The Housing Situation and Needs of Recent Immigrants in the Montréal,Toronto, and Vancouver CMAs:...

Research reportDistinct Housing Needs Series

The Housing Situation and Needs of RecentImmigrants in the Montréal,Toronto, andVancouver CMAs: An Overview

CMHC—Home to CanadiansCanada Mortgage and Housing Corporation (CMHC) hasbeen Canada’s national housing agency for more than 60 years.

Together with other housing stakeholders, we help ensurethat Canada maintains one of the best housing systems in theworld.We are committed to helping Canadians access a widechoice of quality, affordable homes, while making vibrant,healthy communities and cities a reality across the country.

For more information, visit our website at www.cmhc.ca

You can also reach us by phone at 1-800-668-2642 or by fax at 1-800-245-9274.

Outside Canada call 613-748-2003 or fax to 613-748-2016.

Canada Mortgage and Housing Corporation supportsthe Government of Canada policy on access toinformation for people with disabilities. If you wish toobtain this publication in alternative formats,call 1-800-668-2642.

The Housing Situation and Needs ofRecent Immigrants in the Montréal,Toronto,

and Vancouver CMAs: An Overview

Canada Mortgage and Housing CorporationSeptember 2006

Prepared by:

Daniel Hiebert,Annick Germain, Robert Murdie,Valerie Preston, Jean Renaud, Damaris Rose, Elvin Wyly,Virginie Ferreira, Pablo Mendez, and Ann Marie Murnaghan

For correspondence:

Daniel HiebertDepartment of Geography

University of British Columbia1984 West Mall

Vancouver, British ColumbiaV6T 1Z2

CMHC offers a wide range of housing-related information, For details, call 1-800-668-2642 or visit our website at www.cmhc.ca

Cette publication est aussi disponible en français sous le titre : Aperçu de la situation et des besoins en matière de logement desimmigrants récents dans les RMR de Montréal, Toronto et Vancouver 65320.

This research project was (partially) funded by Canada Mortgage and Housing Corporation (CMHC). The contents, views and editorialquality of this report are the responsibility of the author(s) and CMHC accepts no responsibility for them or any consequences arisingfrom the reader's use of the information, materials and techniques described herein..

Library and Archives Canada Cataloguing in Publication

Hiebert, DanielThe housing situation and needs of recent immigrants in the Montréal, Toronto, and Vancouver CMAs : an overview / prepared byDaniel Hiebert ... [et al.].

(Distinct housing needs series)(Research report)Issued also in French under title: Aperçu de la situation et des besoins en matière de logement des immigrants récents dans les RMR de Montréal, Toronto et Vancouver.Includes bibliographical references.ISBN 978-0-662-45314-7Cat. no.: NH18-28/1-2007E

1. Immigrants--Housing--Québec (Province)--Montréal Metropolitan Area. 2. Immigrants--Housing--Ontario--Toronto Region. 3. Immigrants-Housing--British Columbia--Vancouver Metropolitan Area. 4. Immigrants-Housing--Canada--Statistics. I. Canada Mortgage and Housing Corporation II. Title. III. Series. IV. Series: Research report (Canada Mortgage and Housing Corporation)

HD7288.72.C3H53 2007 363.5'969120971 C2007-980045-9

© 2006 Canada Mortgage and Housing Corporation.

All rights reserved. No portion of this book may be reproduced, stored in a retrieval system or transmitted in any form or by anymeans, mechanical, electronic, photocopying, recording or otherwise without the prior written permission of Canada Mortgage andHousing Corporation. Without limiting the generality of the foregoing no portion of this book may be translated from English intoany other language without the prior written permission of Canada Mortgage and Housing Corporation.

Printed in CanadaProduced by CMHC

The Housing Situation and Needs of Recent Immigrants in the Montréal, Toronto, and Vancouver CMAs: an Overview is Volume 2 of a five-volume series for the project

The other volumes are:

Volume 1: Immigrants and Housing: A Review of Canadian Literature From 1990 to 2005 by Robert Murdie, Valerie Preston,Sutama Ghosh and Magali Chevalier

Volume 3: The Housing Situation and Needs of Recent Immigrants in the Montréal Metropolitan Area/La situation résidentielle desimmigrants récents dans la Région métropolitaine de Montréal by Damaris Rose, Annick Germain, and Virginie Ferreira (2006)

Volume 4: The Housing Situation and Needs of Recent Immigrants in the Toronto CMA by Valerie Preston, Robert Murdie, AnnMarie Murnaghan and Daniel Hiebert (2006)

Volume 5: The Housing Situation and Needs of Recent Immigrants in the Vancouver CMA by Daniel Hiebert, Pablo Mendez, andElvin Wyly (2006).

The authors gratefully acknowledge the financial support of Canada Mortgage and Housing Corporation and The Housingand Homelessness Branch (formerly National Secretariat on Homelessness) of Human Resources Social Development Canada,for this research. York University also generously provided funds for part of the project. We also thank the CanadianMetropolis research centres (RIIM, CERIS, and IM) for logistical support, and Statistics Canada for allowing us permission to use the Metropolis Core Tables for this research project. We also extend thanks to the personnel at the British ColumbiaInter-university Research Data Centre for their assistance in accessing data from the Longitudinal Survey of Immigrants to Canada.

Canada Mortgage and Housing Corporation iii

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

The Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3

Basic Household and Housing Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5

Vulnerable Households . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9

The Initial Experience of Immigrants in the Housing Market:

Insights From the LSIC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19

References Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29

Appendix 1:The Residential Mobility of Newcomers to Canada:

The First Months . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31

Appendix 2: Immigrants and Housing in Australia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41

Canada Mortgage and Housing Corporation v

table of contents

Table 1 Tenure, Income and Monthly Housing Costs by Immigration Period, 2001,Toronto, Montréal, and Vancouver CMAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21

Table 2 Percent of Total Household Income Devoted to Monthly Housing Costs by Ethnic Origin and Visible Minority Subgroup, 2001,Toronto, Montréal, and Vancouver CMAs . . . . . . . . . . . . . . . . . . .22

Table 3 Housing Price and Rental Changes, between 1996 and 2001, Montréal,Toronto,and Vancouver CMAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

Table 4 Housing Price Changes and Ownership Rates between 1996 and 2001 by Immigration Period, Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

Table 5 Owners' Housing Spending and Income by Immigration Period, 2001, Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24

Table 6 Tenants' Housing Spending and Income by Immigration Period, 2001, Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24

Table 7 Percent of Owners and Tenants Paying at Least 30% of Income on Housing Payments by Immigration Period, 1996 and 2001, Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . .25

Table 8 Percent of Owners and Tenants Paying at least 50% of Income on Housing Payments by Immigration Period, 2001, Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . . . . . . . . . . .25

Table 9 Percent of Owners and Tenants Paying at Least 30% and 50% of Income on Housing Payments by Visible Minority and European Origin Households, 2001, Montréal,Toronto, and Vancouver CMAs . . . . . .26

Table 10 Homeownership by Admission Class, 2001, Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . . . . . . .26

Table 11 Household Structure by CMA, 2001, Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . . . . . . . . . . . .27

Table 12 Immigrant Class by Crowding and CMA, 2001, Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . . . .27

Table 13 Crowding by Ethno-cultural Group and CMA, 2001, Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . .28

Table 14 Cost of Housing relative to Income for Non-homeowners by CMA, 2001,Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28

Table 15 Non-homeowners Paying at Least 30% of Income for Housing by Ethno-cultural Group and CMA, 2001, Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . . . . . . . . . . .29

Table 16 Construction of the Housing Stress Index (for Non-homeowners Only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29

Table 17 Non-ownership Households by Degree of Financial Stress and CMA, 2001,Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29

Table 18 Non-ownership Households Facing Housing Stress (Moderate to Extreme) by Ethno-cultural Group and CMA, 2001, Montréal,Toronto, and Vancouver CMAs . . . . . . . . . . . . . . . . . . . . . . .30

list of tables

Access to adequate, suitable andaffordable housing is an essential step inimmigrant integration. Immigrants firstseek a place to live and then look forlanguage and job training, education fortheir children, and employment (Lapointe1996, Murdie et al. 2006). Housing isalso an important indicator of quality oflife, affecting health, social interaction,community participation, economicactivities, and general well-being (Engelandand Lewis 2005). This overview providesa synopsis of the findings of a largecomparative study of immigrants in thehousing markets of Canada's largestmetropolitan centres, Montréal, Toronto,and Vancouver. It is the final installmentof five separate reports:

� Immigrants and Housing: A Review ofCanadian Literature From 1990 to2005 by Robert Murdie, Valerie Preston,Magali Chevalier, and Sutama Ghosh (2006).

� The Housing Situation and Needs ofRecent Immigrants in the MontréalMetropolitan Area/La situationrésidentielle des immigrants récents dansla Région métropolitaine de Montréalby Damaris Rose, Annick Germain,and Virginie Ferreira (2006)

� The Housing Situation and Needs ofRecent Immigrants in the TorontoCMA, by Valerie Preston, RobertMurdie, Ann Marie Murnaghan, andDaniel Hiebert (2006)

� The Housing Situation and Needs ofRecent Immigrants in the VancouverCMA by Daniel Hiebert, PabloMendez, and Elvin Wyly (2006)

� The Housing Situation and Needs ofRecent Immigrants in the Montréal,Toronto and Vancouver CMAs : AnOverview by Daniel Hiebert, AnnickGermain, Robert Murdie, ValeriePreston, Jean Renaud, Damaris Rose,Elvin Wyly, Virginie Ferreira, PabloMendez, and Ann Marie Murnaghan(2006)

The first volume in our series summarizesand synthesizes the key Canadian literatureon the relationship between immigrationand housing. Separate reports are devotedto a detailed analysis of the housingsituation of immigrants in each of Canada'smajor Census Metropolitan Areas (CMAs).We have considered each of these centresseparately because of importantmetropolitan variations in immigrationand housing markets in Canada. Montréal,Toronto, and Vancouver each have a

particular history of immigration anddistinct geographical patterns of immigrantsettlement. In essence, immigrants entera specific and complex housing submarketwhen they settle in a particular place(Canada Mortgage and HousingCorporation 2004). The specificity ofthe housing market in each metropolitanarea interacts with the distinct patternsof immigration, so that in each place,immigrants confront a locally-specific setof housing opportunities and challenges.In this report, we review the findings ofthe three separate CMA studies, hopingto provide both a sense of the largerpicture of immigrant settlement and thehousing markets of Canadian metropolitanareas, and also reveal the importantdifferences between Montréal, Torontoand Vancouver.

Our individual reports have drawn uponnew and important information aboutthe housing situation of immigrants, andhave been organized into four sections.Each has begun with a review of thehistory of immigration in the particularmetropolitan area in question, and recenttrends in the housing market. Thesediscussions have set the context forunderstanding the social and housingcircumstances encountered byimmigrants when they first arrived in Canada.

Canada Mortgage and Housing Corporation 1

Chapter 1

Introduction

The next section of each CMA reportreviews the housing conditions ofimmigrants currently living in themetropolitan area. We have emphasizedthe effects of immigrant status, period ofarrival, and ethnic and visible minoritystatus on the housing situation ofimmigrants. Drawing on special tabulationsfrom the 2001 census (made available byStatistics Canada to researchers affiliatedwith the Metropolis Project), and wherepossible invoking comparisons with 1996census data, we have also compared thecircumstances of immigrants in thehousing market with those of householdsthat are comprised of Canadian-bornindividuals. We have further disaggregatedthe immigrant population, by exploringdifferences in the housing situations ofparticular ethno-cultural groups.

We have documented the success of manyimmigrants in attaining homeownershipand the characteristics of immigranthouseholds that are in the rental market.Our findings highlight the situations ofthose who are experiencing affordabilityproblems. In this, the third section ofthe individual reports, we have followedconventions developed by Canada Mortgageand Housing Corporation (CanadaMortgage and Housing Corporation 2004),and have examined the relationshipbetween housing costs and householdincomes of immigrant households spendingat least 30 per cent of their total pre-tax

income on housing, as well as a smallergroup of immigrant households spendingat least 50 per cent of total income onhousing. Again, we have disaggregatedthe metropolitan populations in thesecategories by immigrant status, period ofarrival, visible minority subgroups andethnic origins.

In the fourth section of the reports, we haveturned to new data from the first wave ofthe Longitudinal Survey of Immigrants toCanada (LSIC), conducted by StatisticsCanada and Citizenship and ImmigrationCanada on a representative sample ofimmigrants who landed in Canada betweenOctober 2000 and September 2001. Thesedata have enabled us to explore the waysthat very recently-arrived immigrants findhousing and the extent to which their initialhousing situations are adequate, suitableand affordable. Significantly, LSIC includesinformation on the admission class ofimmigrants—information not collected forthe census or in other major Canadiansocial surveys. Previous research, basedon single case studies or surveys in a singlecity (Renaud 2003; Rose and Ray 2001;Murdie 2005; Bezanson 2003), has suggestedthat refugees and refugee claimants havemore difficulty than other classes ofimmigrants in the housing market. LSICallows us to explore the relationshipbetween immigration category at landingand early housing outcomes in the threemetropolitan areas, and to relate thesefindings to the local housing market.

We have structured this synopsis of thethree metropolitan reports in the sameway, starting with a statement about thechanging trajectories of immigration andthe housing markets of Montréal, Torontoand Vancouver. We then summarize thehousing characteristics of the immigrantpopulation, compared with the Canadianborn, and then focus on households thatare in vulnerable circumstances. With thebroad sweep of immigration and housingin mind, we turn to LSIC, which helps usunderstand how these long-term patternsare established in the first few months ofthe settlement experience. We concludethis report by summarizing our thoughtson these findings in relation to Canadianimmigration and housing policies. Finally,we have provided two appendices, onesurveying the major findings generated bythe Longitudinal Survey of Immigrants toAustralia, and the other examining thelongitudinal side of the LSIC in muchgreater detail, through an event analysisof immigrants' acquisition of housingover their first half-year in Canada.

Research ReportThe Housing Situation and Needs of Recent Immigrants in the Montréal, Toronto, and Vancouver CMAs: An Overview

Canada Mortgage and Housing Corporation2

Throughout the 20th century we canthink of Montréal, Toronto andVancouver as Canada's major connectionpoints to the global economy. Each hasbeen connected differently, though.Historically, Montréal has had the strongestlinks to Europe, Toronto to the UnitedStates, and Vancouver to Asia. To a largedegree, the fortunes of these metropolitancentres mirrored those of their associatedregions, and the intensity of Canadian tradewith them. When Europe dominated theworld, Montréal led the Canadian urbanhierarchy. With the ascendance of theUnited States, Toronto overtook Montréalin both demographic and economicterms. The more recent rise of the Asianeconomies has contributed to economicgrowth in Vancouver.

