Commonwealth Rent Assistance and the spatial concentration of low income households in metropolitan Australia

Transcription

1 Commonwealth Rent Assistance and the spatial concentration of low income households in metropolitan Australia authored by Bill Randolph and Darren Holloway for the Australian Housing and Urban Research Institute UNSW-UWS Research Centre June 2007 AHURI Final Report No. 101 ISSN: ISBN:

2 ACKNOWLEDGEMENTS This material was produced with funding from the Australian Government and the Australian States and Territories. AHURI Ltd gratefully acknowledges the financial and other support it has received from the Australian, State and Territory governments, without which this work would not have been possible. AHURI comprises a network of fourteen universities clustered into seven Research Centres across Australia. Research Centre contributions, both financial and in-kind, have made the completion of this report possible. DISCLAIMER AHURI Ltd is an independent, non-political body which has supported this project as part of its programme of research into housing and urban development, which it hopes will be of value to policy-makers, researchers, industry and communities. The opinions in this publication reflect the views of the authors and do not necessarily reflect those of AHURI Ltd, its Board or its funding organisations. No responsibility is accepted by AHURI Ltd or its Board or its funders for the accuracy or omission of any statement, opinion, advice or information in this publication. AHURI FINAL REPORT SERIES AHURI Final Reports is a refereed series presenting the results of original research to a diverse readership of policy makers, researchers and practitioners. PEER REVIEW STATEMENT An objective assessment of all reports published in the AHURI Final Report Series by carefully selected experts in the field ensures that material of the highest quality is published. The AHURI Final Report Series employs a double-blind peer review of the full Final Report where anonymity is strictly observed between authors and referees. i

3 CONTENTS CONTENTS...II EXECUTIVE SUMMARY...1 Introduction...1 Methodology...2 Spatial concentration of private rental low income households...3 Socio-economic characteristics of private rental low income households...3 Changes in the spatial concentration of private rental low income households...3 The location of low cost housing stock...4 Commonwealth Rent Assistance in Public housing waiting lists in The relationship between public housing waiting lists, Commonwealth Rent Assistance, low cost rental stock and private renter low income households in The changing location of Commonwealth Rent Assistance recipients in The changing relative distribution of Commonwealth Rent Assistance recipients in The labour market profile of privately renter low income working households INTRODUCTION BACKGROUND Introduction Economic changes Demographic and social changes Commonwealth Rent Assistance Housing and employment Disadvantage in the private rental market Housing policy issues METHODOLOGY Introduction Imputing missing household income figures from Census data Other Data sources WHAT IS A LOW INCOME HOUSEHOLD? Introduction Our preferred approach Comparing the populations of this project to the Yates et al (2003) study Examining the changes in PRLIHs over time LOW INCOME HOUSEHOLDS IN THE PRIVATE RENTAL MARKET Introduction The relative position of PRLIH in the housing market Location of low income households in the private rental market Concentrations at the LGA/SLA level Concentrations at the suburb and postcode level Profile of low income households in the private rental market Dwelling type Rent paid...42 ii

4 5.4.3 Household type Age Country of birth Education Occupation Industry Income CHANGES IN THE DISTRIBUTION OF LOW INCOME PRIVATE RENTERS IN Introduction Changes in the number and location of private renter low income households, Summary AN EXAMINATION OF THE LOW COST PRIVATE RENTAL STOCK Introduction An analysis of very low cost private rental stock LGA level An analysis of low cost private rental stock LGA level Summary An analysis of very low cost private rental stock postcode level An analysis of low cost private rental stock postcode level THE RELATIONSHIP BETWEEN LOW INCOME PRIVATE RENTERS AND HOUSING ASSISTANCE Commonwealth Rent Assistance in Public housing waiting lists in The relationship between public housing waiting lists, Commonwealth Rent Assistance, low cost rental stock and PRLIHs in The changing location of Commonwealth Rent Assistance recipients in Absolute changes in the distribution of Commonwealth Rent Assistance recipients in Relative changes in the distribution of Commonwealth Rent Assistance recipients in THE LABOUR MARKET PROFILE OF LOW INCOME WORKING PRIVATE RENTERS Introduction Occupation and industry of PRLIWHs in six case study areas Occupation Industry Summary Journey to work of private renter low income working households in six case study areas Summary Journey to work of private renter low income working households by main workplace destinations Summary CONCLUSION REFERENCES APPENDIX 1: NUMBER AND PROPORTION OF PRLIHS BY LGA iii

5 Sydney Melbourne Adelaide APPENDIX 2: THE NUMBER AND PROPORTION OF VERY LOW COST AND LOW COST PRIVATE RENTAL STOCK FOR POSTCODES IN SYDNEY, MELBOURNE AND ADELAIDE IN Occupation and industry for PRLIWHs in Canterbury in Occupation and industry for PRLIWHs in Wyong in Occupation and industry for PRLIWHs in Darebin in Occupation and industry for PRLIWHs in Moreland in Occupation and industry for PRLIWHs in West Torrens East in Occupation and industry for PRLIWHs in West Torrens West in APPENDIX 4: THE ORIGIN OF PRLIWHS WHO ARE EMPLOYED IN THE AREAS WHICH EMPLOY A SIGNIFICANT NUMBER OF PRLIWHS, Sydney Melbourne Adelaide iv

6 EXECUTIVE SUMMARY Introduction The private rental sector in Australia has been undergoing considerable change in recent years. Concerns have been raised about the lack of attention given to the private rental sector by policy makers, as recent government policies have concentrated on promoting home ownership or, to a lesser extent, on dealing with structural failures of the public housing stock (Beer, 1999; Yates and Wulff, 2000). Yet recent research indicates that landlords are withdrawing from the lower end of the private rental market. The process of inner city gentrification has removed large numbers of lower rental stock from its traditional strongholds (Forster, 2004; Randolph, 2004). While urban consolidation policies have led to a resurgence of the higher value rental investment sector in CBD locations, there has been a countervailing shift outwards to the middle and outer suburbs in the location of lower value rental stock. Concerns have also been raised that lower income households are being concentrated into the private rental sector for longer periods of time and in certain locations (Wulff, et al, 2001). Further, this may also act to constrain the labour market opportunities for these households. During the 1960s and 1970s the expansion of public housing provided relatively easy access to affordable housing, especially for those on low incomes. At the same time, the rapid expansion of home ownership and government policies supporting home owners allowed many to enter the home purchase market. Private rental was seen as the least preferred tenure as households increasingly moved into home ownership or public housing. However, during the 1980s and 1990s economic, demographic, social and policy changes contributed to the changing role of the private rental market in Australia. In particular, there was a move away from direct public housing provision to providing rent subsidies - Commonwealth Rent Assistance (CRA) - which is paid directly to low income individuals. The private rental sector came to be seen as a preferred solution to the provision of low cost housing for those with low incomes, which involved a lower subsidy per dwelling, at least in the short term. However, as with public housing, CRA to private tenants is highly targeted on those in receipt of benefits and/or pensions. As a result, the bulk of this subsidy is directed to households reliant on welfare payments for their income. Apart from some family households, by and large, low paid working households do not receive CRA. As this group do not qualify for public housing nor are likely to be able to afford home ownership, given historically high property prices, they have been largely ignored by policy makers in the on-going policy debates on housing policy. In effect, the working poor in Australia are the forgotten sector in housing policy, falling between CRA, public housing and home ownership policy. By and large, they fend for themselves in the private rental market assisted in policy terms only by the poorly understood impacts of negative gearing on investment in lower value housing. With the impending shift toward more flexible and contingent employment practices, together with welfare to work polices determined to ensure those who are able to work do so, the role of the private rental sector in supporting the low paid workforce will become even more significant. This report focuses on the position of low income households including those in employment (i.e. the working poor ) in the private rental market to better understand the outcomes of the current policy framework on the housing opportunities open to them. Despite their low policy visibility, this group are important. While there are many households who still use the private rental market as a stepping stone to home purchase, often as part of the new Generation X/Y young professional diaspora who are spending much of their 20s and 30s in more mobile lifestyles who rent in inner city locations before settling for home purchase once they can afford to do so, there is evidence that suggests an increasingly significant proportion of low income households in the private rental market are residing in this sector over long time periods (Wulff and Maher, 1998). Furthermore, research has also raised concerns that low income households in the private rental market are experiencing affordability problems, raising concerns over the operation of Commonwealth Rent Assistance policies. More recent AHURI research has identified the low paid service sector as the primary job sector for whom housing affordability problems are most profound (Yates, Randolph and Holloway, 2006). Unaffordable rental housing severely inhibits these 1

