity of the resident population.

Beaujot, R. 1991. Population Change in Canada: The Challenges of Policy Adaptation. Toronto: Oxford University Press.

Beaujot, R. 1998. AMigration Trends and Policies in North America: The Case of Canada@, Paper prepared for an OECD Conference in Mexico. Dept of Sociology, University of Western Ontario, London, Ontario, mimeo

Bourne, L.S. and Ley, D. Eds. 1993. The Changing Social Geography of Canadian Cities. Montreal: McGill-Queens University Press.

Bourne, L.S. and Olvet, A. 1995. New Urban and Regional Geographies in Canada: 1986-91 and Beyond. MR# 33, Centre for Urban and Community Studies, University of Toronto.

Bourne, L.S. and Flowers, M. 1999. Changing Urban Places: Mobility, Migration and Immigration in Canada. Research Paper 196, Centre for Urban and Community Studies, University of Toronto.

Canadian Journal of Regional Science, Special Issue. 1997. AMetropolis@, Vol. XX: 1-2 , Spring-Summer.

Citizenship and Immigration Canada. 1998. Citizenship and Immigration Statistics 1995. Ottawa: Minister of Public Works and Government Services. Cat# MP22-1/1995.

City of Toronto, 1998. Profile Toronto: Immigrants in Toronto. Bulletin #2, June 1998. Urban Planning and Development Services.

Clark, W.A.V. 1986. Human Migration. Beverly Hills: Sage Publications.

Clark, W.A.V. 1998. AMass Migration and Local Outcomes: Is International Migration to the U.S. Creating a New Urban Underclass?@, Urban Studies, 35, 3, 371-383.

DeVoretz, D. Ed. 1995. Diminishing Returns: The Economics of Canada=s Recent Immigration Policies. Toronto: C.D. Howe Institute.

Economic Council of Canada. 1991. New Faces in the Crowd: Economic and Social Impacts of Immigration. Ottawa: Economic Council study No. 22-171.

Ellis, M. and Wright, R. 1998. AThe Balkanization Metaphor in the Analysis of U.S. Immigration@, Annals, Association of American Geographers, 88, 4 : 686-698.

Foot, David. 1994. ACanada=s Unemployment-Immigration Linkage: Demographic, Economic and Political Linkages@, Canadian Journal of Sociology, 19, 4, 513-524.

Foot, D. and Stoffman, D. 1996. Boom, Bust and Echo: How to profit from the Coming Demographic Shift. Toronto: Ross.

Frey, W. 1995. AThe New Geography of Population Shifts: Trends Towards Balkanization@, in F. Reynolds, ed. The State of the Union, Volume 2. New York: Russell Sage Foundation, pp. 271-336.

Frey, W. And Liaw, K.-L. 1998. AImmigrant Concentration and Domestic Migrant Dispersal@, Professional Geographer, 50, 2, 215-232.

Frisken, F. Ed. 1994. The Changing Canadian Metropolis: A Public Policy Perspective. Berkeley: University of California, Institute of Government Studies Press.

Long, L. 1988. Migration and Residential Mobility in the United States. New York: Russell Sage Foundation.

Moore, E. and Rosenberg, M.. 1993. AMigration, Mobility and Population Redistribution@, in L.S. Bourne and D. Ley, eds. The Changing Social Geography of Canadian Cities. Montreal: McGill-Queen=s University Press, pp.121-137.

Murdie, R. 1998. "The Welfare State, Economic Restructuring and Immigrant Flows: Impacts on Socio-Spatial Segregation in Greater Toronto", in S. Musterd and W. Ostendorf, eds. Urban Segregation and the Welfare State: Inequality and Exclusion in Western Cities. London: Routledge, pp. 64-93.

Newbold, K.B. and Liaw, K.-L. 1990. ACharacterization of Primary, Return and Onward Interprovincial Migration in Canada: Overall and Age-specific Patterns@, Canadian Journal of Regional Science, 13, 1: 17-34.

Pandit, K. 1997. ACohort and Period Effects in U.S. Migration: How Demography and Economic Cycles Influence the Migration Schedule@, Annals, Association of American Geographers, 87, 3 : 439-450.

Shaw, R.P. 1985. Internal Migration in Canada. Determinants over Three Decades. Toronto: NC Press.

Simmons, J.W. 1977,1978. Migration in the Canadian Urban System: Part I, Spatial Patterns; Part II, Simple Relationships. Research Papers 85 and 92, Centre for Urban and Community Studies, University of Toronto.

Statistics Canada. 1992. Immigration and Citizenship. Ottawa: Statistics Canada Cat. No.93-316.

Statistics Canada. 1997. Annual Demographic Statistics 1996. Ottawa: Statistics Canada Cat. No. 91-213.

Walker, R., Ellis, M., and Barff, R. 1992. ALinked Migration Systems: Immigration and Internal Migration Flows in the U.S.@ Economic Geography, 68, 3, 234-248.

Virtual Library		Updated February 20, 2004