The grand story of immigration is generallyaligned with these broad patterns, especiallyafter the redefinition of policy that occurredin the 1960s. Of the three centres, Montréalretains the most direct connection withEuropean immigration, drawingsignificantly from France and its formercolonies, such as Haiti and the countriesof the Maghreb. Montréal also receivesthe largest share of refugees of the threeurban centres, for complex reasons (seethe individual report on Montréal for

further discussion). Conversely, Vancouverdraws the majority of its immigrantsfrom eastern and southern Asia. Despiteseveral repressive regimes in those regions,they are not a major source of refugeesdestined for Canada. But, as a region ofrapid economic development, Asia generatesa large number of skilled worker andentrepreneurial/ investor immigrants.Vancouver's immigrant profile, broadly,reflects the state of affairs in Asia andtherefore includes a high ratio of skilledworkers and, particularly, individualsadmitted in the entrepreneur and investorprograms. It is important to rememberthat metropolitan Toronto has the largestshare of first-generation immigrants ofany OECD city and, by far, the largestabsolute number in Canada. All classesof immigrants settle in Toronto andToronto, in essence, attracts the world(hence the title of the recent book, TheWorld in a City; Anicef and Lanphier2003). More than the other twometropolitan areas, Toronto's immigrantprofile resembles that of the country as awhole. Asians are therefore the singlelargest group arriving in metropolitanToronto, but they are joined by immigrantsfrom virtually all corners of the earth.

These specific migration circuits translateinto distinct cultural, and economic, profilesof immigrants in the three centres. Montréalreceives the least affluent immigrants,generally speaking. While we have notprovided specific tables on this in ourvarious reports, the difference in savingsreported by immigrants (in LSIC) in thethree centres is large, with immigrants inMontréal the least “protected” in thismanner. We can simply turn this pointaround for Vancouver, which receives adisproportionate share of affluentimmigrants. This sub-group retains thelargest levels of savings (at six months) ofimmigrants in any of the three metropolitanareas; as we will see below, this hasimportant consequences for the participationof immigrants in the housing market ofVancouver. Finally, Toronto receives thelargest and most complex immigrantflow, with a particularly variegated set ofeconomic outcomes.

The three metropolitan areas also havedistinct housing markets. Nowhere isthis clearer than in the average price of ahouse, which was approximately $374,000in Vancouver, $312,000 in Toronto and$195,000 in Montréal (2004 figures). Tosome degree, the more difficult economicsituation of immigrants in Montréal is


Chapter 2

The Context

compensated for by lower housing prices,though it is important to note that thevacancy rate for rental units has fallendramatically in Montréal in recent years. Infact, the vacancy rate is low in all threemetropolitan areas, leading to high rentsin Toronto and Vancouver, and risingrents in Montréal.

The low vacancy rates across the centres aregenerally not mitigated by the constructionof new rental housing. In all cases, themarket has turned to the production ofowner-occupied housing, with astunning increase in the construction ofcondominiums in the three metropolitanareas. Immigrants face a critical challenge,that of finding their “place” in cities withstatic rental stocks and a rapidly risingdemand due to large-scale immigration.And they must do so with incomes thatare generally well below average.

There are, of course, many nuances to thestory that basic census data only partiallyaddress, such as the prevalence of distinctdwelling types in the three centres (e.g., thewalk-up apartments in Montréal, high-riseapartments in Toronto, and basementsuites throughout the neighbourhoods ofVancouver). These subtleties have a largeimpact on the actual settlement patternsof immigrants in the three centres and,moreover, the quality of the dwellings theyinhabit. We can generalize about oneaspect of settlement, though. In all threemetropolitan areas, the scale of immigrantreception is far too rapid to beaccommodated in a single neighbourhood,or even a single region of the city (suchas traditional inner-city receptionneighbourhoods). Immigrant admissionsare also far too diverse in terms of source

regions, immigration class, and socio-economic background, for immigrantsto settle in a single type of neighbourhood.The social landscape of immigrantsettlement is intricate and includes areasthroughout most of the metropolitan regionsof the three centres, and across the fullsocio-economic spectrum. In all threecentres, immigrant settlement has beendispersing into inner- and middle-rangesuburbs, at least, and in some cases intodistant suburbs. The latter developmentis more prominent in Greater Vancouverand the Greater Toronto Area, than theMontréal region. In any case, the spatialcomplexity of immigrant settlement isrelated to an equally complex encounterwith the housing markets of the threemetropolitan areas.

Research ReportThe Housing Situation and Needs of Recent Immigrants in the Montréal, Toronto, and Vancouver CMAs: an Overview


Throughout Canada, immigranthouseholds tend to be larger than thosewith Canadian-born primary maintainers(generally, the profile of second-generationimmigrant households is very similar tothat of households where both parents wereborn in Canada). The gap in averagehousehold size between immigrants andthe Canadian born is fairly constant for thethree centres, between 0.3 and 0.5 persons.But there is also a close correspondencebetween the cost of housing and the sizeof households. That is, households arelarger in the more expensive centres,Vancouver (3.0 persons on average) andToronto (3.1 persons), than in Montréal(2.8 persons).

Immigrants are also more likely than theCanadian-born to reside in “traditional”families with two parents and children,which helps explain their larger averagehousehold size. Again, there is variationacross the metropolitan areas on thismeasure. Montréal, for example, has alarger proportion of non-familyimmigrant households. The proportionof multiple-family households is muchhigher in Toronto and Vancouver, wherepeople are likely adjusting to the cost ofhousing by assembling larger householdunits to pool their incomes so they can

afford housing. This also suits the culturalneeds of some immigrant householdswho prefer to live as extended families,at least in the initial stage of settlement.

Several of these patterns are intensifiedwhen we differentiate between homeownersand tenants. Immigrants (and theCanadian-born) who own homes are themost likely to reside in “traditional”, i.e.dual-parent-plus-children households, orin multiple-family households. Wesuspect that the latter statistic is actuallyan indication that the multiple-familystrategy is effective, enabling householdsto purchase a dwelling. On the other hand,non-family and lone-parent households aremuch more likely to rent their dwellings.This is, of course, related to income level,as smaller households, particularly thosewith only one adult in the workforce, areless likely to have incomes high enoughfor homeownership.

Disaggregating cultural groups reveals anumber of important nuances. In all threeCMAs, Visible Minority groups arecharacterized by larger households, more dual-parent families, and moremultiple-family households. In many casesthe differences between Visible Minorityand European-origin groups are considerable.

Patterns of household formation are, to adegree, group specific. Of course, the coarsecategories of “Visible Minority” and“White”, or “European-origin”, hidealmost as much as they reveal, as thereare pronounced variations within them.The household dynamics of the Black vs.South Asian-origin populations are a goodcase in point, with the former group farmore likely than average to reside insingle-parent households, and the latterdisproportionately likely to reside inmultiple-family households. Similarly,within the “White” category, the householdstructures of the Italian- and UK-originpopulations are very different (the latterare far less likely to reside in dual-parent-with-children households). Significantly,in broad terms, group-by-group householdstructures are quite similar across the CMAs.For example, South-Asian-origin householdsin Montréal, Toronto and Vancouver aremore similar than South-Asian- and LatinAmerican-origin households in any givenCMA. This point suggests that adjustmentsto immigration intake, particularlychanging the composition of immigrantsfrom different source regions, will havespecific impacts on the housing marketsof Canadian cities.


Chapter 3

basic householdand housing patterns

The group-by-group household profilesbecome even more intricate when wedifferentiate between homeowners andtenants. To simplify this synopsis we willhighlight just two prominent findings.First, the South-Asian-origin group,regardless of the CMA in question, ismuch more likely than average to employa multiple-family household strategy toachieve homeownership. For example,23 per cent of all South-Asian-originhouseholds that owned a home inVancouver were in this category-comparedwith 3.3 per cent of the total population.Secondly, among tenants, European-origingroups, by far, are more likely to includenon-family households. This is likely relatedin part to the fact that the proportion ofrecent immigrants is low for theEuropean-origin population, and theproportion of older individuals is higher.

Table 1 provides a summary of datapresented on the relationship betweenincome and housing cost for all ownersand tenants in the three CMAs (includingimmigrants and the Canadian-born).First, note the strikingly different levelsof homeownership between centres,especially for recent immigrants (around11 per cent in Montréal compared with32 per cent in Toronto and 41 per centin Vancouver). Immigrants in Vancouverand Toronto have invested quickly, andheavily, in housing. In retrospect, wecan see that this has (so far) proven to bea highly successful gamble, as real estateprices in both cities have risen dramaticallysince 2001. But why have immigrants inMontréal not adopted the same strategy?In part, the differential in ownershiprates reflects both the historically lowerhomeownership rates in the City ofMontréal and the lower incomes ofimmigrants in Montréal relative to Torontoand Vancouver. This leads to a second

major point apparent in the data: thepercentage of income dedicated to housing,for both owners and tenants, is fairlyconsistent across the CMAs. For example,homeowners in Montréal, on average, payjust under 14 per cent of their income onhousing, which is only one per cent belowthe corresponding figure for Toronto andless than three per cent below that forVancouver. There is something close toan equalization effect operating in thehousing markets of the three centres,meaning that housing prices and rents reflectincome levels fairly closely (or, perhaps,we could turn the causality around andsay that income levels reflect the cost ofhousing). Residents, particularlyimmigrants, spend a little less of theirincome on housing in Montréal and alittle more in Vancouver, but these arerelatively modest differences. In Vancouverand Toronto, the high rents charged bylandlords, together with the prospect ofrising real estate prices, may have “pulled”immigrants toward homeownership. Thesefactors have been more muted in Montréal.

A third major finding of this study is thatimmigrants in Vancouver and Torontohave higher homeownership rates thanwe would predict given their levels ofincome. The situation in Vancouver isparticularly instructive. There, recentimmigrant homeowners dedicate justunder 30 per cent of their income tohousing, on average. This ratio of sheltercost to income is often used by banks asa maximum allowable figure for obtaininga mortgage, and yet a very large numberof immigrant households exceed it. Wedo not have scope in this synopsis to discussthe causes of this surprising turn of events(which is addressed in the Vancouverreport), but simply make the point thatthere is an “immigrant effect” in thehousing markets of Vancouver and

Toronto, meaning that immigrants are ableto purchase housing more often thanwould be expected.

Fourthly, we note the large discrepancyin income between households that havebeen able to purchase housing and thosewho are in the rental market. Moreover,homeowners enjoy a considerableeconomic advantage, in that on averagethey pay a lower proportion of theirincome on housing, than tenants—allthe while building equity and thereforepersonal or family wealth.

There is considerable variation inhomeownership rates between differentcultural groups. At the most general scale,comparing Visible Minority groups withthose of European origin, we see a ratherstriking difference across the threemetropolitan regions. In Montréal, theincidence of homeownership is almosttwice as high for the latter group (60 vs.32 per cent). In Toronto, the gap inownership rates is considerably less, butstill important (73 vs. 55 per cent). InVancouver, however, there is essentiallyno gap at all (67 vs. 66 per cent), despitethe fact that Visible Minority householdsreceive much lower incomes, on average.Or, another way to think of these figuresis that the homeownership rate forimmigrants from European backgroundsis fairly consistent across the CMAs, butthe rate for Visible Minorities is highlyvariable. We can obtain some understandingof this pattern by examining the experiencesof specific groups in more detail. Italianshave achieved the highest rate ofhomeownership in all three cities. InToronto, those who claimed Chinese andUK origin rank respectively second andthird on homeownership. In Vancouver,the corresponding groups are those ofChinese and South-Asian origin, whilein Montréal they are of UK- and Polish-



origin. There are even greater similaritiesat the opposite end of the spectrum. Thegroups with the lowest ownership ratesare: Toronto (Black, Latin American,Arab/West Asian); Montréal (LatinAmerican, Filipino, Arab/West Asian); andVancouver (Latin American, Arab/WestAsian, Black). In other words, the situationof ethno-cultural communities in the threeCMAs is not that different, but the relativeimportance of those groups in thepopulation composition of the CMAsvaries considerably (e.g., all three groupswith the lowest ownership rates inVancouver are small, while they are muchlarger in Montréal and Toronto). Thisfinding calls for greater scholarly attentionto the apparently similar processes ofmarginalization in all three CMAs.

Another facet of the issue of marginalizationis explored in Table 2. In all three CMAsVisible Minority homeowners dedicate ahigher proportion of their income tohousing than their European counterparts.This is especially the case for Vancouver,the city with the highest house prices. Itappears that many European homeowners,with a longer period of residence in Canada,have paid off their mortgage or havesubstantially lower mortgage costs thanVisible Minority households. In additionto a longer period of residence on average,European homeowners often purchasedhousing when prices were considerablylower than currently. In contrast tohomeowners, tenants generally spend agreater proportion of their income onhousing with much less difference betweenthose of Visible Minority and Europeanorigins. This is especially true in Torontoand Vancouver, cities that generally hadlower rental vacancy rates and higher rentalhousing costs than Montréal during thisperiod. Especially for Toronto andVancouver, groups with below-average

incomes and limited financial resourcesseem to be “blocked” in the rental housingmarket, a situation that surely becameworse with the escalation in real estate pricesthat followed the 2001 data examined here.On the other hand, there is relatively littlevariation in the share of income devotedto housing across the ethno-cultural groupsof Montréal (with the exception of Filipinotenants). In Montréal, ethno-cultural groupsappear to have found housing thatmatches their income levels, especially inthe rental market.

In each of the metropolitan reports, wehave surveyed selected 1996 census datato explore the dynamics of immigrantsin the housing market in the five-yearperiod ending in 2001. Unfortunately,the 1996 data are not as detailed as the2001 information; nevertheless, there isenough information to reveal importanttrends that warrant monitoring. Comparingthe situation of recent immigrants in thetwo census periods (1986-1996 vs. 1991-2001) we find that average householdsize fell in all three CMAs but especiallyin Toronto and Vancouver. We speculatedthat this is the result of a greateremphasis on economic-class immigrationas opposed to family reunification. Inkeeping with the drop in average householdsize, we found that the proportion ofmultiple-family newcomer householdsdeclined in all three CMAs.

There were quite different dynamics atwork in the housing markets of the threeCMAs in the five-year period between1996 and 2001: Toronto saw a sizeablerise in the real price of housing as well asin the rental market; house prices rose inMontréal but rents declined a smallamount; the average price of a house fellin Vancouver in real terms, which wasalso true of rental accommodations,though only by a tiny amount (Table 3).