7 households ability to enter the home purchase market, and as a result of the operation of the market, they are being concentrated in certain areas of cities and are being constrained in their access to labour market opportunities, especially those in the central city (Randolph and Holloway, 2005). Many of those who work in these households are employed in low paid jobs that are nevertheless important to the functioning of the city economy. This project explores the ramifications of these processes through a detailed analysis of spatial trends in the provision of low income private rental dwellings and the location of low income working households who rent privately in Adelaide, Sydney and Melbourne. It examines the contention that lower income working renters are becoming more concentrated in certain areas of these cities. The report analyses the labour market position of low income working private renters, including their workplace locations, shedding light on the issue of whether affordable housing is needed to provide key sectors of the low income workforce with appropriate housing in areas near where they work. Finally, the research explores the spatial relationship between the distribution of low income households who rent privately, the provision of Commonwealth Rent Assistance (hereafter Commonwealth Rent Assistance or CRA) and the location of households on public housing waiting lists in each of the three cities. This analysis will inform policy making with regard to the spatial outcomes of housing policy, especially in relation to the changing structure of the lower income labour markets on which these households rely. Methodology Currently there is no precise definition as to what constitutes a low income household. Part of this project addresses the housing and workplace relationships of low income households with at least one person in work, or the working poor. That is, an analysis focuses on low income households as well as those low income households in the workforce. The income cut-off points are therefore designed to encapsulate a sufficient sample for both of these groups. The income threshold chosen for the research was the bottom quintile, or 20 th percentile, of household income for households with at least one person employed. Because income levels vary around the country, the income threshold was varied to reflect the income distribution in each of the three cities examined in this project Sydney, Melbourne and Adelaide. As the research uses the 2001 Census for the analysis, the upper income thresholds at this time used in this study are: $ for Australia as a whole; $ in Sydney; $ in Melbourne; $ in Adelaide. Most of the data used in this study was collated from a value added unit record file created by the Australian Bureau of Statistics (ABS) from the 2001 Census of Population and Housing for the study by Yates et al (2003). The main advantage of using this file was that each individual, and therefore household, was assigned with an actual income figure (not an income category). Further, all individuals who did not state their income were assigned an income figure. From this methodology we were able to analyse the social, spatial and labour market characteristics of 931,655 employed low income households in Australia. In Sydney, there were 117,545 low income households in the private rental market of which 69,873 had at least one person in work (60%). In Melbourne, there were 92,982 low income households of which 51,354 included at least one worker (55%). In Adelaide, there were 32,559 low income households of which 15,562 (48%) included at least one worker. Data from Centrelink on the number of Commonwealth Rent Assistance recipients and the average amount paid by post-code were also used in the study. Data for the years 2000 to 2005 (i.e. including the years following the property boom) were used. Data on public housing waiting lists by geographic area are also used to compare with spatial locations of private renter low income households. 2

8 Spatial concentration of private rental low income households In Sydney, there were particularly high concentrations of private rental low income households (PRLIHs) in Canterbury and Wyong local government areas (LGAs), and in the suburbs of Liverpool, Campsie and Cabramatta. In Melbourne there were high concentrations of PRLIHs in Darebin and Moreland local government areas and in the suburbs of Frankston, Reservoir and Dandenong. In Adelaide there was a particularly high concentration of PRLIHs in West Torrens East statistical local area (SLA), as well as higher concentrations in West Torrens West SLA and Holdfast Bay-North. At the suburb level in Adelaide there were high concentrations of PRLIHs in Morphett Vale, Salisbury and Prospect. Socio-economic characteristics of private rental low income households An analysis of a selected range of socio-economic variables from the 2001 census provided a profile of the social characteristics of households in the sector and how they differed from the population as a whole. A key finding is that 53% of PRLIHs in Sydney, 42% in Melbourne and 36% of PRLIHs in Adelaide live in flats. In fact, twice the proportion of PRLIHs in Sydney live in flats compared to the rest of the population, while the figure is approximately three times in Melbourne and Adelaide. Between 60% and 70% of PRLIHs in Sydney, Melbourne and Adelaide are lone person households and one parent families. This is clearly at odds with the general population, where the equivalent figure is approximately a third. The importance of this housing submarket for single adult and single parent (i.e. single income) low income households reflects their basic lack of market capacity in the home ownership sector. The age profile of PRLIHs in all three cities is younger than for the population as a whole. In all three cities the proportion of persons aged under 34 is significantly higher than that for the population as a whole. Conversely, the proportion of persons aged over 55 in PRLIHs in Sydney, Melbourne and Adelaide is significantly lower than that for the population as a whole. In Adelaide the proportion of individuals in PRLIHs born in Australia is similar to that for the population as a whole. However in Sydney and Melbourne the proportion of persons in PRLIHs born in Australia is less than that for the population as a whole. That is, there is an over-representation of overseas born persons in PRLIHs in Sydney and Melbourne. Further, despite the low incomes of households in PRLIHs in Sydney, Melbourne and Adelaide, the educational profile is similar to that for the population as a whole. PRLIHs, not surprisingly, pay less rent than other renters. Some 71% of PRLIHs in Sydney, 83% in Melbourne and 82% in Adelaide pay between $100 and $250 per week. Individuals in PRLIHs tend to be employed in lower skilled occupations and are over-represented in Intermediate and Elementary Clerical, Sales and Service jobs as well as Intermediate Production and Transport Workers. Conversely, individuals in PRLIHs are significantly under-represented at the top end of the occupational hierarchy, particularly in Managerial and Professional occupations. Interestingly, the industry profile of individuals in PRLIHs in Sydney, Melbourne and Adelaide is not greatly different from the population as a whole, indicating that this sector of the housing market supplies a low income workforce to a wide range of industries and businesses in these cities. Changes in the spatial concentration of private rental low income households Between 1991 and 2001 there was a 25% increase in the number of PRLIHs in Sydney, 14% in Melbourne and 21% in Adelaide. Generally speaking, there was a movement of PRLIHs away from inner city areas in the three cities (although Melbourne LGA itself is a notable exception) between 1991 and 2001 towards more suburban locations. Nevertheless, the areas with the highest number of PRLIHs in 1991 tended to also have the largest numbers in 2001 (for example, Canterbury in Sydney, Darebin and Moreland in Melbourne, and West Torrens in Adelaide). However, there was an interesting development in Adelaide over the period. Areas around Playford in northern Adelaide and Onkaparinga in the south 3

9 recorded a large increase in the number of PRLIHs between 1991 and 2001 from very low bases. The location of low cost housing stock There are clear concentrations of low cost private rental housing stock in certain areas in the three case study cities. In Sydney the low cost private rental stock is concentrated in the west, south west and Central Coast to the north. In Melbourne the low cost private rental stock is concentrated in the south-east, west and inner north, while in Adelaide the low cost stock is concentrated in the north and western parts of the city. In other words, the inner city areas in each of the three cities lack low cost private rental stock, although there appears to be more low cost stock in the inner suburbs of Adelaide and Melbourne than in Sydney. In Sydney there are significant concentrations of low cost stock in seven LGAs where over 75% of the private rental stock is classified as low cost. This compares with only one LGA in Melbourne and two SLAs in Adelaide. Commonwealth Rent Assistance in 2001 In 2001 there were just over 54,000 Commonwealth Rent Assistance recipients in Adelaide, 155,000 in Melbourne and 189,000 in Sydney. In Sydney, 57 postcode areas had over 1,000 recipients, compared to 45 in Melbourne and just 7 in Adelaide. In Sydney, two postcode areas, 2166 (Cabramatta) and 2170 (Liverpool) had over 5,000 recipients each (5,859 and 5,388 respectively). In Melbourne the largest number of Commonwealth Rent Assistance recipients were recorded in postcode 3175 (Dandenong) and 3021 (Albanvale/Kings Park/St Albans), which had 4,520 and 3,324 recipients respectively. In Adelaide the largest number of Commonwealth Rent Assistance recipients were recorded in postcode 5108 (Salisbury) and 5162 (Morphett Vale), which had 2,190 and 1,599 recipients respectively. Public housing waiting lists in 2001 The largest number of public housing waiting list applicants were, generally, in the same areas as those where larger numbers of Commonwealth Rent Assistance recipients were recorded. In 2001 in Adelaide, the highest number of applicants on the public housing waiting list was located in and around Salisbury in the north and Onkaparinga in the south. In Melbourne the largest concentrations of public housing waiting list applicants was found around Brimbank, Darebin, Dandenong and the Mornington Peninsula. In Sydney the situation is generally the same. High numbers of public housing waiting list applicants can be found in the areas in the south-west and western parts of the city where there are also high numbers of Commonwealth Rent Assistance recipients. Interestingly though, despite large numbers of Commonwealth Rent Assistance recipients in Gosford-Wyong there are relatively few applicants from these areas on the public housing waiting list. This is likely to reflect the lower proportion of public housing in this region and the lower rents in the private rental sector, thereby acting as a deterrent for potential applicants. The relationship between public housing waiting lists, Commonwealth Rent Assistance, low cost rental stock and private renter low income households in 2001 A correlation analysis at the postcode level in Sydney, Melbourne and Adelaide in 2001 reveals a very high correlation between the numbers of applicants on public housing waiting lists, number of Commonwealth Rent Assistance recipients and the location of low cost private rental stock and PRLIHs. In Sydney, Melbourne and Adelaide the correlation coefficient between PRLIHs and Commonwealth Rent Assistance recipients is greater than 0.9. The correlation between the number of PRLIHs and the number of public housing applicants is 0.9 in Adelaide and 0.8 in Sydney and Melbourne. The correlation between the number of Commonwealth Rent Assistance recipients and the number of public housing applicants is 0.9 in Adelaide and Melbourne and 0.8 in Sydney. Finally, the correlation between the location of the low cost private rental stock and the number of PRLIHs is greater than 0.9 in the three cities. This is also similar to the correlation coefficients between the low 4