These changes in the housing marketintersected with equally complex shifts inincome. Table 4 summarizes inflation-adjusted income figures provided in themetropolitan reports, as well as the levelsof homeownership. The Torontoeconomy was the most buoyant in the1996-2001 period. Household incomeswere highest in Toronto in 1996 andconsiderably more so in 2001.Significantly, immigrants in general, andnewcomers in particular, shared in theoverall economic improvement registeredin these years. Despite increasing realestate prices and rents, then, the level ofhomeownership rose in Toronto, especiallyfor newcomers. Income levels in Montréalhave been the lowest of the three CMAs,as we have seen. As in the Toronto case,real incomes rose in Montréal (thoughnot as rapidly as in Toronto), especiallyfor households identified as recentimmigrants. The level of homeownershipincreased generally in Montréal in keepingwith rising overall prosperity. However,homeownership rates are very low amongnew immigrants. Moreover, homeownershiprates of the Canadian born haveovertaken those of immigrants. Indeed,long-standing immigrants had a lowerrate of homeownership in 2001 than in1996. This outcome appears to be relatedto the unexpectedly low rate ofhomeownership of immigrants whoarrived in the 1980s, a fact that is beyondthe scope of this synopsis. Unlike theToronto and Vancouver cases, recentimmigrants in 2001 were only veryfractionally more likely to be homeownersthan their 1996 counterparts. Realhousehold incomes in Vancouver in 1996were fairly close to those in Toronto, butdid not keep pace between 1996 and2001; the rate of increase in Toronto wasmore than double that in Vancouver for



all of the groups identified in Table 4.The overall level of homeownership wasslightly higher in Vancouver than Torontoin 1996, but this relationship reverseditself by 2001. In contrast to Montréal,though, the level of homeownership rosefor more settled immigrants inVancouver and fell for newcomers.

This brief and partial glimpse into thedynamics of immigrants in the housingmarkets of the three CMAs reveals thatthere were different trajectories of housingconsumption in Montréal, Toronto, andVancouver. In essence, immigrants adaptedto changes in the housing markets ofthese metropolitan areas in somewhatdifferent ways. To a degree, the traditional

concept of a progressive housing career—the generalization that immigrants findrelatively cheap housing upon arrival andmove to better residences, and higher levelsof ownership, over time—remains relevant.But our analysis has shown that there areimportant exceptions to this standardexpectation, especially when we takeethno-cultural differences into consideration.



In each of the three reports we focusedon the experiences of households that arestruggling in the housing market byprofiling those we defined as “at risk”(spending at least 30 per cent of theirpre-tax, total income on housing) and“at high risk” (at least 50 per cent). Someof these households—those who own theirdwellings—may be acquiring equity andbe therefore at the starting point of aprogressive housing career, especially thosewho are not far above the 30 per centthreshold of payments identified here.But many of the households who ownproperty in these categories are at risk oflosing their investment (and home) ifthere is any significant interruption intheir income. Those in the rental market,of course, are not building equity. Quitethe opposite, by remaining outside theownership side of the market, they maybe falling further behind as housingprices and rents escalated, especially after2001. Moreover, they are dedicating alarge share of their income to housingand may be vulnerable to homelessness ifthey face a sustained loss of income.

The “at risk” population of homeownersis identified in Table 5. Broadly, thisincludes one-sixth of homeowners inMontréal, one fifth in Toronto, and one-quarter in Vancouver (though it isimportant to remember that the absolutenumbers are highest in Toronto giventhe much larger stock of immigrants in

that metropolitan area). These proportionsare understandable given the relationshipbetween household incomes and real estateprices in the three CMAs. Vancouver, forexample, is situated in the middle of thethree in terms of income, but at the topin terms of house prices. The result is ahigh ratio of at-risk homeowners. Assuggested earlier, some of these householdswould have seen a remarkable increase intheir equity since 2001, as long as they wereable to maintain their mortgage payments.As would be expected, in all three CMAs,newcomers who have purchased housingare most likely to be financially stretched.Nearly half of those who landed between1996 and 2001, owning a house and livingin Toronto, fall into this category, andmore than half in Vancouver. We wonderwhere new immigrants have obtained somuch capital given the relatively highrisk involved. Previous research on earliergroups of immigrants who attained highrates of homeownership in spite ofmodest socio-economic status (e.g. thePortuguese in Montréal and Toronto) leadsus to presume that many of those withrelatively low incomes were able to obtainappropriate credit on the basis of savingsand financial resources accumulated intheir home countries. Also, as notedearlier, some newcomers may be able toafford ownership housing by poolingincomes in multiple-family households.

Data explored in the individual studies alsoreveal another facet of the income-costsqueeze for those households whodedicate at least 30 per cent of their incometo house payments: on average they paymore, in absolute terms, in monthlypayments than all homeowner or tenanthouseholds. For example, in Toronto,newcomers who landed between 1996and 2001 and who owned homes, spentan average of $1,443 on monthly paymentsrelated to their housing, but those whowere in the at-risk category spent anaverage of $1,616. At the same time, thelatter group has well-below-averagehousehold incomes ($36,533 comparedwith $64,570).

The basic statistics on at-risk tenants areprovided in Table 6. Again, the proportionof this group is highest in Vancouver andToronto, followed by Montréal. But thereis relatively less variation in the ratio ofat-risk tenants across the three CMAs;broadly, 40 per cent of tenants in generalare in this category, and half of those whoare recent immigrants. In essence, thebottom of the housing market is “flatter”across the three CMAs. As in the case ofhomeowners, at-risk tenants face apunishing income-cost squeeze, theirhousehold incomes are well belowaverage, yet they pay above-average rents.


Chapter 4

Vulnerable Households

The proportion of at-risk households wasactually higher in 1996 compared with2001 (Table 7). In the intervening period,rising incomes translated into smallerproportions of households paying at least30 per cent of their income on housing.This was the case for both owners andtenants, in Toronto and Montréal. InVancouver, the proportion of at-risktenants declined as well. There is oneexception to this general pattern ofimprovement in housing affordability—thestable percentage of homeowners inVancouver spending more than 30 per centof household income on housing. Housingaffordability has not improved forhomeowners at risk in Vancouver, but ithas not deteriorated either. For recentnewcomers who arrived less than 10 yearsago, affordability has improved. Thepercentage of homeowners spending morethan 30 per cent of household income onhousing has declined from 50.1 per centto 45.9 per cent. This trend suggests thatrecent immigrant homeowners inVancouver may well enjoy the sameimprovement in affordability as those inMontréal and Toronto in the future.

The basic patterns in Table 8, whichprovides figures for households in ourhigh risk category (spending at least 50 per cent of their income on housing)largely coincide with those just explored.That is, fewer households face thisdramatic cost-income squeeze in Montréalcompared with Toronto and Vancouver.Homeowners, across the CMAs, are alsoless susceptible to this form of vulnerability.However, recent immigrants in all threecentres are much more likely than averageto fall into this unfortunate category.The situation is particularly difficult inVancouver, where one-third ofnewcomers—regardless of their tenurestatus—spend at least half of their householdincome on housing. Given that somehomeowners may have other sources ofwealth, this is particularly troublesomefor renters.

Finally, Table 9 adds another element tothis discussion, by comparing the degreeof vulnerability between Visible Minorityand European-origin households. It isimportant to keep in mind, when viewingthese data, that only immigrants areincluded. However, Visible Minorityhouseholds are much more likely to berecent immigrants, while people fromEuropean backgrounds are more likely tohave lived in Canada longer. Approximately

one in twelve European-origin homeowners,across the three CMAs, fall into the highrisk category and one in five are at risk.The corresponding figures for members ofVisible Minority groups are much higher,in fact nearly twice as high in Vancouver.As would be expected given our earlierfindings, tenants are far more stretched,financially, than homeowners, with muchhigher housing costs relative to theirincomes. However, we have been somewhatsurprised to learn that the proportions ofEuropean-origin vs. Visible Minority atrisk and high risk households are quitesimilar, especially in Montréal and Toronto.We are unsure how to interpret this result,since we know that incomes are loweramong Visible Minority groups. Wesuspect that there are important, unseen,demographic effects in these data. Forexample, it is possible that theEuropean-origin groups include a highnumber of retired households, as well asyoung single people in the case of Montréal,while Visible Minority groups include ahigh number of newcomers, relativelyfew of whom are retired. That is, bothcategories may contain around the sameratio of low-income households, butthey are different types of households.Unfortunately we cannot verify thispoint given the data at our disposal.



In our metropolitan reports, weconsidered a new source of informationon the settlement of immigrants, theLongitudinal Survey of Immigrants toCanada (LSIC), and conducted ananalysis of the early housing conditions,experiences and needs of newcomersapproximately six months after theirarrival in Canada.1 Our overarching goalhas been to understand how the initialimmigrant settlement experience is relatedto our longer-term findings, based onthe census analysis discussed so far inthis report. There are actually four majoringredients in LSIC that make it especiallyapplicable to our research: it enables usto see, for the first time, the initial housingcircumstances of immigrants; it includesa wider set of variables than the census,such as household savings; it specifies theadmission class of immigrants; and it is alongitudinal survey, tracking changes incircumstances over time. In this report,we will concentrate on the first three ofthese special features of LSIC. We planto devote our attention to the dynamicside of LSIC in future work, followingthe release of the full three waves ofsurvey results. However, we include apreliminary investigation of one longitudinalfeature in LSIC in Appendix 1 of thisreport, which compares the process ofhousing acquisition over time in the

three metropolitan areas. There is also afifth aspect of LSIC that makes it attractivefor housing research. The Canadian versionof this survey was actually adapted froman Australian model. In fact, theAustralian government has conductedthree separate longitudinal surveys inrecent years. We have provided a briefsummary of the results of thesesurveys—with respect to housing andimmigration in Australia—in Appendix 2.

Unfortunately there is also a majordisadvantage associated with LSIC data: weare unable to provide detailed breakdownson most variables due to the combinationof the relatively small sample in LSICand the confidentiality regulations ofStatistics Canada. Our tables havetherefore had to be constructed withrather coarse categories.

LSIC data corroborate the influence ofthe specific nature of the housing marketsof the three centres, discussed earlier.The majority of newcomers in Montréallive in low-rise apartments (57 per cent),while the modal categories of dwellingtype are high-rise apartments in Toronto(47 per cent) and single detached housesin Vancouver (36 per cent). In terms oftenure, the majority of newcomers in allthree centres are tenants (88 per cent inMontréal, 73 per cent in Toronto, and

74 per cent in Vancouver). The figures forhomeownership are much more variable,though, and range from a high of 20 per centin Vancouver, through 17 per cent inToronto, to the much lower figure of 5 per cent in Montréal. Surprisingly, 6 per cent of newcomers in Vancouverhad already paid in full for their homesby the time the survey was recorded;corresponding figures were 3 per cent inToronto and 1 per cent in Montréal. Recallthe high real estate prices characteristicof Vancouver and Toronto. Two types ofimmigrant households are able to purchasehomes quickly—some in cash—in thesemetropolitan areas: those that havealready-established family members inCanada; and those that bring substantialcapital when they arrive. This point iscorroborated by data on the relationshipbetween admission class andhomeownership (Table 10). In all threecentres, the groups with the highestpropensity to purchase housing wereeither family-class immigrants or whatwe have defined as “other economic” forour research, which is mainly comprisedof business-class immigrants.

Given the different profiles of immigrantsarriving in the three metropolitan areas,it should come as no surprise that theethno-cultural composition of newcomers


Chapter 5

The Initial Experience ofImmigrants in the HousingMarket: Insights from the LSIC

1 We discuss the nature of these data and our analytical approach in Mendez et al, 2006.

who purchase homes soon after theyarrive differs between CMAs. In Montréal,immigrants from Europe have the highestpropensity to be homeowners (nearly 10 per cent). In Toronto, South Asians arethe leading group in this category (over19 per cent), which is probably related totheir frequent adoption of multiple-familystrategies. Finally, newcomers who identifiedthemselves as East Asian had the highestrate of homeownership in Vancouver(nearly 23 per cent), though South Asianswere not far behind. In the Vancouvercase, early homeownership is most closelyassociated with the group that contains thelargest number of business-class immigrants.

Around half of the immigrant householdssettling in the three metropolitan areaswere dual-parent families with children(Table 11).2 The ratio of this householdtype was somewhat lower in Montréaland higher in Vancouver, which standsto reason given the different profile ofimmigrants in each centre (more refugeesin Montréal, more economic-classimmigrants in Vancouver). LSIC providesevidence that the multiple-family strategyis adopted quickly, especially in Torontoand Vancouver, where one in seven oreight households (respectively) falls intothis category. The incidence of thishousehold type is lower in Montréal, whichtakes in a higher percentage of youngcouples without children and young singlepeople, and also has a more accessiblehousing market, in terms of rental prices.

A relatively high proportion of immigranthouseholds in the three metropolitan areaslive in crowded circumstances (defined inthis study as those households that havemore than one person per room used

exclusively for business purposes).3 Again,for reasons cited above, the degree ofcrowding is lower in Montréal than inthe other two centres, but even there oneimmigrant household in five faces thisproblem. Further, the level of crowdingis particularly high for refugee householdsin Montréal. In Toronto and Vancouver,immigrants who came to Canada to joinfamily already here, or who came as refugees,are most likely to live in crowdedcircumstances. The level of crowding,generally, is highest in Toronto, reflectingthe cost of rent in that metropolitan area.

The level of crowding was dramaticallydifferent for European vs. Visible Minorityethno-cultural groups. In all threemetropolitan areas, the latter group wasat least twice as likely to be have morethan one person per room in their dwelling.Among Visible Minority groups, thereseem to be two forms of crowding. Onthe one hand, some groups, such asSouth Asians, have both high levels ofhomeownership and crowding. Crowding,in this case, is associated with ahomeownership strategy. But for othergroups, most notably the extreme case ofWest Asians in Montréal, crowding isassociated with low incomes.

A clearer picture of the relationshipbetween income and housing is providedin Table 14; in this and subsequent tableson income we focus exclusively on thelarge majority of newcomers who havenot purchased a home. We emphasizetwo simple but important points aboutthese data. First, half, or more, of thenewcomers living in the three centreswho do not own their homes spend atleast 50 per cent of their total household

income on housing. Secondly, this ratiois similar across the CMAs. In essence,the higher cost of rental housing iscompensated for by greater access toemployment and correspondingly higherhousehold incomes. Approximately 48per cent of LSIC respondents in Torontohad found employment by the time ofthe survey, compared with 39 per cent inVancouver and 31 per cent in Montréal.

Given the sharp difference in the level ofcrowding experienced by European-originversus Visible Minority newcomersdiscussed earlier, it may be surprising forreaders to learn that both groups facesimilar financial constraints in the housingmarkets of the three metropolitan centres(Table 15). In fact, while there are a fewexceptions, such as South Asians inVancouver and South East Asians inMontréal (a relatively small group),generally speaking, a considerable majorityof households across all three CMAs, inall ethno-cultural communities, dedicatemore than 30 per cent of their income tohousing. In other words, European-originand Visible Minority immigrant householdsdevote similar proportions of their incometo housing, but the former groupappears to be more adequately housed.

However, newcomers do not rely onincome alone to support their housingneeds. Most bring savings with them toCanada. In fact, applicants to the SkilledWorker class, the largest single form ofentry into Canada, will only be admittedto Canada if they have sufficient fundsto support their needs for six months whilethey settle (but only three months in thecase of skilled workers selected for entryto Québec). Given that the first wave of



2 Note that the totals do not match across LSIC tables. The reason for this discrepancy is discussed in the metropolitan reports.3 Readers should note that this is not the same as the National Occupancy Standard, which takes into account household composition variables such as age, gender, and

parental and marital status and determines if the unit has sufficient bedrooms for the size and makeup of the household .

LSIC interviews took place around thesix-month point, we cannot understandthe housing consumption of thesenewcomers without knowledge of theirsavings as well as their income. Fortunately,information on both is included in LSIC.In order to assess the role that newcomers'savings might play in the context of theirresidential situation, we developed a specialhousing stress index for the non-homeownersubset of survey respondents. Ourassumption is that savings act as a financialbuffer that would allow newcomers tocover some portion (up to 100 per cent)of their monthly housing costs. Combiningthis savings information with reportedincome-to-rent allocations, we createdthree categories of housing stress,ranging from No Housing Stress toExtreme Housing Stress (Table 16).

Our housing stress index reveals a newpicture of the financial challengesconfronting newcomers to Canada.Immigrants to Vancouver arrive with thehighest average level of savings and, sixmonths later, are relatively well-positionedin the housing market compared withthose in the other two CMAs. Only 43 per cent of newcomers in Toronto (asopposed to 58 per cent in Vancouver)either have sufficient income or savingsfor another year of reasonable housingexpenditures, and one in fourhouseholds are in the precarious extremestress category. Despite low rental fees inMontréal, the situation is even worsethere. Just over one-third of householdsface extreme stress in the housing market,and another quarter face moderate tohigh stress. Immigrants in Montréal havethe lowest employment rate, the lowestincomes, and the lowest financial buffer

in the form of savings, compared withToronto and Vancouver.

As would be expected, the extent ofhousing stress varies across ethno-culturalgroups, but not necessarily in ways thatwould be predicted given the literatureon immigrant settlement in Canada. Inparticular, in all three CMAs, newcomerswho identified as European (“White”)were more likely than average to beexperiencing housing stress. As notedearlier, European-origin newcomers faceless crowded circumstances in their homes,but this form of housing consumptioncarries substantial costs for households inthis group, and their degree of financialstress is high. The number of VisibleMinority respondents in LSIC issufficient to enable greater group-specificdetail. The situation of East Asians, whodevote a very high proportion of theirincome to housing (Table 15), appearsto be much better when their substantialsavings are taken into consideration; theproportion of those falling into themoderate-to-extreme stress categories isbelow-average in all three centres.Significantly, the degree of housing stressfor most other groups varies dramaticallybetween CMAs. Newcomers whoidentified as Black, for example, face thegreatest level of financial difficulty of anygroup in the Toronto housing market,which is simply not the case in Montréal.Similarly, South Asian newcomers inToronto are experiencing far greaterfinancial challenges securing housing inToronto compared with either Vancouveror Montréal. Although data are limited,it seems that Latin American and WestAsian newcomers are under severehousing stress in Montréal and Toronto,

a product of a combination of lowincomes and levels of savings.

LSIC respondents were asked about theirgreatest difficulties in securing housing,and the forms of help they received intheir housing search. Unfortunately,much of these data have been suppresseddue to confidentiality constraints(sample numbers are small since thesequestions were only answered by thosewho had experienced difficulties in thehousing market). In keeping with all ofthe LSIC data examined thus far,newcomers in Toronto and Vancouveridentified cost as their most significantobstacle in the housing market whereasin Montréal, difficulties obtaining creditor finding a guarantor were just asimportant as cost. The small amount ofdata that have been made available toresearchers suggest that different groupsof immigrants rely on different sourcesof information about housing. InMontréal and Toronto, for example,individuals admitted as Skilled Workersmade extensive use of family and kinshipnetworks when searching for housing,while refugees rarely were able to do so.There are also intriguing group-specificdetails. For example, in Toronto andVancouver, East Asian newcomers weregenerally much more able than membersof other groups to tap into friendshipnetworks when searching for housing.While interesting, data on the relianceupon social networks in the housingmarket are insufficient to enable us tomake any firm generalizations about thisimportant issue.



Any cross-sectional study of a complex anddynamic phenomenon, such as immigrantsettlement, is bound to raise as manyquestions as it answers. We have used tworesources to provide a portrait of immigrantparticipation in the housing markets ofMontréal, Toronto, and Vancouver at thestart of the current decade. We emphasizethe importance of the timing of our data,which were mainly collected in 2001,during relatively favourable economicconditions and on the eve of a major risein housing prices across the country.Moreover, the Canadian government hadrevised immigration policy in the early1990s, raising the ratio of economic-classimmigrants, those who would be mostlikely to succeed in the Canadian labourmarket. Given these contexts, we wouldexpect immigrants to be well situated inthe housing markets of Canadianmetropolitan areas in 2001.

There are several indications that thissupposition is correct. Our briefcomparison of 1996 and 2001 censusdata reveals that immigrants weregenerally better off in terms of housingin the latter year: household incomeshad grown more rapidly than both rentalfees and the price of real estate in theintervening five-year period. Secondly,

LSIC data indicate that a surprisinglyhigh proportion of newcomers—who hadonly been in Canada six months—werehomeowners in 2001-2002. We cangeneralize this point further. Accordingto the census, there appears to be an“immigrant effect” in the housing market;that is, in their first 10 years in Canada,immigrants achieve a higher rate ofhomeownership than their incomes wouldlead us to expect. This is accomplished,we have suggested, through assemblinglarger households to pool incomes, andthrough the use of savings brought toCanada. Thirdly, a relatively small ratioof new immigrants, in all three CMAs,indicated that they faced difficulties inthe housing market (Mendez et al,2006). Finally, there is clear evidencethat the housing situation of immigrantsimproves over time. In fact immigrantswho have been in Canada more than 20 years are, arguably, better housedthan the Canadian-born population.

But, on the other hand, the data exploredhere also generate a number of concerns.This story of prosperity entails great costfor most newcomers, who devote a highproportion of their financial resources tohousing. Savings are depleted, andapproximately half of all household

income is dedicated to housing in theinitial few years of settlement. Further,some immigrants appear to becometrapped in the housing market, since wefind that a significant proportion ofimmigrants who have been in Canadafor 5, 10, and even longer (in the 2001census) pay a high ratio of their incomefor housing. This is particularly the casefor those immigrants who have notmanaged to purchase housing.

These general findings suggest severalimportant policy directions. First, thesupply of affordable rental housing inmajor Canadian metropolitan areas isnot keeping pace with demand. Untilrecently, vacancy rates have been fallingand rents increasing at a time when largenumbers of immigrants are being admittedto Canada. Even though the rental markethas improved in some cities, many of thevacancies are at the most expensive endof the market. A large majority of thesenewcomers rely on the rental marketand, as noted above, may haveaffordability problem. An increasedsupply of affordable rental housingwould help immigrants and theCanadian-born alike.


Chapter 6

Conclusions

Secondly, much more detailed informationabout the Canadian housing marketshould be made available to immigrantsprior to their journey to Canada.

Thirdly, there is no specific institutionthat is responsible for helping newcomersenter the Canadian housing market.Governments themselves do not offerorientation services to immigrants; thisrole is performed by non-governmentorganizations (NGOs), which areprovided funds by government for thispurpose. When governments determinethe level of funds that they will allocateto NGOs, they do so on the basis of aspecific set of services that should beprovided to newcomers. That is, NGOsare funded to provide certain types ofsettlement services, such as languageinstruction and employment counseling,and not others. They adjust the servicesthat are offered accordingly. For themost part (that is, in most provincialjurisdictions), NGOs are allocated fundsto provide basic housing orientation andto refer immigrants to other, morespecialized, housing services. They aregenerally not funded to directly helpimmigrants find housing (with the exceptionof Government Assisted Refugees). This,we believe, should change. NGOs shouldbe encouraged, with appropriate fundsprovided, to develop additional expertiseand services in the area of housing. Mostnewcomers rely on friends and familymembers for housing assistance;

nevertheless, it is important for NGOsto reach those who would benefit frommore formal assistance.

Beyond these general findings, our reportsshed light on the specificities of metropolitanhousing markets in Montréal, Toronto,and Vancouver, especially with respect tothe place of immigrants within them.Montréal receives a much smaller numberof immigrants in proportional terms thanthe other two CMAs examined here. Thevast majority of new immigrants enterthe rental market in Montréal, which isoften characterized as a “city of tenants”more generally. Even those immigrants whohave been in Montréal a long time stillhave much lower rates of homeownershipthan their counterparts in Toronto andVancouver (though their rates are highrelative to the Montréal population as awhole). In part, lower homeownership ratesreflect lower incomes, for immigrantsand the whole population. But there isanother side to this issue: rents are relativelyreasonable in Montréal. Immigrantstherefore are able to live in smallerhousehold units and face less crowdingthan in the other CMAs. Nevertheless,many immigrants in Montréal, especiallynewcomers, face a similar housingaffordability crisis. On this level, despite themore open housing market in Montréal,low-income immigrants living there facethe same challenges as those elsewhere.

In contrast, Toronto receives the largestabsolute number of immigrants in Canadaand also has the largest share of immigrants,including newcomers, of the three CMAs.As noted earlier, this is an exceedinglyvariegated population in terms of sourceregions, class of entry, and socio-economicposition. In general terms, incomes arehigh in Toronto relative to the rest ofCanada, and immigrants in Toronto havemuch higher rates of employment andincome than those in Montréal orVancouver. Homeownership rates aretherefore relatively high for immigrantsin Toronto, especially those who havebeen in Canada 20 years or more. This isthe case even for recent immigrants, seenin the 1996-2001 census cohort as wellas in LSIC data on newcomers. LSIC, infact, provides an important nuance to thispoint: family- and business-class immigrantshave, by far, the highest ownership ratesamong newcomers. However, not allimmigrants share in this prosperity.Homeowners struggle with highpayment/income ratios, and tenants inToronto face the highest rents in Canada.Despite higher immigrant incomes inToronto (relative to Montréal andVancouver), immigrants there face acuteaffordability challenges.



Many attempt to overcome financialbarriers by “doubling up”, that is, livingin multiple-family households. This isparticularly true of visible minoritygroups, and much less so for those ofEuropean origin. Together, LSIC and thecensus show that these problems emergealmost immediately for newcomers, andyet are of long duration. Many whobecome marginalized in the Torontohousing market, remain marginalized,despite the more general trend ofprogressive housing careers. We areconcerned that the likelihood of long-term marginalization is particularlyhigh for refugees.

There are also many distinct aspects of theVancouver situation. Vancouver receives,by far, the largest share of immigrants fromEast and South Asia, and both groupshave achieved relatively high levels ofhomeownership. It is interesting to notethat these groups have high rates ofownership across all three CMAs, butthey are most significant, numerically, inVancouver. Therefore, we see thestrongest indication of the “immigranteffect” in Vancouver’s housing market,and the concept of a progressive housingcareer is most applicable in Vancouver.This outcome actually happens veryquickly, since the particular groups ofimmigrants that settle in Vancouver—especially from East

Asia—arrive with savings that are investedin the real estate market. South Asians,on the other hand, are most prone toutilizing multiple-family strategies togain homeownership. It is important toacknowledge that, whether ownership ispurchased or attained through crowding,it entails sacrifice (whether of savings orspace, or both). Also, the progressivehousing career that is built by many isnot universally shared; there is clearevidence, in Vancouver as in Montréaland Toronto, of a significant number ofimmigrants that face long-termmarginalization in the housing market.

Summarizing our findings as succinctlyas possible, our investigation ofimmigrants in the housing markets ofthe largest Canadian metropolitan areasshows that: over time, many immigrantsexperience “progressive” housing careers(i.e. over time, improving their livingconditions and eventually joining the“mainstream” by becoming homeowners);this is achieved through considerablesacrifice by newcomers. Not all share inthis positive trajectory, and we see abifurcation in the long-term fortunes ofimmigrants in the housing market; thisbifurcation begins to take shape quickly, andthose who do move into homeownershipfrequently purchase housing even thoughthey do not have high incomes (the“immigrant effect”). There are alsostructural differences in the housingsituations of European-origin and visibleminority immigrants, even though thereare important differences within thesecoarsely-defined categories.




Chapter 7

Tables

Table 1 Tenure, Income and Monthly Housing Costs by Immigration Period, 2001,Toronto, Montréal, and Vancouver CMAs

Number ofhouseholds

Per centowners

Averageincomeowners

($)

Averagemonthlyowner’smajor

payments($)

Per centowners

paymentsof income

Per centrenters

Averageincomerenters

($)

Averagemonthly

grossrent ($)

Per centtenantsrent ofincome

Toronto Total 1,625,980 63.4 94,706 1,171 14.8 36.6 45,728 870 22.8

Immigrants 859,665 63.7 84,016 1,144 16.3 36.3 41,399 832 24.1

Immigrated 1996-2001 109,440 32.0 64,570 1,443 26.8 68 38,591 890 27.7

Montréal Total 1,411,835 50.3 72,951 845 13.9 49.7 34,624 568 19.7

Immigrants 297,785 48.1 68,016 883 15.6 51.9 32,994 572 20.8

Immigrated 1996-2001 36,450 10.8 61,445 1,082 21.1 89.2 28,611 550 23.1

Vancouver Total 750,250 61.2 77,083 1,057 16.5 38.8 41,640 814 23.5

Immigrants 303,490 66.3 70,196 1,053 18.0 33.7 38,897 798 24.6

Immigrated 1996-2001 45,060 40.9 48,678 1,207 29.8 59.1 35,067 820 28.1

Source: Census of Canada, 2001, special tabulations prepared for the Metropolis Centres of Excellence



Table 2 Per Cent of Total Household Income Devoted to Monthly Housing Costs by Ethnic Origin and Visible Minority Subgroup, 2001,Toronto, Montréal, and Vancouver CMAs


Toronto Montréal Vancouver

Owners Renters Owners Renters Owners Renters

Total 16.3 24.1 15.6 20.8 18.0 24.6

Total visible minority population 19.3 24.7 17.9 21.4 20.7 25.7

Black 22.1 25.3 20.1 22.2 22.7 22.5

South Asian 19.6 24.4 15.5 20.2 21.8 20.8

Chinese 17.3 26.3 17.2 22.8 19.4 28.0

Southeast Asian 19.6 22.3 15.8 18.4 25.5 29.2

Filipino 18.2 19.2 16.3 15.9 21.6 20.7

Arab/West Asian 23.3 29.2 18.3 22.9 25.0 30.8

Latin American 22.4 23.2 20.6 22.5 25.6 26.1

All other ethnic origins 14.0 23.1 14.6 20.1 14.6 23.2

British Isles 12.9 21.9 11.3 19.8 14.4 22.6

French 14.1 22.1 13.9 19.6 14.7 22.3

Canadian 16.5 24.8 14.7 20.2 16.7 22.3

Other European ethnic origins 14.2 23.5 15.0 19.9 14.7 23.7

Polish 16.4 22.9 14.8 20.8 18.5 23.2

Italian 13.0 23.5 15.3 18.9 14.4 23.3

Other single or multiple ethnic origins 18.2 24.1 15.3 21.6 16.5 25.9



Table 3 Housing Price and Rental Changes, Between 1996 and 2001, Montréal,Toronto,and Vancouver CMAs