10 cost private rental stock and the number of Commonwealth Rent Assistance recipients. The correlation between the number of public housing applicants and the location of the low cost private rental stock is also significant at approximately 0.8 for the three cities. The implication of this analysis is that the locations of Commonwealth Rent Assistance recipients, waiting list applicants and the low income, low value rental housing sub-market are remarkably closely associated in each city. However, while some of these households are supported by Commonwealth Rent Assistance and others may eventually benefit from housing assistance through public housing, there are others, mainly those in work, who will receive little or no assistance in their housing needs. The changing location of Commonwealth Rent Assistance recipients in A separate analysis of the changes in Commonwealth Rent Assistance between 2000 and 2005 showed that the areas with the largest number of Commonwealth Rent Assistance recipients in 2000 were generally the same areas with the highest number in 2005, although the order in which they appear has changed. In Sydney the five postcode areas with the largest number of recipients in 2000 were the same in 2005, although 2170 (Liverpool) replaced 2166 (Cabramatta) as the postcode area with the largest number of recipients. The postcode areas with the largest increase were in Sydney s west and south-west, while the gentrifying inner city, the urban fringe and Cabramatta recorded the largest decreases. In Melbourne, the four postcode areas with the largest number of Commonwealth Rent Assistance recipients in 2000 had the largest concentrations in They were also in the same order, with postcode 3175 (Dandenong) leading the way. Between 2000 and 2005 there was an increase in the number of recipients on the western and south-eastern fringes of Melbourne, while there were decreases in the gentrifying inner city and inner east. In Adelaide the three postcode areas with the largest number of Commonwealth Rent Assistance recipients in 2000 were also the largest in 2005, and in the same order, with 5108 (Salisbury) having the largest number of any postcode area. Between 2000 and 2005 there was a decrease in the number of Commonwealth Rent Assistance recipients in parts of inner Adelaide and the inner north-western coast, while there were increases in some of the more disadvantaged suburbs in the north and Morphett Vale in the south. Interestingly, between 2000 and 2005 Adelaide was the only city of the three case studies that had an increase in the total number of Commonwealth Rent Assistance recipients. The changing relative distribution of Commonwealth Rent Assistance recipients in The research involved examining the relative change in the proportion of Commonwealth Rent Assistance recipients between 2000 and To do this, the number of Commonwealth Rent Assistance recipients in a postcode area was expressed as a proportion of the metropolitan wide total in both years. The change between the two years was then calculated. This method allows the changing number of Commonwealth Rent Assistance recipients in each city over time to be accounted for. The analysis here confirms that, in general, there has been a decrease in the proportion of Commonwealth Rent Assistance recipients in the inner cities of Sydney, Melbourne and Adelaide, and in some outer areas on the fringes of the three cities, especially in coastal locations. Conversely, there was an increase in the relative proportion of Commonwealth Rent Assistance recipients in certain middle suburban areas in each of the three cities. Over this five year period, therefore, there was a notable shift of Commonwealth Rent Assistance recipients into middle and outer suburban areas. 5

11 The labour market profile of privately renter low income working households The labour market characteristics of privately renting low income working 1 households (PRLIWHs) in the two areas in each city with the highest number of this group were analysed. The six case study areas were Canterbury and Wyong LGAs in Sydney, Darebin and Moreland LGAs in Melbourne and West Torrens (East) and West Torrens (West) SLAs in Adelaide. Generally speaking the individuals in these case study areas are employed in the middle to lower end of the occupational hierarchy. Individuals are more likely to be employed as Intermediate Clerical, Sales and Service Workers, Intermediate Production and Transport Workers and Labourers and Related Workers. Conversely, individuals in PRLIWHs are less likely to be employed as Managers and Administrators, Professional Workers or Advanced Clerical and Service Workers. However, this varies slightly between case study areas. For example, in the West Torrens case study areas, compared to the areas in Sydney and Melbourne, there was an over-representation of Managers and Administrators. In terms of their industry of employment, employed individuals in PRLIWHs in our case study areas were more likely to be employed in Retail Trade industries, Manufacturing, Accommodation, Cafes and Restaurants, and Property and Business Services. In Melbourne and Adelaide there were also larger numbers of individuals in PRLIWHs employed in Health and Community Service industries. The workplace locations of workers in PRLIWHs varied with each location, but generally there was a marked pattern of the most important employment locations either within the local LGA or in the respective CBDs of each city. In Canterbury LGA in Sydney and Darebin and Moreland LGAs in Melbourne the largest proportion of employed persons in PRLIWHs worked within the LGA, with the next largest proportion working in the respective CBDs. In the Adelaide case study areas of West Torrens (East and West) the largest proportion of employed PRLIWHs worked in the Adelaide CBD, followed by the SLA itself. However, in Wyong LGA a much higher proportion of employed persons in PRLIWHs worked in the LGA itself as well as the nearby LGA of Gosford. Few individuals in Wyong worked in the Sydney CBD, a clear reflection of the commuting distance and costs involved. This has significant implications for local employment prospects for this group of low income workers in the Central Coast LGAs. Finally, where do individuals in PRLIWHs work in each city? Despite the outward movement of the sector, the research shows the strong attraction of the core employment areas for such workers. The main workplace destinations of PRLIWHs were City of Sydney LGA and Parramatta LGA in Sydney, City of Melbourne LGA and Port Phillip LGA in Melbourne, and City of Adelaide SLA and West Torrens-East SLA in Adelaide. In Sydney LGA, Melbourne LGA and Adelaide SLA there were a number of prominent nearby areas from which PRLIWHs commuted. But for Parramatta LGA, Port Phillip LGA and West Torrens-East SLA, the majority of PRLIWHs who worked in these areas also resided in these areas. 1 i.e. households where at least one person was working 6

12 1 INTRODUCTION In recent years there has been increasing concern about the role of the private rental market within the Australian housing system (Yates and Wulff, 200). Research has indicated that landlords are withdrawing from the lower end of the private rental market, and that investment returns and the effectiveness of negative gearing are lower in this sector of the housing market (Wood, 2001). Inner city gentrification has also removed large numbers of lower rental stock, pushing the investment market for this housing into the middle and outer suburbs. As such, concerns have been raised that lower income households who are reliant on private renting are being concentrated for longer periods in this market and in certain locations in the city. Further, this may also be acting to constrain the labour market opportunities for these households. This project explores the ramifications of these processes through a detailed analysis of spatial trends in the provision of low income private rental housing and low income households who rent privately at the local level in Adelaide, Sydney and Melbourne. In particular, it explores the extent to which lower income rental markets are becoming more concentrated or more dispersed in these cities. The proposed research will also contribute to the debate on the effect of the private rental market in constraining access to labour market opportunities for lower income working households. It explores this through an analysis of the labour market position of low income private renters, including their workplace locations, helping to address the issue of whether affordable housing is needed to provide key sectors of the low income workforce with appropriate housing in areas near where they work. Finally, the research explores the spatial relationship between the distribution of low income households who rent privately, the location of the lower cost rental stock, the location of recipients of Commonwealth Rent Assistance (RA) and the location of households on public housing waiting lists, to better understand the connection between these elements of housing assistance for those in the lower end of the housing market. It also analyses whether the location of RA recipients has changed over the recent past during a period of intense house price inflation. In particular, the project had the following aims: To measure the concentration of low income households in the private rented sector in Adelaide, Sydney and Melbourne at the local scale, and changes in the concentration or dispersal of this housing market sector in these cities over the decade in relation to changes in the overall distribution of the rental market. To explore the extent to which lower income private renters are concentrated in certain localities in terms of the housing market segment they occupy, defined in terms of their property characteristics and rental levels. To establish whether the spatial concentration of Commonwealth Rent Assistance recipients and households on the public housing waiting list is more or less concentrated than the spatial distribution of lower income private renters who are in employment. To establish the labour market position of those lower income households in the private rental sector who are in employment, and the relationship between their residential location and place of work in the three cities. These aims will be addressed through a series of five interlinked Research Questions focusing on lower income households in the private rental sector:- To what extent is the location of lower income households in the private rental sector concentrated within Adelaide, Melbourne and Sydney, and has the level of concentration increased or reduced over the last decade? Have any changes been in line with the overall shift in the location of the private rental market in these cities during this time? Are lower income households disproportionately concentrated in certain segments of the private rental market in each city? To what extent do concentrations of private rental lower income households reflect the distribution of lower cost private rental opportunities across each city? Are economically 7