1996 2001 Change in realdollars

Average house price (MLS) Montréal 105,729 125,744 8.3

Toronto 196,476 251,508 17.4

Vancouver 288,268 285,910 -11.4

Average 2-bedroom apartment rent Montréal 491 529 -2.9

Toronto 819 1,027 14.8

Vancouver 845 919 -1.8

NOTE: a change of 10.6% in the CPI was used in the calculation of change in real estate prices and rent levelsSource: CMHC, Canadian Housing Observer 2005

Table 4 Housing Price Changes and Ownership Rates Between 1996 and 2001 by Immigration Period,Montréal,Toronto, and Vancouver CMAs

1996 2001 Change from1996-2001

Number ofhouseholds

Averagehousehold

income

Per centowners

Number ofhouseholds

Averagehousehold

income

Per centowners

Change inhouseholdincome

Change inper centowners

Montréal Total 1,335,975 44,715 48.5 1,411,835 53,903 50.3 9.9 1.8

Immigrants 275,130 41,890 49.1 297,785 49,836 48.1 8.4 -1.0

Immigrated<10 years 74,270 27,693 16.8 74,780 36,037 17.5 19.5 0.7

Toronto Total 1,479,955 60,382 58.5 1,625,980 76,774 63.4 16.5 4.9

Immigrants 751,635 54,577 59.3 859,665 68,562 63.7 15.0 5.2

Immigrated <10 years 220,940 39,503 33.7 235,205 52,342 41.3 21.9 7.6

Vancouver Total 684,690 54,316 59.5 750,250 63,313 61.2 6.0 1.7

Immigrants 258,190 51,789 66.5 303,490 59,664 66.3 4.6 -0.2

Immigrated <10 years 78,880 39,117 56.5 95,085 46,446 52.3 8.1 -4.2

NOTE: a change of 10.6% in the CPI was used in the calculation of change in household incomeSource: Census of Canada, 1996, 2001, special tabulations prepared for the Metropolis Centres of Excellence



Table 5 Owners’ Housing Spending and Income by Immigration Period, 2001, Montréal,Toronto, and Vancouver CMAs


Per cent of householdsspending at least 30%

on housing

Average householdtotal income ($)

Average monthlyowner’s majorpayments ($)

Montréal Total 15.9 28,351 1,172

Immigrants 23.3 29,246 1,200

Immigrated 1996-2001 36.7 27,917 1,167

Toronto Total 21.5 40,100 1,540

Immigrants 25.2 38,822 1,533

Immigrated 1996-2001 49.7 36,533 1,616

Vancouver Total 24.0 38,914 1,544

Immigrants 29.4 36,878 1,559

Immigrated 1996-2001 53.3 28,149 1,495

Table 6 Tenants’ Housing Spending and Income by Immigration Period, 2001, Montréal,Toronto, and Vancouver CMAs


Per cent of householdsspending at least 30%

on housing

Average householdtotal income ($)

Average monthly rentalpayments ($)

Montréal Total 36.4 13,322 603

Immigrants 40.0 13,441 612

Immigrated 1996-2001 46.2 11,132 573

Toronto Total 42.2 20,805 897

Immigrants 44.7 19,851 872

Immigrated 1996-2001 52.1 18,711 935

Vancouver Total 43.2 18,970 849

Immigrants 46.0 18,746 869

Immigrated 1996-2001 52.8 16,809 904



Table 7 Per Cent of Owners and Tenants Paying at Least 30% of Income on Housing Payments by Immigration Period, 1996 and 2001, Montréal,Toronto, and Vancouver CMAs


Montréal Toronto Vancouver

1996 2001 1996 2001 1996 2001

Owners Total 19.3 15.9 23.9 21.5 23.4 24.0

Immigrants 27.7 23.3 27.7 25.2 29.5 29.4

Immigrated less than10 years ago 43.3 33.5 48.0 41.6 50.1 45.9

Tenants Total 57.2 36.4 44.0 42.2 46.7 43.2

Immigrants 50.5 40.0 50.5 44.7 49.9 46.0

Immigrated less than 10 years ago 55.0 41.8 52.0 46.3 55.0 48.5

Table 8 Per Cent of Owners and Tenants Paying at Least 50% of Income on Housing Payments by Immigration Period, 2001, Montréal,Toronto, and Vancouver CMAs



Owners Tenants Owners Tenants Owners Tenants

Total 6.3 18.1 7.8 20.0 9.6 22.3

Immigrants 9.2 20.4 9.9 21.6 13.5 24.4

Immigrated 1996-2001 15.7 27.8 25.0 30.9 33.3 33.1



Table 9 Per Cent of Owners and Tenants Paying at Least 30% and 50% of Income on Housing Payments by Visible Minority and European-Origin Households, 2001, Montréal,Toronto, and Vancouver CMAs


Total population Visible minorityhouseholds

European-originhouseholds

Montréal Housing payments at least30% of income

OwnersTenants

23.340.0

28.040.1

21.339.8

Housing payments at least50% of income

OwnersTenants

9.220.4

11.621.7

8.218.8

Toronto Housing payments at least30% of income

OwnersTenants

25.244.7

31.444.1

20.145.9


OwnersTenants

9.921.6

12.921.8

7.521.2

Vancouver Housing payments at least30% of income

OwnersTenants

29.446.0

36.347.7

19.543.5


OwnersTenants

13.524.4

18.127.1

6.920.4

Table 10 Homeownership by Admission Class, 2001, Montréal,Toronto, and Vancouver CMAs

Source: LSIC, master file

Number Family class Skilledworker

class

Othereconomic

classes

Refugees Total

Montréal Total 21,350 4,450 14,900 750 1,250 21,350

Homeowners (no., %) 1,200 15 2 20 x 6

Toronto Total 74,250 19,400 49,000 3,350 2,500 74,250

Homeowners (no., %) 12,950 39 8 37 x 17

Vancouver Total 24,000 6,450 13,250 3,250 1,050 24,000

Homeowners (no., %) 4,850 34 11 37 x 20.2



Table 11 Household Structure by CMA, 2001, Montréal,Toronto, and Vancouver CMAs



Number 21,550 73,350 24,500

Couple with child(ren) (%) 44 50 56

Couple without child(ren) (%) 23 17 16

Lone parent family (%) 4 3 3

Single person (%) 16 8 6

Family and non-family person(s) (%) 4 6 5

Multi-family household (%) 10 16 13

Table 12 Immigrant Class by Crowding§ and CMA, 2001, Montréal,Toronto, and Vancouver CMAs


§ Readers should note that crowding is defined in this study as those households that have more than one person per roomand follows the Census definition of crowding as a situation where there is more than one person per room in a household,excluding bathrooms, entrance halls, and rooms used exclusively for business purposes.This definition is not the same as theNational Occupancy Standard, which takes into account household composition variables such as age, gender, and parental andmarital status and determines if the unit has sufficient bedrooms for the size and makeup of the household.


Total Crowded (%) Total Crowded (%) Total Crowded (%)

Family 4,300 23.3 19,350 30.7 5,850 35.0

Skilled Worker 14,850 15.5 48,100 24.1 12,800 19.1

Other Economic 700 28.6 3,150 27.0 2,800 16.1

Refugee 1,150 65.2 2,500 52.0 950 42.1

Total 21,000 20.2 73,100 26.9 22,400 23.9



Table 13 Crowding by Ethno-cultural Group and CMA, 2001, Montréal,Toronto, and Vancouver CMAs



Total More thanone personper room

(%)


(%)


(%)

South Asian 1,450 48 26,500 32 4,050 42

Black 2,000 25 2,650 21

Filipino 4,650 33 2,200 25

Latin American 1,500 10

South East Asian (excl. Filipino) 1,000 15

Arab 1,350 30

West Asian (excl.Arab) 650 69 3,500 24

East Asian (Chinese, Japanese, Korean) 3,650 16 19,450 30 10,750 21

All other visible minority groups 6,950 22 2,650 25

Non-visible minority (European) 6,100 8 11,750 12 3,000 12

Total 20,800 20 72,350 27 22,650 24

Table 14 Cost of Housing Relative to Income for Non-homeowners by CMA, 2001, Montréal,Toronto, and Vancouver CMAs



Number of non-homeowning newcomers 20,250 61,400 19,400

Free lodging (%) 3 4 3

Spent below 30 per cent of income on housing (%) 22 16 20

Spent 30 to 49.9 per cent of income on housing (%) 22 18 17

Spent 50 per cent or more of income on housing (%) 50 56 51

Don’t know or refused (%) 4 7 9



Table 15 Non-homeowners Paying at Least 30 Per Cent of Income for Housing by Ethno-cultural Group and CMA, 2001, Montréal,Toronto, and Vancouver CMAs


Number 14,350 45,100 13,150

South Asian (%) 59 77 42

Black (%) 60 74

Latin American (%) 67 78

South East Asian (Including Filipino) (%) 44 68 70

Arab (%) 80 75

West Asian (minus Arab) (%) 77 86

East Asian (Chinese, Japanese, Korean) (%) 90 85 85

All other visible minority (%) 85

Non-visible minority (European) (%) 69 81 72

Total (%) 74 80 75


Table 16 Construction of the Housing Stress Index (for Non-homeowners Only)

Proportion of incomespent on housing

And/Or Savings remaining at timeof interview

No housing stress 0% to 30% OR Savings equal at least 12months of housing costs

Moderate to high housingstress

30% to more than 100% AND Savings equal less than 12months of housing costs

Extreme housing stress 50% or more AND Savings equal less than 3months of housing costs

Table 17 Non-ownership Households by Degree of Financial Stress and CMA, 2001, Montréal,Toronto,and Vancouver CMAs



Number 18,100 49,650 16,000

No housing stress (%) 40 43 58

Moderate to high housing stress (%) 26 33 25

Extreme housing stress (%) 34 24 17




Table 18 Non-ownership Households Facing Housing Stress (Moderate to Extreme) by Ethno-cultural Group and CMA, 2001, Montréal,Toronto, and Vancouver CMAs


Total number 10,750 28,000 6,700

South Asian (%) 13 66 30

Black (%) 17 72

Latin American (%) 68

South East Asian (Including Filipino) (%) 57 54

West Asian and Arab (%) 80 69

East Asian (Chinese, Japanese, Korean) (%) 15 34 32

All other visible minorities (%) 12 67

Non-visible minority (White) (%) 63 68 57

Total 60 57 42

Anisef, P., and C. M. Lanphier (2003).The world in a city. Toronto: Universityof Toronto Press.

Bezanson, R. Z. (2003). Make Yourself atHome: Exploring Housing and Resettlementwith Afghan Refugee Households inKitchener-Waterloo, Ontario. Waterloo:Unpublished Master's Thesis, Departmentof Geography, University of Waterloo.

Canada Mortgage and HousingCorporation (2004). 2001 Census HousingSeries Issue 7: Immigrant Households. Ottawa:Canada Mortgage and Housing Corporation.

Engeland, J., R. Lewis, et al. (2005).Evolving Housing Conditions in Canada'sCensus Metropolitan Areas, 1991-2001:Trends and Conditions in CensusMetropolitan Areas. Ottawa: StatisticsCanada. [http://www.cmhc-schl.gc.ca/en/homadoin/maintrst/stda/stda_008.cfm]

Lapointe Consulting Inc. with R. Murdie(1996). Housing and Immigration -Immigrants and the Canadian HousingMarket: Living Arrangements, HousingCharacteristics and Preferences. Ottawa:Canada Mortgage and HousingCorporation.

Mendez, P., D. Hiebert, E. Wyly (2006).“Landing at Home: Insights onImmigration and Metropolitan HousingMarkets from the Longitudinal Surveyof Immigrants to Canada” CanadianJournal of Urban Research 15(2): 81-104.

Murdie. R. (2005). “Pathways to Housing:The Experiences of Sponsored Refugeesand Refugee Claimants in AccessingPermanent Housing in Toronto, Canada.”Paper presented at the EuropeanNetwork for Housing Research Conference,Reykjavík, Iceland, July, 2005.

Murdie, R., V. Preston, S. Ghosh, andM. Chevalier (2006). Immigrants andHousing: A Review of CanadianLiterature From 1990 to 2005. ReportSubmitted to Canada Mortgage andHousing Corporation, August, 2006.

Renaud, J. et al (2003). What a DifferenceTen Years Can Make! The SettlementExperience of Immigrants Admitted toQuébec in 1989. Québec : Les Publicationsdu Québec. 222 pp.

Rose, D. and B. Ray (2001). "The HousingSituation of Refugees in Montréal ThreeYears after Arrival: The Case of AsylumSeekers who Obtained PermanentResidence." Journal of InternationalMigration and Integration 2(4): 493-527.


Chapter 8

References Cited

Jean Renaud and Karine Bégin,Centre d'études ethniques Universityof Montréal, Virginie Ferreira andDamaris Rose, INRS-Urbanisation,Culture et Société, Institut national dela recherche scientifique

Originally published in Canadian Journalof Urban Research, 2006, Volume 15,Issue 2, Supplement pages 67-81.

AbstractThis paper looks at a key aspect of newimmigrants' settlement experience—finding a home. Specifically, we examinethe factors determining the propensity,over the first six months of settlement, toremain in or move on from the firstresidence occupied since arrival inCanada. We consider in turn the effectsof various household and individualcharacteristics, and examine how theseeffects vary by urban region. Our datasource is the first wave of observationsfrom the (LSIC), carried out after anapproximate 6 month stay (n=12,040).Semi-parametric survival models are usedfor the analysis. We find that while theresidential mobility of this cohort in theinitial months after arrival is associated withcertain individual- and household-levelcharacteristics, the strongest associationis with the type of housing occupied. The

city of residence of these newcomers,however, has little bearing on theirhousing transitions.

Keywords: housing, residential mobility;immigration; immigrant settlement process;Canadian cities; longitudinal analysis

IntroductionWhile shelter is a basic need for all, it takeson a particular material and symbolicsignificance when people settle in a newcountry in that it represents the fulcrumfor a new start (Ray 1999; Ryan andWoodill 2000). Finding a home is one ofthe first settlement actions, if not the first,that a newcomer takes when making theinitial contact in terms of day-to-dayliving with the receiving society. In thissituation, an understanding of the firstsettlement activities is vital. Newcomersto Canada are highly diversified in termsof economic resources, links to the labourmarket, social networks and knowledgeof residential markets in the cities wherethey first settle (Statistics Canada 2005).It can be assumed, therefore, that whilesome succeed at once, or almost at once,in obtaining housing that will satisfy theirneeds and aspirations for some time,others tend to move more than once inorder to gradually improve their residentialquality of life, to be closer to those with

whom they have social ties, or to improvetheir job prospects. However, immigrantswhose economic status remains precariousface a more limited range of residentialoptions; this can entail frequent and moreor less forced moves resulting from, say,rent increases, which may impair theirpersonal stability and their social integration.For those immigrants, residential stabilitymay mean that they have succeeded infinding affordable housing, likely withhelp from their social network (Bernècheand Martin 1986; Miraftab 2000; Ray1998); this may or may not satisfy theirother needs. Residential mobility orstability among newcomers, and theirresidential trajectories or “careers” resultingfrom these dynamics, may thus have avariety of meanings, and the residentialadjustments they make in the initialmonths and years of settlement do notoccur in a vacuum and must be interpretedin light of the settlement actions takenby immigrants in other areas of theirlives (Murdie et al. 1999; Özüerkren andvan Kempen 2002), taking into accountthe filters and barriers they face in theresidential market (Bolt and vanKempen 2002; Murdie 2002).


Appendix 1

The Residential Mobility ofNewcomers to Canada: TheFirst Months

The residential transitions of recentimmigrants should thus be examinedlongitudinally, which was not possible ona Canada-wide basis until quite recently,with the completion of the LongitudinalSurvey of Immigrants to Canada (LSIC),and with the availability to researchers ofthe microdata from that survey. Thetarget population for the LSIC consistedof immigrants aged 15 or older arrivingin Canada between October 2000 andSeptember 2001 who submitted animmigration application to a Canadianmission abroad. Some 12,040 immigrantswere selected to participate in the firstwave of interviews after about six months'residence (Statistics Canada 2003 and2005).1 The same immigrants were alsoasked to take part in a second and a thirdwave of interviews after two and four years,respectively, of settlement. The surveygathered data on various aspects ofimmigrant settlement. Subjects such asthe socio-demographic characteristics of therespondent, language proficiency, socialnetworks, household composition andincome were addressed. Questions werealso asked about the respondent's housing,employment and education. Data werealso gathered on moving-in and moving-out dates, and on selected features of thehousing occupied in each place therespondents lived in. Since much of thedata included dates, dynamic study ispossible of the residential settlement ofthe immigrants, although the structure ofthe survey does not enable us to considerthe reason for each successive move as anexplanatory factor in the analysis.

We therefore took the opportunitypresented by the LSIC to explore thesurvey's potential to shed light on theresidential transitions that newcomersexperience. We restricted our study tothe residential adjustments occurring inthe first months of settlement, as onlythe data from the first wave of interviewswere available at the time. The objectiveof this paper will be, first, to describeimmigrants' residential mobility, andsecond, to answer two specific questionsabout their initial residences: Whatfactors affect how soon they leave theirfirst home? Are there factors peculiar tothe urban settings in which immigrantsare placed? In addressing these twoquestions, we will first identify thefactors—whether individual orhousehold-related—that hasten or delaythe transition. This will help usdetermine what characteristics affect thestability or mobility of newcomers. Analysisof the second question will showwhether cities offer different urbancontexts, and thus whether new immigrantsface different situations, depending onwhether they settle in Montréal, Toronto,Vancouver or elsewhere in Canada.

Methods and variablesIn our study of residential transitions, weused survival model analysis (Allison1991; Cleves et al., 2002). This involvesstudying the speeds of transition fromone status to another and identifying thefactors that affect them. It shows for eachunit of time the probability that a givenevent will occur. Survival curves and

semi-parametric regressions (or Coxregressions) were plotted.2 In this case,residential mobility is considered fromfour different angles related to the analysisof the transitions. First, we looked atinterurban and interprovincial mobility.Two survival curves were estimated,showing the rates at which immigrantschange their metropolitan region ofresidence (MRR) or census agglomerationof residence (CAR) and province. However,the main purpose of the analyses was toexplain mobility from one residence toanother, while continuing to pay specialattention to the dynamics related to theurban settings of residence. Survival curveswere prepared for the promptness of leavinga residence depending on the rank of theresidence occupied (first, second or thirdresidence since arrival), and the promptnessof leaving the first residence dependingon the city of residence.

The first variables introduced to explainhow soon people leave a dwelling relateto socio-demographic characteristics: therespondent's age, sex, immigration category(economic, family or refugee) andmembership in various kinds of visibleminorities. Added to these were variablesrelated to human capital. Two variablesrelate to knowledge of English and French.These are the indicators of knowledge ofthe official languages based on threequestions that ask respondents to assesstheir ability to speak, read and write eachof them. Another variable considered wasthe highest level of schooling attainedoutside Canada, whether primary orbelow, secondary, post-secondary or



1 For the first wave of observations, designed to describe the situation of respondents after 6 months' residence in Canada, longitudinal respondents were interviewed over aperiod varying from 5 to 10 months or so in the host society. Thus, length of residence is not the same for all respondents.

2 The regression results presented were resampled using “bootstrap” weights. This method of resampling is used to test data reliability and consists of extracting randomsubsamples (with replacement) from within the original sample to obtain an approximation of the actual variance. For the LSIC, Statistics Canada supplied a series of 1,000bootstrap weights for recalculating the variance for each estimate produced, and for determining its quality. Using these weights, we can determine whether the differencesobserved in the regression are statistically significant for the cohort studied.

university. Another variable was whetherthe respondent had lived in Canadabefore immigrating.

As the promptness of departure from adwelling is a phenomenon that involvesthe household,3 its characteristics mustbe considered among the causal factors.Such factors include savings and averagemonthly family income,4 as indicators offinancial independence in the first monthsin Canada. Two variables related tohousehold composition are also included:the number of members of the immigratingunit5 the respondent belongs to, and itscomposition (single adult, two or moreadults with children, two or more adultswithout children, one adult with children,or children alone).

Furthermore, the social networks alreadydeveloped by the immigrant and in placewhen they arrive are likely to provideassistance in finding a home, in particularby supplying temporary lodging, or moreor less complete information on the marketavailability of affordable accommodation(Moriah et al. 2004; Ray 1998; Rose andRay 2001). To take these effects intoaccount, the presence of family orfriends in Canada at the time of therespondent's arrival is also included.

Two questions relating to housing wereidentified and included in the analysis.6

The first relates to any arrangements madeprior to migration to occupy the initialdwelling. The second relates to the varioustypes of accommodation occupied bythe respondent: their own home, or thatof immediate family or in-laws; the homeof a friend; the home of a relative outsidethe immediate family; a hotel or motel;the home of an employer; temporarylodgings; an immigrant or refugee centre;or, lastly, some other type of accommodation.

The next variable contributes informationon the immigrant's region of residence:Montréal, Toronto, Vancouver or someother urban region, and indicates whetherthe processes relating to the promptnessof leaving a dwelling differ according toits urban context. To that end, terms ofinteraction designed mainly to showwhether the effect of the previouslyintroduced variables differs with the cityof residence were added to the regression.In the process of defining the finalanalysis model, interactions between theimmigration category and the socialnetwork variables, and between cities ofresidence and immigration category,knowledge of the official languages andmembership in a visible minority werecarried out at the same time. Of these

interactions, only those that appearedsignificant in the first test were retainedfor the purposes of our analysis.

Analyses of residential mobilityThe descriptive analyses of residentialmobility will focus on the respondents'first 30 weeks of settlement.7 After 30 weeks,although respondents will have begun tomove out of the picture, enough remainto produce reliable estimates.

Interurban and interprovincial mobilityMoving to another city or province ofresidence during the first months entailsa major residential adjustment that mightresult from getting a new job or fromthe desire to be closer to relatives or toconcentrations of members of the sameethnocultural group (Citizenship andImmigration Canada 2000 and 2001).The results obtained indicate that whenthis type of mobility occurs, it takessome time to set in motion—usuallyafter 20 weeks of residence—and affectsa very small proportion of newcomers.After 30 weeks, only 4 per cent ofimmigrants will have moved to anothercity, and the proportion drops to 1.5 per cent for those changing province.



3 With regard to the unit of analysis and the phenomenon being studied, the residential dynamics and the resulting choices involve the entire household. However, the unitof analysis for the LSIC was the longitudinal respondent. While it would have been appropriate to use immigrant-dependent households as our unit of analysis, this was notpossible given the design of the survey, which was intended to assess the individual experiences of immigrants. For more information, refer to the user's guide.

5 Since total income received in Canada and from outside Canada by the economic unit supported by the longitudinal respondents was not available until the interviewswere held, it was divided by the number of months the respondent had spent in Canada in order to obtain an approximation of average monthly income. Thus, income isassumed to have been stable through the first months of settlement, which is likely not the case. It would have been helpful to have more accurate data on family incomelevels: start of gainful employment, and increases or decreases in the income of both spouses.

5 In the LSIC, household characteristics are available only as at the time of arrival. In order to ensure recognition of the time factor in the sequence of events, data on therespondent's immigrating unit were given priority for the period considered, that is, between the time of arrival and the first interview. “'Immigrating unit' means the 'groupof people who applied to come to Canada under the same visa form and, for the purpose of the survey, who arrived either with the longitudinal respondent or three monthsbefore or after the longitudinal respondent'” (Statistics Canada 2003: p. 9). However, it is not necessarily all the individuals belonging to the immigrating unit, or only thoseindividuals, who will be living with the respondent and forming the respondent's household. In some cases, the number of individuals in an immigrant household may beunderestimated, and in other cases overestimated. Nevertheless, this is the best estimate available in the survey for the relevant period of settlement.

6 With no detailed information on all the initial dwellings occupied by the respondents, given the structure of the survey, the analysis could not include rental costs, the rateof effort or the tenure (owner or tenant). These factors could have proved to be key determinants of residential stability.

7 Note, however, that regression analyses are done for the entire period during which respondents were observed. In the first wave, observation ended after 10 months' residence.

Promptness of moving out by rank of dwellingResidential mobility, however, is certainlynot a matter only of changing the city orprovince of residence. It may be thoughtthat the first type of mobility experiencedby the new arrival is mobility betweendwellings.8 Our focus will be the rate atwhich newcomers move, particularly withrespect to the first dwelling occupiedupon arrival in Canada (ranked first).

We prepared a survival curve and tablefor moves out of the first three dwellingsoccupied by newcomers since their arrival(see figure 1). They indicate more specificallythe proportion of individuals who are stillin their first, second, third or subsequentdwelling (ranked 1, 2 and 3) over time.Immigrants leave their initial dwelling soonand at a relatively constant rate in thefirst months of settlement. After 30 weeksin Canada, nearly 50 per cent of respondentswill have left their initial dwelling. Theyare slower to leave the second dwellingthan they were to leave the first, andeven slower to leave the third than theywere to leave the first two, which isconsistent with the idea that overall, theresidential situation improves with eachmove. Of immigrants at risk of leavingtheir second dwelling, just over 15 per centwill have done so after 30 weeks' residencethere, and this proportion will be onlyslightly smaller with respect to leavingthe third dwelling. In other words, theresidential trajectories of a substantialproportion of new immigrants show nosign of residential stabilization in thefirst weeks of settlement. The regression

analyses of departure from the firstdwelling presented below provides aprofile of the most mobile individuals atthe very beginning of settlement.

Promptness of departurefrom the initial dwellingIn order to determine whether thepromptness with which immigrants leavetheir first dwelling differs with the placeof residence, an additional survival curveand table were prepared (see figure 2).They indicate that in the first 20 weeks,respondents seem to leave their firstdwelling at the same pace, regardless ofwhether they live in Montréal, Toronto,Vancouver or another MRR or CAR, whichsuggests that generally, the first residentialadjustments reflect the same types ofexperience in the early stages of settlement,regardless of variations from city to cityin the residential market or the profile ofthe newcomers. Thereafter, a slight gapopens up between Montréal and othercities, with immigrants in Montréaltending to leave their dwelling lesspromptly than immigrants elsewhere inCanada. However, this initial trend shouldbe explored further, using regression analysis.

We move now to the results of the Coxregressions, which yielded betterdocumentation of the processes surroundingthe move out of the initial dwelling. Sevenmodels were constructed, in a hierarchicalarrangement. The first includes one setof variables, and in each subsequentmodel a new set of variables was added.In Table 1, for each model, we presentthe coefficients for each variable, their

significance and a set of data on theregression, particularly the P 2 or likelihoodratio (LR), which indicates whether themodel is significant, and the P 2 thatdetermines the contribution of eachsignificant variable to the model. Thelast column of the table is associatedwith the seventh model. It presents theP 2 for each variable, showing the impactof each of them on the promptness ofleaving the first dwelling.

The results indicate that the introductionof each set of variables in the regressionmodels in turn adds further explanatoryforce to assist our understanding ofresidential transitions. The various modelsshow some measure of stability as theadditions are made: a majority of variablesand categories of variables retain theirsignificance and effect. The seventh andlast model best explains the promptness ofleaving the first dwelling (LR P 2=15785.51,sig=0.001). The introduction of the termsof interaction further improves theexplanation (P 2=21.44, sig=0.001), butto a lesser extent than the addition of theother variables.

Table 1 also shows which variables explainin significant terms the moves out of thefirst dwelling: the age of the respondent;the category of immigration; membershipin a visible minority; knowledge of French;level of education; previous residence inCanada; average monthly family income;composition of the immigrating unit;presence of family in Canada; and typeof initial dwelling in Canada. The othervariables have no significant effect onhow soon the respondent moves out ofthe first dwelling.



8 In the various analyses conducted, all the immigrants in the survey are likely to undergo a mobility experience in the initial months of settlement, whether mobility isexpressed in its more general form (a straightforward move) or the more specific form (such as a change of city or province). With regard to inter-city or inter-provincemobility, transitions may or may not take place at the time of the move from the initial dwelling. In some cases, the change of MRR or province of residence occurs uponleaving the first dwelling; in others, it occurs when they leave subsequent dwellings (second, third or more). Thus, in the analyses of the move from the initial dwelling,some immigrants who went through this first change of residence may have moved beyond the boundaries of their MRR of initial residence. However, the percentage ofrespondents in this situation is very small.

The P 2 of the variables indicates the extentto which each variable explains thetransition. The type of dwelling seems tobe the variable that helps most to explainit (P 2=2638.07, sig=0.001). Next inorder of importance are the variablesrelating to composition of the immigratingunit, membership in a visible minority,category of immigration, previous residencein Canada, and level of education. Thecontribution of the other variables is less important.

It is crucial to remember that in thelatter model, the addition of the terms ofinteraction prevents individualconsideration of the variables or categoriesof variables. They have to be consideredin relation to the terms of interactionassociated with them, since the coefficientsof the variables lack the effect associatedwith the terms of interaction. Thus, thecategories “family” and “West Asian” cannotbe interpreted in isolation. To obtain theoverall effect of these categories, theymust be related to the coefficients of theterms of interaction: “other MRR orCAR / family” and “West Asian / Montréal.”