13 active lower income private renters more or less concentrated in suburbs where this housing market sector as a whole is concentrated? What is the relationship between lower income households in the private rental market and forms of housing assistance? Are households in receipt of Commonwealth Rent Assistance distributed across the city in the same way as those lower income private renters in employment? What is the spatial relationship between lower income private rental households and public housing waiting lists in these cities? What is the labour market profile of lower income private rental households, at the citywide level with the highest concentrations of this group? Where do employed lower income private renters work and what is the relationship between their residential location and their workplace location? Are there any indications that lower income private renters are effectively excluded from job-rich locations as a result of their residential location, or do the majority of employed lower income private renters work relatively locally? The chapters that follow provide the core empirical findings of the research. Chapter 2 first sets the context for the research by reviewing the evidence of the current changes affecting the private rental market in Australian cities in recent years. Chapter 3 outlines the methodology used in the research, in particular, the derivation of the value-added Census data file used. Chapter 4 focuses on how the study defined low income for the purposes of this study. Chapter 5 addresses the first research question, namely are low income working households renting privately concentrated in part of the city and in addition profiles their characteristics. Chapter 6 goes on to establish whether the location of the sector has changed in recent years. Chapter 7 identifies the location of the low cost rental stock in the three cities analysed and chapter 8 analyses the relationship between the low cost rental stock, waiting list applicants, Commonwealth Rent Assistance recipients and the low paid renter sector. Chapter 9 explores the labour market profile of the sector, its relationship to employment concentrations and its journey to work characteristics in six case study local government areas. Chapter 10 offers some final observations from the findings in relation to the overarching research questions. 8

14 2 BACKGROUND 2.1 Introduction At the end of World War II, detached dwellings on individual plots of land dominated Australia s housing stock. The great Australian dream of owning a separate house had widespread support among the Australian population. This sentiment was promoted strongly by government policies that encouraged a culture of home-ownership. The era of the mass homeownership and the quarter acre lot had arrived. During this period, private rental accommodation was seen as a tenure of transition (Kendig 1981). That is, young households moved into private rental accommodation while saving to purchase their own dwelling. Moreover, at present, owner occupied housing still constitutes the majority of net household wealth in Australia (Yates 2000). It was not until the late 1970s and 1980s that academics began to investigate the nature of the private rental market in Australia (Paris 1985). Australia s home-ownership rate grew rapidly in the post-war period reaching its high point of 72 per cent in 1966 (Beer 1999). But in 1945, private rental housing was the single largest tenure in Australia, accounting for 45 per cent of households (Beer 1993). However, within 20 years this had fallen by 50 per cent. This was due to a number of factors. Landlords sold their properties to tenants as rent controls introduced in 1941 made rental properties an unprofitable investment (Kendig 1979, Beer 1999). There was a rapid expansion of home ownership through mortgage funded debt, combined with favourable government policies for home owners, and this allowed many households to enter the home purchase market and to pay higher prices for housing than investors were willing or able to do. In effect, the expansion of home ownership recapitalised the private rental stock, encouraging its sale to homeowners (Hamnett and Randolph, 1987). In addition, the expansion of public housing during the 1960s and 1970s supported relatively easy access to these dwellings, especially to working households on low incomes, thereby removing much of the demand for private rental accommodation for this group (Paris 1993, Hayward 1996). However, during the late 1970s and 1980s a number of factors contributed to an increasing re-investment in the private rental market. In particular, demographic, economic and social changes in the last thirty years in Australia have contributed to the changing nature of the supply of private rental accommodation and the demand for this kind of housing. 2.2 Economic changes Economic restructuring during the late 1970s witnessed a parallel increase in unemployment in Australia, which placed greater demand on the public housing system in Australia. A number of commentators have contended that during this period public housing became welfare housing (Paris et al 1995). Although the Federal government increased the number of public housing dwellings in Australia during the 1980s and early 1990s to nearly 400,000 dwellings, there were still 210,000 households on the waiting list (Industry Commission 1993, Hayward 1996). Restructuring also meant deregulation of the financial markets in the mid 1980s, which saw the rate of increase in house prices outstrip the growth in wages and salaries (Yates 1987). Deregulation resulted in interest rates climbing to 17 and 18 per cent by the late 1980s and early 1990s. Many first time home buyers were confronted with high housing prices and high interest rates which made home ownership relatively unaffordable during the late 1980s. Further, the changing nature of the labour market during this period also influenced the home ownership market. During the early 1990s the shift towards a more flexible economy, through the increased use of sub-contractors, the subsequent growth of self-employment and part-time workers (particularly amongst women), the increase in dual income households, and the decline in manufacturing employment in Australia made the entry into home ownership more difficult (Beer 1999, Beer and Forster 1996). In particular, the consequences of such shifts made securing a loan more difficult for those on low or fluctuating incomes. 9

15 Yates (2003) found that despite considerable tax changes since the late 1980s the indirect assistance provided through the tax system continues to provide assistance to those households who need it least. On average, outright owners receive more than five times the amount of assistance provided to purchasers, and high income outright owners receive a total tax benefit of approximately $9,000 per household per year (compared to the average of $4,200 for all outright owners). 2.3 Demographic and social changes In conjunction with the economic shifts in Australia, a number of demographic and social factors have also influenced the changing nature of the housing system in Australia. In particular, the decreasing size of households and the increase in one and two persons households have substantially impacted on the housing system in Australia. For example, Wulff (1997) and Wulff and Maher (1998) found that households are experiencing extended periods of time in the private rental market. These longer term renters are more likely to live alone, their economic position tends to be poorer (as a higher proportion receive pensions and benefits), and they are more likely to be divorced. At its simplest level, lone person households only have one income to rely on and therefore are much less able to compete in the housing market against two or multi-income households. The widening gap between housing market capacity of single persons and other households is therefore compounded, especially for those in low paid employment or those not in the workforce. The increasing separation and divorce rate in Australia has, consequently, increased the number of sole-parent and lone person households in Australia. These households increasingly rely on the private rental sector for accommodation, at least temporarily, as they re-establish their housing position after family break-up. Wulff et al (2001) note that between 1986 and 1996 half of the increased demand for lower priced private rented dwellings came from single person households. By 1996, single person households (lone person and single parent households) constituted 42 per cent of all private renters, up from 35 per cent in The increasing trend over the last thirty years for couples to have children later in life is also impacting on the private rental market. For whatever reasons, these households also chose to be in the private rental market. Wulff and Maher (1998), for example, found that never married persons aged constituted the largest group of continual long-term renters. 2.4 Commonwealth Rent Assistance In more recent times there has also been a shift of government policy away from public housing provision to rent subsidies. Commonwealth Rent Assistance (RA) is a major form of direct housing assistance paid to individuals on low incomes who rent in the private market. RA is paid directly to the individual with the total amount paid being a proportion of the rent. The amount paid therefore varies by rental level up to a threshold. It also varies by household type. A number of concerns, however, have been raised about the effectiveness of RA. Firstly, RA is capped and therefore its value is eroded in areas where housing costs are high (Brotherhood of St Laurence 2003). Berry and Hall (2001) argue that RA has failed to sufficiently alleviate housing stress for lower income households. Their study concluded that singles and couples without children who are eligible to receive RA could not afford to rent a dwelling in either Sydney or Melbourne. Further, Burke et al (2002) contend that RA offers limited assistance to students in high rent housing markets, where many tertiary institutions are located. RA is also only effective if there is a sufficient supply of affordable dwellings in the private rental market. Unfortunately, affordable housing stocks have been decreasing in recent years (Yates and Wulff 2000). Wood (2001) contends that the taxation system, through negative gearing provisions and capital gains exemptions, has contributed to this trend by encouraging investors to purchase high value properties. 10