Detailed examination of the seventhmodel shows the effect of the significantvariables and categories of variable.Socio-demographic characteristics helpexplain the phenomenon in terms of agevariables, a few groups within the cohortbelonging to visible minorities, and

particularly the family immigration category.With regard to age, the older the respondent,the slower the pace of leaving the firstdwelling, which is consistent with theresults of earlier studies in Quebec (Renaudet al. 1993; Renaud and Gingras 1998)and which suggests that residential stabilityis more important when a newcomer hasalready achieved a certain age level uponarrival in the new country. However,membership in a visible minority acceleratesdeparture from the first dwelling, bycomparison with the “White” referencegroup: this is the case with West Asiansliving outside the Montréal MRR, andwith Koreans.

Respondents coming to Canada for reasonsof family reunification generally experiencea significant reduction in the likelihoodof moving out of their initial dwellingduring our observation period, bycomparison with economic immigrants.The slower departure rate could beassociated with the very definition of thisimmigration category. These respondentsseem more likely than economic immigrantsto have accommodation arranged beforearrival, since they are joining a familymember already settled in Canada, whoin most cases will have been able to makesuitable arrangements to receive them.Living in accommodation more appropriateto their family situation, these respondentsseem less likely to move from their initial

residence. Also, in some cases, the initialdwelling may not be satisfactory, but thetransition is slowed by the difficulty offinding affordable accommodation thatis sufficiently large, particularly in themajor urban areas. The fact thatrespondents living outside the Montréal,Toronto and Vancouver MRRs leavetheir initial dwelling sooner could alsorelate to market conditions that are morefavourable to family households.

The set of variables relating to humancapital shows that knowledge of Frenchand a level of educational attainmentabove secondary contribute to an earlymove out of an immigrant's initialdwelling. These moves could be promptedby occupational factors. These attributesalso favour the ability to access informationon the residential market and on thehousing system without reliance oninformal networks, which means thatpeople are better informed about thepossibilities of improving their residentialsituation (Moriah et al. 2004; Rose andRay 2001). On the other hand, previousresidence in Canada reduces the likelihoodof a transition. Having lived in Canadabefore immigrating could give immigrantsmore familiarity with the housingmarket and with Canadian institutionsand how they work, and the possibilityof having developed a network withinthe city that enables them to findsuitable housing from the outset.





Model

Variables 1 2 3 4 5 6 7 PP 2

Socio-demographic characteristics

Age 0.996** 0.997* 0.994*** 0.994*** 0.995** 0.995** 0.995** 9.07**

Sex (female) 1.027 1.020 1.024 1.018 0.995 0.998 0.998 0.00

Category of immigration (economic) 28.52***

Family 0.339*** 0.358*** 0.443*** 0.478*** 0.743*** 0.743*** 0.701*** 27.40***

Refugee 0.985 1.027 1.099 1.130* 1.104 1.081 1.031 0.15

Visible minority (White) 40.23***

Chinese 1.020 1.015 0.960 0.946 0.979 0.996 1.004 0.01

South Asian 1.171*** 1.099* 1.031 1.063 0.924 0.946 0.956 0.78

Black 1.087 1.093 1.043 1.067 0.875 0.872 0.881 2.80

Filipino 1.132 1.013 0.972 1.027 0.883 0.895 0.906 1.75

Latin American 1.150 1.198 1.206 1.212* 1.045 1.060 1.171 2.10

Southeast Asian 1.065 1.076 0.911 0.929 0.730 0.736 0.741 2.94

Arab 1.190** 1.178* 1.133 1.163* 1.109 1.139 1.128 2.96

West Asian 1.486*** 1.483*** 1.390*** 1.428*** 1.113 1.138 1.255** 7.42**

Korean 1.292*** 1.344*** 1.155 1.160 1.227** 1.265** 1.277** 8.84**

Japanese 0.402* 0.558 0.494 0.477 0.605 0.617 0.622 1.58

Visible minority n.i.e 1.333 1.254 1.254 1.313 1.272 1.307 1.329 1.62

Multiple visible minorities 1.131 1.067 0.940 1.005 0.893 0.821 0.832 0.51

Whites and visible minorities 1.564 1.417 1.397 1.427 1.170 1.199 1.199 0.37

Human Capital

French 1.005 1.072 0.997 1.031 1.043** 1.039* 4.71*

English 1.059*** 1.001*** 1.072*** 1.028 1.027 1.026 2.47

Educational attainment (none or primary) 14.75**

Secondary 1.072 1.074 1.071 1.172 1.166 1.150 2.27

Postsecondary 1.167* 1.188 1.177 1.360*** 1.367*** 1.347** 9.38**

University 1.151** 1.199* 1.178 1.308** 1.311** 1.291** 7.36**

Previous residence in Canada (none) 0.519*** 0.563*** 0.554*** 0.738*** 0.726*** 0.731*** 20.79***

Household characteristics

Monthly family income 0.999*** 0.999*** 0.999** 0.999** 0.999** 8.08**

Size of the immigrating unit 1.004 1.007 1.020 1.014 1.016 0.73

Composition of the immigrating unit (adult only) 51.70***

2 or more adults with children 1.261*** 1.246*** 1.478*** 1.507*** 1.507*** 35.47***

2 or more adults without children 1.262*** 1.256*** 1.378*** 1.377*** 1.379*** 37.51***

One adult with children 0.890 0.895 1.146 1.152 1.150 2.52

Children only 0.399** 0.408** 0.518 0.536 0.532 4.13*

Amount brought in savings 1.000* 1.000* 1.000 1.000 1.000 0.30

Table 1 Risk ratios for the Cox Regression on the Move From the Initial Dwelling9

9 The categories in parentheses were omitted from the regression analyses.



Model

Variables 1 2 3 4 5 6 7 PP 2

Social network

Relatives in Canada (none) 0.868*** 1.130** 1.127** 1.111** 6.42*

Friends in Canada (none) 1.059 0.971 0.979 0.980 0.33

Housing characteristics

Arrangements for initial dwelling (none) 0.943 0.941 0.940 2.27

Type of accommodation (own home or a relative's) 2638.07***

Home of a friend 5.313*** 5.300*** 5.303*** 1570.95***

Home of a distant relative 3.127*** 3.149*** 3.178*** 456.51***

Hotel or motel 10.790*** 10.884*** 10.869*** 1229.24***

Residence of an employer 2.745** 2.769** 2.787** 8.71**

Centre or temporary residence 8.954*** 8.887*** 8.942*** 1118.51***

Other 2.997*** 3.023*** 3.022*** 125.33***

MRR or CAR of residence

City of residence (Toronto) 3.20

Montréal 0.931 0.967 0.25

Vancouver 1.035 1.019 0.15

Other MRR or CAR 1.084* 1.068 2.31

Terms of interaction

Vancouver/French 1.080 2.18

Other MRR or CAR/Family 1.258* 6.28*

Relative/Refugee 1.159 3.04

West Asian/Montréal 0.922** 9.78**

Latin American/Other 0.954 3.07

West Asian/Other MRR or CAR 0.981 2.34

* p<0.05 n 11956 11956 10959 10959 10875 10720 10720

** p<0.01 Event 5608 5608 5240 5240 5163 5029 5029

*** p<0.001 Cases covered 6348 6348 5719 5719 5712 5691 5691

-2LL 99747.844 99572.608 91946.124 91920.512 87383.318 85166.026 85144.584

dl 17 23 30 32 39 42 48

LR P 2 1182.25*** 1357.49*** 8983.97*** 9009.58*** 13546.07*** 15764.07*** 15785.51***

Contribution of each set of variables (P 2) 175.23*** 7626.48*** 25.61*** 4537.19*** 2217.29*** 21.44***

Table 1 Risk ratios for the Cox Regression on the Move From the Initial Dwelling (Con’t)

Monthly family income and thecomposition of the immigrating unit arethe only significant variables in consideringthe characteristics of the household. Bothmay relate to different groups of individuals,since the first relates to the family, whereasthe second relates to the immigratingunit—the respondent and all thoseindividuals who arrived in Canada withhim or her. In terms of income, the resultsindicate that the higher it is, the lower thechances that the transition will take place;families that are better-off are more likely tofind satisfactory accommodation uponarrival. In terms of the composition ofthe immigrating unit, the categories thatappear significant are households of twoadults with or without children. Both ofthose types of immigrating units are morelikely to move than those made up of asingle adult. The desire to acquire a moreprivate and stable residential situationboth for children and for couples beginningtheir lives in a new country could helpmake the first residential adjustment occursooner, particularly in cases where thefirst dwelling does not belong to them.

Among the variables relating to the socialnetwork, having family in Canada at thetime of arrival is the only significant one,and it increases the likelihood of leavingthe initial dwelling. It may be that familyin Canada is a help in quickly findinghousing that appeals, if there are relativesalready settled in the city where therespondent arrives, but such an interpretationmust be made with caution, because in theLSIC, this variable includes both familymembers living in the same MRR orCAR, and those living in another province.

With respect to housing characteristics,only the type of housing has a significanteffect. Thus, immigrants who live in housingthat is not their own tend to changeresidences more quickly than those livingin their own home or in the home of aclose relative. These results were more orless what we might have expected.Beginning life in Canada in housing thatis not one's own, with the sharing andcrowding that this can imply, may increaseimmigrants' desire to find a place of theirown or to live with a close relative, therebyenjoying a residential quality of life thatoffers more privacy, comfort and stability.

Lastly, with regard to the urban areavariable, none of the categories is significant.The model shows that a respondentliving in Vancouver, Montréal or anotherMRR or CAR does not move outsignificantly sooner or later than oneliving in Toronto. This result accordinglyled us to reconsider our initial observationsconcerning the slight gap between Montréaland other urban areas. Only certain specialcases of interaction between the place ofresidence of the respondents and anotherindividual attribute made it possible toidentify a few specific effects associated withthe geographical location of newcomers.Immigrants in the family reunificationclass living outside the three main MRRsare quicker to leave their initial dwelling,whereas West Asians living in Montréalshow more residential stability; this couldbe attributable to the difficulties ineconomic integration experienced by thelatter (Godin 2004) and the supposedlynegative effect on their ability to findand move into more satisfactory housing.

ConclusionThe objective of this study was to explorethe potential of a new source oflongitudinal data in documenting theresidential mobility of new immigrantsin the first months of settlement in Canada,whether the mobility takes the form of amove out of their metropolitan area orprovince of residence or, less drastically, achange of housing. Detailed examinationof the promptness of moving out of theinitial dwelling made it possible not onlyto identify the factors associated with greaterresidential mobility, or on the contrarygreater residential stability, but also to seewhether the associated processes differwith the immigrant's place of residence.We hoped thus to make up for the lackof studies comparing the major cities ofimmigration in Canada and to gain abetter understanding, through theenhancement of longitudinal data, of thedynamics of residential settlement. Theresults are somewhat surprising.

While immigrants are unlikely to moveto another city or province in the firstmonths of settlement, we cannot saythat they are not very mobile in the earlydays. Many of them—about 50 per cent, infact—moved at least once, either within thesame city, or otherwise. However, contraryto what one might have expected, the rateat which newcomers leave their initialdwelling does not generally seem to beaffected by variations in residentialmarkets from one large city to another,or between the three large metropolisesand the other urban areas. The promptnessof moving out of the initial dwelling isaffected rather by the characteristics of



the housing and the household, bynewcomers' individual characteristics andattributes in terms of human capital, andby social networks. In particular, the typeof housing occupied, householdcomposition, membership in a visibleminority and educational attainment arethe determinant variables in acceleratingtransitions, whereas the category ofimmigration, previous residence in Canada,age and income have a significant effect ondelaying the move from the initial dwelling.

While the dynamics of the move fromthe first dwelling vary little from city tocity over the brief period of settlementconsidered, that is, with respect to thefirst wave of observations from the LSIC,things could be different in the mediumterm, when immigrants may be moreexposed to the dynamics of residentialmarkets, which differ considerably fromcity to city. In conducting this study, wesought to use the newly available datafrom the first wave of the LSIC to shedlight on the residential aspects of the firststeps in settling in a new country. The periodcovered by the analysis does not, however,correspond for all respondents to thetime spent in the same accommodation.Data from subsequent waves of the LSICwill support analyses of this type covering alonger period of observation, particularlyas it corresponds to a period of inflationin the housing market in certain largecities (Canada Mortgage and HousingCorporation 2004). It will be possible tosee whether the mechanisms observedare maintained, whether some arepeculiar to the beginning of settlement,or whether other dynamics emerge aftera certain time has passed.

Lastly, with regard to method, the studyshowed the added value of the longitudinalapproach in gaining a better understandingof the housing dynamics that newcomersexperience, even though we did encountersome limitations of the LSIC, particularly

the impossibility of determining possibleassociations between the timing of residentialtransitions and certain variables with amajor explanatory potential, such as rentalcosts, the degree of effort and the modeof occupancy. That said, the otherCanada-wide longitudinal surveys of theimmigrant cohort (such as the LongitudinalImmigration Database and the NationalPopulation Health Survey) containmuch less information on housing andresidential mobility, and in this sense theLSIC represents substantial progress,particularly as housing is a crucial fulcrumfor making a fresh start in a new country.

AcknowledgmentsThis research was conducted as part ofthe Metropolis Project, and funded in partby Canada Mortgage and HousingCorporation and by the NationalHomelessness Initiative. For the valuableassistance we received from the CentreInter-universitaire québécois de statistiquessociales (www.ciqss.umontreal.ca), wewant to thank the Centre's staff, andmore particularly, Isabelle Pronovost ofStatistics Canada. Our thanks go also toBenoit Laplante of INRS-UCS.



00.

05

2.0

05.

05

7.0

00.

1t

nev

ee

htd

ec

neir

ep

xe

gniv

ah

to

n

noitr

op

orP

0 10 20 30Duration (weeks )

rank = 1 rank = 2

rank = 3

Kaplan -Meier survival estimates , by rank0

0.0

52.

00

5.0

57.

00

0.1

tn

eve

eht

de

cn

eire

px

eg

niva

ht

on

n

oitro

por

P


rank = 1 rank = 2

rank = 3

Kaplan -Meier survival estimates , by rank

Rank 1 Rank 2 Rank 3

n : 11 968 5 599 640 Events : 5 504 572 41 Censored cases : 6 464 5 027 599

Figure 1 : Survival curve for the timing of exit from the dwelling, by rank

00.

05

2.0

05.

05

7.0

00.

1

tn

eve

eht

de

cn

eire

px

eg

niva

ht

on

n

oitro

por

P


city = Montr éal city = Toronto

city = Vancouver city = Other CMA or CA

Kaplan -Meier survival estimates , by city

00.

05

2.0

05.

05

7.0

00.