16 2.5 Housing and employment Social and economic change over the last few decades has impacted on the polarisation of household income in Australia (Gregory and Hunter 1995, Raskall 2002) and to increase social and spatial inequalities (Yates 2002a). In general, lower income households may be forced to locate in areas with limited employment opportunities because of the limited housing options they face. For example, Winter and Stone (1998) concluded that less skilled and lower paid workers were being marginalised by being forced into those categories at the bottom of the tenure hierarchy (public housing and those who are constrained to the private rental market). In Australia, the size of the private rental market means that the bulk of low income households who rent, rent privately. In 1999 the Commonwealth Government appointed a reference group to consult with the community and provide advice to the Government on welfare reform. In its final report, the McClure Report (2000) also expressed a concern that low cost housing in the most disadvantaged regions may tend to attract unemployed and underemployed people, adding locational disadvantage to their existing barriers in accessing employment. O Connor and Healy (2002) found strong and stable geographic links between housing markets and labour markets within the Melbourne metropolitan area at the regional level (p. vi). In their study of Melbourne they concluded that policies introduced in Melbourne during the 1990s had increased the spatial separation of specific labour market groups. They argued that the Core area is job and skills rich but housing expensive. However, the fringe areas of Melbourne do not have the highest concentrations of labour market disadvantage. Rather these areas are concentrated in the middle suburban areas of Melbourne. O Connor and Healy therefore suggest that housing policy cannot be expressed without a better understanding of job-housing links. That is, knowledge of the number, type and location of jobs needs to be considered in planning policies associated with new suburban developments. Further, Stilwell (1989) also concludes that economic polarisation in Sydney intensified during the 1980s due to structural economic change, and that the changing conditions in labour markets and housing markets interact to reinforce spatial inequality. That is, as house prices increase, the existence of house price inequalities reinforces the spatial segregation in that the losers from economic structural changes are forced into localities where house prices remain just within reach. As a consequence, Stilwell suggests, like O Connor and Healy (2002), the viability of development schemes should be called into question unless employment opportunities for the growing population can be established. Randolph and Holloway (2004 and 2005) have also confirmed evidence for the increasing concentration of lower income households in middle distance metropolitan suburbs in housing markets associated with lower value rental. In both Sydney and Melbourne it is clear that areas of older housing between the gentrifying inner cities and the new middle to upper income fringe suburbs are a growing characteristic of our largest cities (see also Dodson 2003 and Reynolds and Wulff 2005 for Melbourne). Finally, National Shelter and ACOSS (2003) concluded in their examination of Commonwealth Rent Assistance that Commonwealth Rent Assistance is currently not effective, on its own, in encouraging and/or allowing recipients to locate to areas with better job prospects (p. 16). In particular, the study found a tendency for high numbers of Commonwealth Rent Assistance recipients to live in areas with lower rents and above average unemployment levels, and this was more geographically pronounced in NSW and Queensland. 2.6 Disadvantage in the private rental market Due to some of the economic, demographic and social changes mentioned above, as well as some of the policy shifts of recent years, it can be argued that large sections of the private rental market are associated with some of the most disadvantaged of all groups in the Australian housing system. For example, Flood (1993) found that in outright owners received net subsidies of $2,000 per household per year, purchasers received $1,000 in 11

17 subsidies, public tenants $3,020 in subsidies per year, while private tenants only received $994 in subsidies per year (cited in Beer 1999). As early as 1985, Rossiter and Vipond found that private tenants, on average, spent 51 per cent of their income on accommodation costs, compared to 34 per cent for purchasers, 23 per cent for public tenants, and only 7 per cent for outright owners. In 1989, Wilson concluded from her survey of RA recipients that nearly three-quarters of them paid more than 30 per cent of their income in accommodation costs. The National Housing Strategy (1991) also concluded that while less than 10 per cent of all income units were in housing stress in 1998, the predominant group in stress were in the private rental market. Private renters comprised 20 per cent of all income units, but of all those units experiencing housing stress, nearly 60 per cent were private renters. Between 1986 and 1996, owners became richer while renters became poorer (Yates 2002b). In her report on the housing outcomes arising from social, spatial and structural change, Yates concluded that between 1986 and 1996 the average weekly income of private renters decreased by 6 percentage points to $714 per week, whilst average weekly income for home purchasers increased by 3 percentage points to $1,036 per week. More recently, Randolph and Holloway (2002) contended that in 1999, 155,689 lower income households 2 were in housing stress in Sydney. Of these, 106,713 (69 per cent) were private renters. This represented 64 per cent of all lower income households who rent privately, and 32 per cent of all private renters in Sydney. Importantly, the Randolph and Holloway (2002) study also found that while lower income private renter households in stress were geographically widespread across the Sydney metropolitan area, there were clear local concentrations of lower income renter households experiencing housing stress. The spatial concentrations of lower income private renters in stress were not only confined to high value inner city areas, but were numerically more concentrated in middle suburban areas of western Sydney. This is similar to the findings of O Connor and Healy (2002) in Melbourne. Following up on their 2002 work in Sydney, Randolph and Holloway (2005) continued to explore these relationships through an examination of the relationship between housing tenure and disadvantage in Sydney and Melbourne. They concluded that socio-economic disadvantage in both cities was not tenure specific. In fact only a quarter of households in Census Collectors Districts (CDs) in Sydney and Melbourne classified as severely disadvantaged lived in public housing. Of those CDs in severe disadvantage with low and moderate levels of public housing, around one-third in Sydney and one-fifth in Melbourne were privately rented. Importantly, Randolph and Holloway (2005) concluded that between 1996 and 2001 the local government areas in Sydney and Melbourne that had an increase in the number of CDs in severe disadvantage were characterised by very low levels of public housing. That is, the shift in disadvantage is happening in the private housing market. As public housing waiting lists remain stubbornly high, low income households who do not qualify for an increasingly socially targeted social housing stock and are priced out of the increasingly unaffordable homeownership market have little alternative but to turn to the private rental market. However, little is known about the spatial concentrations of such households in Australia or of its current characteristics (Gleeson and Randolph 2002; Mee 2002). As we noted above, government housing policies are not well targeted on this group at the present time. If policies are to be developed that better address the needs of low paid workers in the private rental market then we need a better understanding of the housing market position of these households, both in terms of the market segmentation characteristics and their locational characteristics. This research project addresses this gap in our knowledge of this large group of low income working private renters. 2 Lower income households are households with below median household income. 12

18 2.7 Housing policy issues In their seminal piece of work on the private rental market in Australia, Wulff et al (2001) concluded that between 1986 and 1996 there was a decline in the low cost private rental stock of 28 per cent, despite an increase of 34 per cent across the entire private rental stock during this period. More recent evidence by the same authors (Wulff et al, 2005) shows that there was also a decline in low cost private rental stock from 1996 to 2001 of 0.2% despite an increase in private rental dwellings of 7.6 per cent over that period. This raises a number of housing policy issues in Australia. Debates in Australia continue over whether demand side assistance is preferred to supply side solutions. In general, the debate has centred on whether Commonwealth Rent Assistance (CRA) is preferred over public housing provision. Since 1992, the amount of CRA payments from the Commonwealth has exceeded payments made to the States for public housing under the Commonwealth-State Housing Agreement (CSHA) (Seelig 2001). Wulff et al (2001) contend, however, that the debate needs to be expanded to address the question of why there has not been a supply side response to the increasing demand for low cost housing. The underlying assumption of CRA is that lower income households can be adequately housed in the private market by supplementing their income. However, there is little consideration of whether there are enough low rent private dwellings in appropriate locations that are affordable and are of an appropriate standard. Evidence suggests that many low income households who work remain poor after meeting their rent obligations (Hulse and Burke 2000). Research by National Shelter and ACOSS (2003) and King and Melhuish (2003) has shown that Commonwealth Rent Assistance affects the level of housing affordability depending on what part of Australia the household lives in. As such both pieces of work suggest that the structure of CRA provision needs to incorporate a regional or geographical factor that takes into account the regional variations in rents by households who receive Commonwealth Rent Assistance. Further, the vast majority of investors in the private rental market in Australia are individuals, not institutional investors (Yates 1996; O Dwyer 1999; Berry 2000). Wood and Watson (1999) suggest that those individuals who invest in the lower end of the private rental market have, on average, lower incomes than rental investors overall. As such, they receive fewer tax advantages than other rental investors, and face higher operating costs. Combined, these factors add up to lower returns for these investors. Hulse and Burke (2000) also suggest that as the low cost rental stock contracts and rents increase, the nexus between labour market processes and housing problems will intensify. They further contend that programs which link the receipt of income supplements to work incentives will have limited success for many recipients unless there are parallel policies to address the underlying structural problems of the private rental market. While the proportion of private rental housing has grown in recent years, and outright owner occupation continues to grow as the population ages, the decreasing proportion of home purchasing and the stagnant proportion of public housing suggest that the private rental market will continue to be an important sector of the Australian housing system. However, as noted above, low income private renters in Australia, particularly those in employment, have been largely invisible to policy makers. Government policies aimed at promoting home ownership and the focus on maintaining existing public housing estates have done little to address the concerns of the lower income private rental market. This project aims to address the gaps in our knowledge about lower income working households in the private rental market by examining whether these households have become more spatially concentrated and how this may have affected their ability to access employment in job-rich locations. 13

19 3 METHODOLOGY 3.1 Introduction This report uses Census data as a key source to examine the spatial and labour market characteristics of low income households in the private rental market. There are a number of issues that need to be addressed at the outset of this project. Firstly, one of the concerns, particularly for this project, is that the normal Census data has a relatively high proportion of partially stated, or missing household income figures (Yates and Wulff 2001; Yates et al. 2003). In the 2001 Census, three per cent of households did not state their income, while a further eight per cent only partially stated their income. Using the data set generated by Yates et al. (2003), this study attempts to address the limitations of household income data from the Census by imputing household income figures for those households who did not, or only partially, stated their incomes at the 2001 Census. Details of this procedure are discussed below. This value added data set was used for 2001 analyses only. The time series analysis presented in chapter 6, based on 1996 Census data, uses a slightly different methodology (see section 4.4). Secondly, also following Yates et al. (2003), we define a private rental household as one that is renting from a real estate agent or a private landlord not in the same household. The final issue that needs to be addressed in this project is what is a low income household? We discuss this issue fully in chapter 4. Before this though, we briefly examine the methodology for imputing missing Census variables to the data set used for this project. 3.2 Imputing missing household income figures from Census data The process of imputing missing household income figures is the same as that used for the Yates et al. (2003) study. This study should be consulted for a more detailed description of the procedure used. Here we provide a brief overview of the process. The data imputation procedure was undertaken by the ABS prior to any other tables being produced. Missing income values were imputed from donor populations, that is, incomes were assigned from a particular sub-group of households with similar demographic characteristics and locational attributes of the households whose incomes were missing. The sub-groups or donor populations were selected on the basis of location (metropolitan and non-metropolitan within each State/Territory), age of household reference person (5 levels), household type (6 levels) 3 and employment status. Employment status was also sub-divided by LGA, sex and age. In essence, the only variable that had missing values was employment status, of which 3 per cent of cases were missing. The individuals for whom employment status was not known were assigned an employment status such that the probability of being employed was the same as that of the corresponding sub-group. These employment subgroups were based on the LGA, sex and age of those individuals whose employment status was known. In all, 180 sub-groups or sub-populations were created. These 180 sub-groups were further split into three categories. Cases where income and employment data were reported for all household members formed a donor category. Cases where income was only partially stated or completely unstated formed one recipient category. All other households formed a second recipient category. Each case or record in the recipient categories was randomly assigned a donor s household income, as long as it was at least as great as the partial income. Overall, imputation was seen as desirable because preliminary checks suggested that for household income, the cases for which data were missing, were not a random subset of all households. 3 Non-classifiable households have been excluded from the analysis 14

20 3.3 Other Data sources Data from Centrelink on the number of Commonwealth Rent Assistance recipients and the average amount of Commonwealth Rent Assistance paid by post-code were also used in the study. Data for the years 2000 to 2005 (i.e. including the years following the property boom) were used, as well as data on numbers on public housing waiting lists by local government area. 15

21 4 WHAT IS A LOW INCOME HOUSEHOLD? 4.1 Introduction The focus of this report concerns low income households in the private rental market and the labour market characteristics of these households who are in employment. But what exactly is a low income household and how might it be defined for this project? Currently there is no concise definition as to what constitutes a low income household. A number of cut-off points have been used by a number of commentators to define a low income household. Some definitions include: Households on welfare payments Households below the Henderson poverty line Households in the bottom income quintile Households with income below half of the median household income Households with incomes in the bottom 40 per cent of the income distribution range For example, Yates and Wulff (2000) used $300 in 1996 as the cut-off point for low income households in their study. That is, households earning less than $300 per week in 1996 were considered a low income household. Yates and Wulff (2000) also considered a low to moderate income household as one earning between $300 and $499 in Siminski and Saunders (2003), on the other hand, define a low income household as one whose income is in the bottom quintile (20 per cent) of the national income distribution, whilst Travers and Richardson (1993) describe the bottom quintile of incomes as being low income households or households with a low standard of living. However, Travers and Richardson (1993) also consider the subjectiveness of the matter, and point out that the bottom 10 or 20 per cent, half of the median, or the Henderson poverty line are all relatively good indicators of lowincome groups. Adding to this complexity, Bray (2001) defined a low income household as those households in the bottom 43.2 per cent of all household incomes. Marshall et al. (2003) referred to low income households as those receiving income support payments. Further, The National Housing Strategy (1991) defined those on low incomes as being in the bottom 40 per cent of the income distribution range of all income units in Australia. Commentators interested in income distribution analyses and inequality have also recently adopted the idea of equivalent income scales. The major underlying factor for such an approach is that a household with a given level of economic stress will be of greater concern if it is larger since this stress will affect more individuals (Saunders 2003). The notion of equivalent household income is derived by applying an equivalence scale to household income. This equivalence scale measures the estimated needs of a household relative to the needs of a single person living alone. The degree of income inequality is then described by a series of standard indicators, in particular the Gini coefficient. One of the limitations of the equivalent income approach is that it assumes that income and resources are shared among the household members according to their needs (Saunders, 2003). For example, the equivalence scale used by the ABS assigns a value of 1.0 to the first adult, 0.5 to every other adult aged over 15, and 0.3 to each child under 15 years of age. Obviously, one of the problems of equivalency scales is that actual income is not represented. For example, in the mean disposable income for a household was $791 per week, however, mean equivalent disposable income was $469 (Saunders, 2003). Further adding to the complexity of defining what a low income household might be is the notion of using income cut-offs varied by different household types. This might be favoured for analyses focusing on households dependent on welfare payments, for example, as there are considerable differences between such payments for different households, especially between those with or without children. It is also clear that different type and sizes of household face different costs, not least in the housing costs they need to meet. An advantage of using such a technique is that all household types would be more equally represented in the resulting sample. For example, by taking the bottom quintile of incomes 16

22 for each household type would ensure that a comparable proportion of each household type is represented in further analyses. However, this is also the downside of using such a technique. Not only would the analysis need to be presented for different household types, but it could also be debated whether some households at income thresholds above basic welfare levels actually have low incomes. For example, Tables 4.1 and 4.2 below show that, indicatively, the bottom and second quintile incomes for family households with older children is over $1,000. Anecdotally, many of these households may have at least two incomes, which pulls the averages up for these households. Even allowing for the differential living costs these households face, can all these households really be considered to have low incomes when we compare the first and second quintile cut off points for lone person households, for example? Then there is the issue that incomes vary geographically. A low income in one city may be relatively more generous in another where housing costs, for example, are lower. Housing costs vary significantly between cities, as do household income levels. If using different household types as an approach is utilised, then we would need a complex matrix of income thresholds by location to capture the complexity of the situation, as Tables 4.1. and 4.2 illustrate. For example, at the 1 st quintile, the average couple family with non-dependent children earns almost 16 per cent less in Adelaide than in Sydney. Table 4.1: Income cut-offs for households in the 1 st distribution range, 2001 quintile of the household income Sydney Melbourne Adelaide Australia Couple family with children under 15 $772 $728 $669 $676 Couple family with dependent students (15-24) $1,004 $941 $892 $892 Couple family with children under 15 and dependent students (15-24) $833 $795 $765 $772 Couple family with non-dependent children $1,024 $913 $861 $873 Couple family without children $447 $419 $397 $397 One parent family with children under 15 $312 $313 $306 $309 One parent family with dependent students (15-24) $417 $426 $394 $394 One parent family with children under 15 and dependent students (15-24) $383 $388 $397 $388 One parent family with non-dependent children $523 $497 $458 $466 Other family $470 $451 $432 $434 Total family households $569 $529 $457 $472 Lone person household $220 $220 $207 $211 Group household $649 $562 $469 $494 Total households $414 $393 $337 $360 (source: ABS customised matrix) 17

23 Table 4.2: Income cut-offs for households in the 2 nd distribution range, 2001 quintile of the household income Sydney Melbourne Adelaide Australia Couple family with children under 15 $1,119 $1,020 $921 $954 Couple family with dependent students (15-24) $1,484 $1,364 $1,280 $1,309 Couple family with children under 15 and dependent students (15-24) $1,245 $1,154 $1,079 $1,112 Couple family with non-dependent children $1,459 $1,282 $1,192 $1,241 Couple family without children $844 $715 $613 $646 One parent family with children under 15 $426 $417 $386 $395 One parent family with dependent students (15-24) $664 $646 $580 $596 One parent family with children under 15 and dependent students (15-24) $531 $528 $511 $507 One parent family with non-dependent children $810 $751 $653 $684 Other family $799 $733 $651 $665 Total family households $980 $891 $739 $791 Lone person household $314 $303 $272 $280 Group household $1,052 $895 $720 $798 Total households $779 $702 $558 $623 (source: ABS customised matrix) 4.2 Our preferred approach For this project we have decided to use the second household income quintile cut-off points for all working households for each city as defining a low income household. We have chosen this for three reasons. Firstly, this project is not interested in the labour market outcomes of low income households by selected household types, but rather employed households 4 on low incomes. Secondly, one of the emphases of this research is those employed households in the private rental market on low incomes, that is, the working poor. Therefore, the low income threshold needs to be appropriate for working households and not those dependent on welfare. Thirdly, by using a lower threshold such as the bottom income quintile of all households, we would exclude a large proportion of working households with what are, to all intents and purposes, very low incomes. These thresholds have been compared to the actual income profiles of households, renters and working households in the three case study cities. The results of this approach can be seen in Tables 4.3 to 4.5. These data have been extracted from the confidentialised unit record file (CURF) of the 2001 Census. As such the values and proportions will differ slightly from the results presented in Tables 4.1 and 4.2, as these tables are calculated using a combination of ABS Income surveys as well as the Census. However, once the incomes of those who did not state, or only partially stated, their income on Census night were imputed using the procedure outlined earlier, the ABS calculates that the exact income cut-offs for the bottom 20% (first quintile) of employed households in 2001 in each of the three cities and Australia would be: Australia $ Sydney $ Melbourne $ Adelaide $ Employed households refer to households in which at least one person is employed 18