1

tn

eve

eht

de

cn

eire

px

eg

niva

ht

on

n

oitro

por

P


city = Montr éal city = Toronto


city = Montréal city = Toronto

city = Vancouver city = Other CMA or CA

Kaplan -Meier survival estimates , by city

n: 11 968 Events: 5 504 Censored cases: 6 464

tn

eve

eht

de

cn

eire

px

eg

niva

ht

on

n

oitro

por

P

Figure 2 : Survival curve for the timing of exit from the dwelling, by city of residence

ReferencesAllison, P.D. 1991. Event History Analysis:Regression for Longitudinal Event Data,Series: Quantitative Applications in theSocial Sciences, No. 46. Beverly Hills: Sage.

Bernèche, F. and J.-C.Martin. 1986.L'implantation géographique des ménagesd'origine haïtienne dans la région montréalaise :étude des regroupements spatiaux de population.Première partie. Montréal: Université deMontréal, Centre de recherches caraïbes,Rapports de recherche, N˚ 17.

Bolt, G. and R. van Kempen. 2002.Moving Up or Moving Down? HousingCareers of Turks and Moroccans inUtrecht, the Netherlands. HousingStudies 17(3): 401-422.

Canada Mortgage and HousingCorporation. 2004 Canadian HousingObserver, 2004. Ottawa.

Canadian Council on Refugees. 1998.Best Settlement Practices: SettlementServices for Refugees and Immigrants inCanada. Montréal. Available athttp://www. web.ca/ccr/bpfina1.htm

Citizenship and Immigration Canada. 2000.The Inter-provincial Migration of Immigrantsto Canada. IMDB Profile Series. Ottawa.Catalogue No. MP22-18/ 4-2000E.

Citizenship and Immigration Canada.2001. Towards a More Balanced GeographicDistribution of Immigrants. Ottawa.

Cleves, M.A., W.W. Gould and R.G.Gutierrez. 2002. An Introduction toSurvival Analysis Using Stata. CollegeStation, TX: Stata Press.

Godin, J.-F. 2004. L'insertion en emploides travailleurs admis au Québec en vertude la grille de sélection de 1996. Partie 1 :Rapport synthèse. Québec : Gouvernementdu Québec, Ministère des Relations avecles citoyens et de l'Immigration, Directionde la population et de la recherche.

Miraftab. F. 2000. Sheltering Refugees:The Housing Experiences of Refugees inMetropolitan Vancouver. CanadianJournal of Urban Research 9(1): 42-63.

Moriah, A., L. Rodriguez and L. Sotomayor.2004. Building Housing Through SocialNetworks: New Colombian Immigrants inToronto. Presentation at the InternationalConference “Adequate and AffordableHousing for All,” International SociologicalAssociation, Research Committee 43 onHousing and the Built Environment, Toronto,June 24-28; electronic copy available athttp://www.urbancentre.utoronto.ca/pdfs/housing conference/Moriah_Rodriguez_Sotomayor_.pdf

Murdie, R. 2002. The Housing Careersof Polish and Somali Newcomers inToronto's Rental Market. HousingStudies 17(3): 423-443

Murdie, R.A., A.S. Chambon, J.D.Hulchanski and C. Teixeira. 1999.Differential Incorporation and HousingTrajectories of Recent Immigrant Households:Towards a Conceptual Framework. DiscussionPaper. University of Toronto: Housing NewCanadians Research Working Group;electronic copy available at http:// www.library.utoronto.ca/hnc/publish/concept.pdf

Özüerkren, A.S. and R. van Kempen. 2002.Housing Careers of Minority EthnicGroups: Experiences, Explanations andProspects. Housing Studies 17(3): 365-379.

Ray, B. 1998. A Comparative Study ofImmigrant Housing, Neighbourhoods andSocial Networks in Toronto and Montréal.Report produced under the ExternalResearch Program. Ottawa: CanadaMortgage and Housing Corporation.

Ray, B. 1999. Plural Geographies inCanadian Cities: Interpreting ImmigrantResidential Spaces in Toronto and Montréal.Canadian Journal of Regional Science /Revue canadienne des sciences régionalesXXII (1/2): 65-86.

Renaud, J. and L. Gingras with theparticipation of A. Carpentier, G.Pinsonneau and M. Faille. 1998. Les troispremières années au Québec des requérantsdu statut de réfugié régularisés. Montréal :Gouvernement du Québec, Ministère desRelations avec les citoyens et del'Immigration, Series “Études, rechercheset statistiques,” N˚ 2.

Renaud, J., S. Desrosiers and A. Carpentier.1993. Trois années d'établissementd'immigrants admis au Québec en 1989.Portraits d'un processus. Montréal :Gouvernement du Québec, Ministère desCommunautés culturelles et del'Immigration, Direction des études et de larecherche, Series “Études et recherches,” N˚ 5.

Rose, D. and B. Ray. 2001. Le logementdes réfugiés à Montréal trois ans après leurarrivée : le cas des demandeurs d'asileayant obtenu la résidence permanente /The Housing Situation of Refugees inMontréal Three Years After Arrival: TheCase of Asylum Seekers Who ObtainedPermanent Residence. Journal ofInternational Migration and Integration /Revue de l'intégration et de l'immigrationinternationale 2(4): 455-492 (Frenchversion); 493-528 (English version).

Ryan, L. and J. Woodill. 2000. A Searchfor Home: Refugee Voices in the RomeroHouse Community. Study requested bythe City of Toronto. Toronto: MaytreeFoundation. Available at http://ceris.metropolis.net/Virtual%20Library/housing_neighbourhoods/ryan1/ryan1.html

Statistics Canada. 2003. LongitudinalSurvey of Immigrants to Canada: Wave 1,Microdata User Guide. Ottawa: StatisticsCanada, Special Surveys Division.

Statistics Canada. 2005. LongitudinalSurvey of Immigrants to Canada. APortrait of Early Settlement Experiences.Ottawa: Statistics Canada. CatalogueNo. 89-614-XIE.



The major characteristics of theLongitudinal Survey of Immigrants inAustralia, LSIA, and its major findingsconcerning the housing of immigrantsare described to situate the LSIC, theLongitudinal Survey of Immigrants inCanada, in terms of its methodology andfindings regarding housing.

There have been three panels of theLongitudinal Survey of Immigrants inAustralia (Black et al. 2003). The first panelconcentrated on the initial and early periodsof settlement, interviewing immigrantsafter approximately six, eighteen, andforty-two months of residence in Australia(Cobb-Clark 2001). Five thousand onehundred and ninety-two permanentsettlers who arrived between September1, 1993 and August 31, 1995 and settledin state and territory capital cities, centresclose to capital cities, and Cairns wereinterviewed in person. The sample wasstratified by visa category and by countryof birth. The second panel is a smallersample of 3,000 immigrants who arrivedin Australia between September 1, 1999and August 31, 2000. In the second panel,the sample emphasized highly skilledimmigrants. Conducted over a period of12 months, the second panel includesonly two interviews at six months andtwelve months of residence in Australia.A third panel survey was conducted veryrecently, in this case relying upon a

questionnaire sent through the postrather than in-person interviews.

The design of the LSIA differs from theLSIC in three respects. The LSIA targetedmigrants in urban centres that were hometo 95 per cent of immigrants, but only71 per cent of the population. Migrantsliving outside urban centres were excludedfrom the survey. The LSIC is conductedin 15 languages, whereas the LSIA isconducted in English using flash cards in10 languages and interpreters. Finally,the timing of the cohorts in each surveydiffers slightly. Interviews will be conductedat six, eighteen, and thirty-six months ofresidence in Canada, slightly differentfrom the timing of interviews in LSIA.

With regards to the housing conditionsof immigrants and refugees, the contentof the two surveys is very similar. Thereare several questions about the cost,affordability, quality, and suitability ofhousing in each cohort. A search forliterature drawing on the housing questionsin the LSIA unearthed few reports andacademic articles. Most published analysesof the LSIA concern the economictrajectories of immigrants (Cobb-Clark2001, Black et al. 2003). The housingsituations of immigrants are discussedbriefly in descriptive reports summarizingthe trends from each section of the survey.Andrew Beer and Sarah Morphett (2002)have completed the most sophisticated

analysis of immigrants' housingcircumstances.

With two panels, the LSIA allows fordescriptions of immigrants' housing aftersix, twelve, and forty-two monthsresidence to track how immigrants andrefugees enter the housing market.Comparative analyses of the first andsecond panels enable researchers toevaluate the impact of the economic andsocial context on immigrants' housingconditions. Comparison of the two panelsof the LSIA illustrates the value ofrepeating longitudinal surveys (Richardson,Miller-Lewis, Ngo, and Ilsey 2002b).The second panel respondents are bettereducated and more likely to speak English.They also arrived in better economic times.Nevertheless, the housing situations inthe first six months are still remarkablysimilar when the first and second panelsare compared.

The housing situations of immigrants inthe first six months of settlement are similarin Canada and Australia. Upon arrival, 9of 10 immigrants in Australia lived inshared accommodations with relatives orfriends (Richardson, Miller-Lewis, Ngo,and Ilsley 2002b). Within six months,the majority of immigrants are living inpermanent accommodation. Only 14 per cent and 9 per cent ofimmigrants and refugees in panels 1 and2 respectively remain in other housing


Appendix 2

Immigrants andHousing in Australia

arrangements such as boarding housesand short-term accommodation sixmonths after arrival. Most immigrantsand refugees move into the private rentalsector. At six months, 51 per cent and54 per cent of settlers in panels 1 and 2were private renters. Approximately onein five households were still living rent-free with family and friends.Homeownership is a minority tenure,even for the highly skilled immigrants inpanel 2. Only 11 per cent of panel 1arrivals and 16 per cent of panel 2 arrivalsare homeowners within six months.Humanitarian settlers, refugees inCanada, are the main occupants ofsocially-assisted housing that is home toonly 4 per cent of all recent arrivals aftersix months of residence.

Changes in housing type often occurwithin six months of arrival (Richardson,Miller-Lewis, Ngo, and Ilsley 2002). Bysix months, approximately half of allimmigrants and refugees are living insingle-detached housing, the predominanttype of housing in Australian cities whereimmigrants and refugees concentrate.Another 13 per cent are living in semi-detached housing, with slightly more thanone third, 36 per cent, living in apartments,evenly divided between low-rise 1 and 2-storey garden apartments and 3- or morestorey high-rise apartments. The distributionof immigrants and refugees among housingtypes does not vary much over time(VandenHeuvel and Wooden n.d.). In eachcohort, humanitarian and independentclass immigrants are most likely to livein apartments, while business class andemployer nominated immigrants are mostlikely to live in single-detached housing.

As expected, housing costs increased overthe period of settlement and between the

two panels of the LSIA. Immigrants inthe skilled stream reported the largestincreases in housing costs while humanitarianclass immigrants who were mainly ingovernment housing had the smallestincreases in housing costs (VandenHeuveland Wooden n.d.). Due to inflation,average housing costs were 20 per centhigher by the second panel than for thefirst. As a result the second panel is payingmore to rent and own housing than earlierarrivals. Higher costs have not intensifiedhousing stress. Only 18 per cent of thesecond panel were spending more than40 per cent of gross income on rent ormortgage payments. Better economicoutcomes for the second panel havereduced the financial burden of higherhousing costs (Richardson, Miller-Lewis,Ngo, and Ilsley 2002a).

In panel 2, the majority of immigrants andrefugees lived in good quality housing(Richardson, Miller-Lewis, Ngo, andIlsley 2002a). Only 6 per cent judgedtheir housing to be of poor quality.However, the quality of housing variesacross immigration classes. Humanitarianimmigrants are more likely to report thattheir housing is of poor or moderatequality than any other class. At the otherextreme, more than 70 per cent ofbusiness and employer nominatedimmigrants live in good quality housing.

Within the first 42 months of settlement,the majority of immigrants and refugeesare renters even though the percentage ofimmigrants and refugees purchasing ahouse increases steadily over time. In thefirst panel, 11 per cent of immigrantshad bought a dwelling within the firstsix months of arrival. The percentage ofhomeowners rises to 15 per cent by 18months and then to 38 per cent within

42 months. Although the finalhomeownership rate is only slightly morethan half the national rate of 68 per cent,the steady rise in homeownership suggeststhat some immigrants and refugees haveprogressive housing careers. However,the concentration of immigrants andrefugees in rental accommodation after42 months indicates that the majority ofnewcomers are still establishingthemselves in the housing market. Beerand Morphett (2002) propose a usefulconceptual framework that distinguishesthree stages of housing consumption:

� Initial settlement - The first six monthsof settlement in which immigrantsmove from temporary and sharedaccommodation to permanent housing.

� Adjustment - The period stretchingfrom six to twenty-four monthsduring which there are substantialand varied changes in tenure.

� Consolidation - After two years of residence, there is a concertedmove towards homeownership formany immigrants.

According to this framework, manyimmigrants and refugees are still in theconsolidation phase after 42 months ofsettlement. Additional research is neededto document over a longer time span thehousing careers of different classes ofmigrants. Humanitarian migrants appearat greatest risk of being stuck in governmenthousing. However, the progress ofhighly-skilled immigrants also warrantsinvestigation. In the first three years ofsettlement, family is an importantresource. Preferential family immigrantsmove into homeownership more quicklythan some other classes of migrantslargely because of family support in theform of rent-free accommodation.



References for Appendix 2Beer, Andrew, and Morphett, Sarah (2002)The Housing and Other Service Needs ofRecently Arrived Immigrants. Melbourne:Southern Research Centre, AustralianHousing and Urban Research Institute.

Black, R., Fielding, T., King, R., Skeldon,R., and Tiemoko, R. (2003) LongitudinalStudies: An Insight into Current Studiesand the Social and Economic Outcomesfor Migrants. Sussex Migration WorkingPaper No. 14.

Cobb-Clark, Deborah (2001) 'TheLongitudinal Survey of Immigrants toAustralia,' The Australian EconomicReview, 34: 467-477.

Richardson, Sue, Miller-Lewis, L., Ngo,Phong, and Ilsley, Diana (2002a) Life ina New Land: The Experience of Migrantsin Wave 1 of LSIA 1 and LSIA 2.Adelaide: National Institute of LabourStudies, Flinders University.

Richardson, Sue, Miller-Lewis, Lauren,Ngo, Phong, And Ilsley, Diana (2002b)The Settlement Experiences of NewMigrants, A Comparison of Wave One ofLSIA 1 and LSIA 2. Report Prepared forthe Department of Immigrant andMulticultural and Indigenous Affairs,National Institute of Labour Studies,Flinders University, Adelaide, Australia.

VandenHeuvel, Adriana and Wooden,Mark (n.d.) New Settlers Have Their Say- How Immigrants Fare Over the EarlyYears of Settlement.http://www.dimia.gov.au/research/publications/overview/newset1.htm



Visit our home page at www.cmhc.ca

6531

9

08/03/07

The Housing Situation and Needs of Recent Immigrants in the Montréal,Toronto, and Vancouver CMAs:...

Documents

Transcript of The Housing Situation and Needs of Recent Immigrants in the Montréal,Toronto, and Vancouver CMAs:...