24 Private renter Low Income Households comprise all of those households with incomes below those cut offs. Incomes are not equivalised. Later in the analysis, Private Renter Low Income Working Households are analysed. This is a subset of Private Renter Low Income Households in which at least one person is employed. 4.3 Comparing the populations of this project to the Yates et al (2003) study As the methodology used in this project is similar to that used in the Yates et al. (2003) study for AHURI, this section compares the number of households that were used in the Yates et al. study to the number of households to be used in this study. 5 Table 4.6 presents these findings. While we have used income cut-off points for each of the three cities in our study, using the Australia wide figures from this study and Yates et al (2003) will provide an example of how the emphasises of this study compare to that of the Yates et al study. Thus if we use the Australia wide low income cut-off point (less than $ per week) from the Yates et al (2003) study we find that this research analysed just under 1.13 million households. However, only 180,000 of these households had at least one person in employment. By using our income cut-off point of $ across Australia we include over 2.7 million households, but importantly, just over 930,000 are employed households. While the focus of the Yates et al study was not on the working poor, using the same cut-off point would have included only very small numbers of working poor households, making sub-group analysis by area problematic. Using the bottom quintile income for employed households has enabled us to focus on low income households who are not in work and in private rental in each city, and also allowed us to analyse the labour market characteristics of those working poor households. This is also important as analyses at spatial levels below the metropolitan area level are more robust than they would be using the low income cut-off point as that used in the Yates et al research. 4.4 Examining the changes in PRLIHs over time The value added census value used in this project and Yates et al (2003) was for In trying to examine the changes in the location of low income private renters over time another methodology had to be employed in this instance. This was because the cost of re-producing the 2001 value added file for 1991 was too expensive. Throughout this time series analysis a low income household was still defined as a household that had an income below that of the cut-off point for the bottom quintile for employed households. This cut-off point is derived for each city (Sydney, Melbourne and Adelaide). In 1991, the household income quintiles were obtained for Sydney, Melbourne and Adelaide. As the 1991 census income categories were used, the bottom quintile (or 20 th percentile) figures for the three cities were rounded up or down to coincide with the nearest income band/category. In the event, the income cut-off point for Sydney, Melbourne and Brisbane was $25,000 ($481 per week) in All households under this income figure and in private rental were selected for further analysis. However, as the analysis required the use of income categories from the census the proportion of households in each of the cities under $25,000 differed from 20%. The proportion of households who earned under $25,000 per annum in 1991 was 17.2% in Sydney, 20.2% in Melbourne, and 24.4% in Adelaide. To compare the number of low income private renters in 2001, the proportions in 1991 were then used to select the number of PRLIHs from the value added census file created for In other words, the households with the bottom 17.2% of incomes in Sydney, 20.2% in Melbourne and 24.4% in Adelaide were selected for 2001 to compare with the 1991 data. The analysis of the changes in PRLIHs over time is presented in chapter 6. While this changes the base numbers involved in the analysis for 2001, this does not invalidate the 5 The employment status of individuals who did not state their employment status on Census night was re-imputed for this project. As such the numbers of persons and households will differ slightly from the numbers presented in the Yates et al study, although the differences are minimal. 19

25 analysis as the focus here is in the changing spatial distribution of this group, not the change in overall numbers per se. In any event, the use of a strictly comparable income baseline in both years means the results can also be taken as indicative of the changing numbers of lower income private renting households. 20

26 Table 4.3: The weekly household income distribution of all households, private renters and employed households in Sydney, 2001 All Households Percent Cumulative Private Renters Percent Cumulative Employed Percent Cumulative Negative income 3, % 0.2% 1, % 0.5% 1, % 0.1% Nil income 10, % 0.9% 5, % 2.2% % 0.2% $1-$119 7, % 1.4% 1, % 2.7% 1, % 0.3% $120-$159 10, % 2.1% 1, % 3.1% 2, % 0.5% $160-$199 31, % 4.1% 3, % 4.2% 2, % 0.8% $200-$299 85, % 9.8% 13, % 8.4% 11, % 1.9% $300-$399 90, % 15.8% 18, % 14.4% 21, % 4.0% $400-$499 76, % 20.8% 16, % 19.4% 33, % 7.3% $500-$599 59, % 24.7% 19, % 25.7% 40, % 11.2% $600-$699 69, % 29.2% 21, % 32.4% 50, % 16.2% $700-$799 56, % 33.0% 17, % 37.8% 48, % 21.0% $800-$ , % 40.6% 30, % 47.3% 102, % 31.1% $1000-$ , % 48.4% 33, % 57.7% 110, % 41.9% $1200-$ , % 55.8% 26, % 65.9% 107, % 52.5% $1500 or more 368, % 80.1% 75, % 89.7% 358, % 87.6% Partial incomes stated 109, % 87.3% 23, % 96.9% 91, % 96.6% All incomes not stated 52, % 90.7% 9, % 99.9% 15, % 98.1% Not applicable 140, % 100.0% % 100.0% 19, % 100.0% Total 1,517, % 319, % 1,019, % (source: ABS, 2001 Census CURF) 21

27 Table 4.4: The weekly household income distribution of all households, private renters and employed households in Melbourne, 2001 All Households Percent Cumulative Private Renters Percent Cumulative Employed Percent Cumulative Negative income 2, % 0.2% % 0.1% % 0.1% Nil income 8, % 0.8% 2, % 1.3% 1, % 0.2% $1-$119 5, % 1.2% 1, % 1.8% 1, % 0.4% $120-$159 10, % 2.0% 2, % 2.9% 2, % 0.6% $160-$199 26, % 4.1% 4, % 4.7% 3, % 1.0% $200-$299 75, % 9.9% 13, % 10.7% 12, % 2.4% $300-$399 91, % 16.9% 17, % 18.2% 22, % 5.0% $400-$499 80, % 23.1% 16, % 25.5% 35, % 8.9% $500-$599 64, % 28.0% 19, % 34.2% 45, % 14.2% $600-$699 72, % 33.5% 16, % 41.2% 51, % 20.0% $700-$799 57, % 37.9% 13, % 47.2% 51, % 25.8% $800-$ , % 46.4% 23, % 57.5% 100, % 37.3% $1000-$ , % 54.4% 22, % 67.2% 98, % 48.4% $1200-$ , % 62.1% 17, % 74.9% 99, % 59.7% $1500 or more 255, % 81.7% 38, % 91.8% 251, % 88.2% Partial incomes stated 94, % 89.0% 12, % 97.4% 76, % 96.9% All incomes not stated 44, % 92.4% 5, % 99.9% 14, % 98.6% Not applicable 99, % 100.0% % 100.0% 12, % 100.0% Total 1,304, % 227, % 879, % (source: ABS, 2001 Census CURF) 22

28 Table 4.5: The weekly household income distribution of all households, private renters and employed households in Adelaide, 2001 All Households Percent Cumulative Private Renters Percent Cumulative Employed Percent Cumulative Negative income % 0.1% % 0.1% % 0.1% Nil income 1, % 0.3% % 0.6% % 0.1% $1-$119 2, % 0.9% 1, % 2.0% % 0.4% $120-$159 4, % 2.0% % 2.8% 1, % 0.8% $160-$199 16, % 5.6% 2, % 7.0% 1, % 1.4% $200-$299 38, % 14.2% 6, % 15.8% 4, % 3.2% $300-$399 40, % 23.0% 7, % 26.9% 10, % 6.8% $400-$499 37, % 31.4% 7, % 37.8% 17, % 13.1% $500-$599 28, % 37.6% 6, % 47.1% 19, % 20.2% $600-$699 31, % 44.5% 5, % 55.8% 22, % 28.4% $700-$799 19, % 48.9% 3, % 60.8% 16, % 34.2% $800-$999 39, % 57.5% 7, % 71.0% 35, % 47.0% $1000-$ , % 64.7% 5, % 78.3% 30, % 58.0% $1200-$ , % 71.8% 4, % 84.9% 31, % 69.3% $1500 or more 58, % 84.7% 5, % 93.1% 57, % 89.9% Partial incomes stated 26, % 90.6% 3, % 97.5% 20, % 97.4% All incomes not stated 12, % 93.3% 1, % 99.7% 3, % 98.6% Not applicable 30, % 100.0% % 100.0% 4, % 100.0% Total 453, % 68, % 276, % (source: ABS, 2001 Census CURF) 23

29 Table 4.6: The number of households using different household income cut-off points from this project and the Yates et al. study Household Income Cut-Off Point Australia This project bottom 20% of employed households ($0-$655.23) Yates et al (2003) low income household ($0-$334.50) Yates et al (2003) low to moderate income household ($ $557.50) Employed Households Unemployed Households Other Households Total 931, ,525 1,647,889 2,701, ,957 85, ,801 1,129, ,599 32, ,203 1,205,143 Household Income more than $ ,991,715 13, ,060 4,410,584 Total Households 4,658, ,460 1,955,064 6,744,795 Sydney This project bottom 20% of employed households ($0-$655.23) Yates et al (2003) low income household ($0-$334.50) Yates et al (2003) low to moderate income household ($ $557.50) 200,143 16, , ,144 24,022 10, , ,585 71,337 4, , ,297 Household Income more than $ ,356 2,949 82, ,474 Total Households 1,000,715 17, ,651 1,366,356 Melbourne This project bottom 20% of employed households ($0-$655.23) Yates et al (2003) low income household ($0-$334.50) Yates et al (2003) low to moderate income household ($ $557.50) 173,466 16, , ,025 25,947 11, , ,073 74,540 4, , ,772 Household Income more than $ ,841 2,289 69, ,299 Total Households 867,328 18, ,535 1,196,144 Adelaide This project bottom 20% of employed households ($0-$655.23) Yates et al (2003) low income household ($0-$334.50) Yates et al (2003) low to moderate income household ($ $557.50) 54,416 8, , ,886 10,317 6,728 63,178 80,223 31,656 1,970 49,599 83,225 Household Income more than $ , , ,597 Total Households 272,079 9, , ,045 (Source: ABS customised matrix) 24

30 5 LOW INCOME HOUSEHOLDS IN THE PRIVATE RENTAL MARKET 5.1 Introduction This chapter profiles low income households in the private rental market. The first section introduces the tenure distribution of employed and unemployed low income households before examining the spatial distribution of low income households in the private rental market. Finally, an analysis of the socio-economic characteristics of low income households in the private rental market is presented. 5.2 The relative position of PRLIH in the housing market In 2001, there were 577,031 low income households in the private rental market in Australia. This represented 21% of all low income households. The largest tenure group among all low income households were outright owner-occupiers (47%), reflecting the high proportion of older and retired people with low incomes. Home buyers accounted for just 12% of this group. In Sydney, 117,545 low income households rented privately (23% of all low income households in the city) while in Melbourne 92,982 low income households rented privately (20%) as did 32,559 households (18%) in Adelaide (see Table 5.1). The higher percentage in Sydney is likely to reflect both the problems lower income households have in buying a home in Sydney as well as the greater investment incentives from higher property markets in this city leading to greater rental supply. In other words, high property prices effectively squeeze lower income households out of home ownership on both the demand and supply side of the equation. Outright owners were again the largest group among low income households in the three case study areas: 51% in Melbourne, 47% in Sydney and 44% in Adelaide. Both Melbourne and Adelaide (14%) had a larger proportion of low income households than Sydney (10%) who were purchasing their dwelling, again most probably an outcome of differential affordability. Importantly though, Adelaide clearly had the largest proportion of low income households in the public rental sector. Some 16% of low income households in Adelaide rented publicly, compared to 12% in Sydney and 7% in Melbourne. However, low income private renters have a relatively high workforce participation rate. In Sydney, 60% (69,873) of the 117,545 low income households in the private rental market included at least one person in work. In Melbourne and Adelaide the figures were 55% (51,354) and 48% (15,562) respectively (Table 5.1). The second significant feature of Table 5.1 is the importance of the private rental sector as the largest tenure for unemployed low income households. In Sydney, Melbourne and Adelaide between 43% and 46% of unemployed low income households rent privately. Only 17% of unemployed low income households in Sydney, 11% in Melbourne and 20% in Adelaide are in publicly rented housing. In contrast, a significantly higher proportion of employed low income households are owneroccupiers or purchasing their home compared to unemployed households. Some 32% of employed low income households in Sydney own their dwelling, as do 34% in Melbourne and 27% in Adelaide. And while 19% of employed low income households are purchasing their dwelling in Sydney, this is much lower than the proportions in Melbourne and Adelaide (26% and 29% respectively). Outright ownership is the dominant tenure for other households, again, a reflection of the high proportion of retired and older people in this category. Nevertheless, around three in ten employed low income households in the three cities rent privately. 25

31 Table 5.1a: Tenure by labour force status for low income households, 2001 Sydney Melbourne Adelaide Employed Unemployed Other Total Employed Unemployed Other Total Employed Unemployed Other Total Fully owned 64,237 2, , ,970 58,779 3, , ,677 14,664 1,191 62,743 78,598 Being Purchased 37,882 1,491 12,067 51,440 44,347 2,446 16,196 62,989 15,858 1,215 7,390 24,463 Renting from Private Landlord 6 69,873 7,495 40, ,545 51,354 7,199 34,429 92,982 15,562 3,787 13,210 32,559 Renting from Public Landlord 11,718 2,811 45,698 60,227 5,119 1,850 23,020 29,989 4,340 1,784 21,886 28,010 Other Landlord Type 4, ,835 13,162 3, ,914 11,028 1, ,229 6,095 Total Rented 86,178 11,046 93, ,914 59,861 9,775 64, ,007 21,318 6,021 39,326 66,665 Other Tenure Type 6, ,721 18,242 5, ,049 13,911 1, ,964 6,551 Not Stated 5, ,310 16,558 5, ,706 15,449 1, ,206 4,610 Total 200,143 16, , , ,466 16, , ,025 54,416 8, , ,886 (Source: ABS customised matrix) 6 Includes persons who rent from a real estate agent, person not in the same household, person in the household, an employee, community or cooperative housing group. 26

32 Table 5.1b: Tenure by labour force status for low income households, 2001 Sydney Melbourne Adelaide Employed Unemployed Other Total Employed Unemployed Other Total Employed Unemployed Other Total Fully owned 32.1% 17.6% 57.9% 46.7% 33.9% 20.8% 63.1% 50.5% 26.9% 13.5% 53.3% 43.5% Being Purchased 18.9% 9.1% 4.0% 9.9% 25.6% 14.5% 6.1% 13.8% 29.1% 13.7% 6.3% 13.5% Renting from Private Landlord 34.9% 45.9% 13.2% 22.6% 29.6% 42.7% 12.9% 20.3% 28.6% 42.8% 11.2% 18.0% Renting from Public Landlord 5.9% 17.2% 15.0% 11.6% 3.0% 11.0% 8.6% 6.6% 8.0% 20.2% 18.6% 15.5% Other Landlord Type 2.3% 4.5% 2.6% 2.5% 2.0% 4.3% 2.6% 2.4% 2.6% 5.1% 3.6% 3.4% Total Rented 43.1% 67.6% 30.9% 36.7% 34.5% 58.0% 24.1% 29.3% 39.2% 68.1% 33.4% 36.9% Other Tenure Type 3.0% 3.2% 3.9% 3.5% 3.0% 3.7% 3.0% 3.0% 2.5% 2.8% 4.2% 3.6% Not Stated 2.9% 2.5% 3.4% 3.2% 3.0% 3.0% 3.6% 3.4% 2.3% 1.9% 2.7% 2.5% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% (Source: ABS customised matrix) 27

33 Table 5.1c: Tenure by labour force status for low income households, 2001 Sydney Melbourne Adelaide Employed Unemployed Other Total Employed Unemployed Other Total Employed Unemployed Other Total Fully owned 26.4% 1.2% 72.4% 100.0% 25.5% 1.5% 73.0% 100.0% 18.7% 1.5% 79.8% 100.0% Being Purchased 73.6% 2.9% 23.5% 100.0% 70.4% 3.9% 25.7% 100.0% 64.8% 5.0% 30.2% 100.0% Renting from Private Landlord 59.5% 6.4% 34.2% 100.0% 55.2% 7.7% 37.0% 100.0% 47.8% 11.6% 40.6% 100.0% Renting from Public Landlord 19.5% 4.7% 75.9% 100.0% 17.1% 6.2% 76.8% 100.0% 15.5% 6.4% 78.1% 100.0% Other Landlord Type 34.9% 5.6% 59.5% 100.0% 30.7% 6.6% 62.7% 100.0% 23.2% 7.4% 69.4% 100.0% Total Rented 45.1% 5.8% 49.1% 100.0% 44.7% 7.3% 48.0% 100.0% 32.0% 9.0% 59.0% 100.0% Other Tenure Type 32.9% 2.9% 64.3% 100.0% 37.7% 4.5% 57.9% 100.0% 20.4% 3.8% 75.8% 100.0% Not Stated 35.3% 2.4% 62.3% 100.0% 33.9% 3.3% 62.8% 100.0% 26.9% 3.6% 69.5% 100.0% Total 38.5% 3.1% 58.4% 100.0% 38.0% 3.7% 58.4% 100.0% 30.1% 4.9% 65.0% 100.0% (Source: ABS customised matrix) 28

34 5.3 Location of low income households in the private rental market Concentrations at the LGA/SLA level The remainder of this chapter focuses on the position of low income households who rent privately. Table 5.2 and Figures 5.1 to 5.3 show the LGAs with the highest concentrations of low income private rental households in Sydney, Melbourne and Adelaide 7. In Sydney, the highest concentrations of PRLIHs are located in Canterbury, Fairfield and Blacktown in the inner and outer west of the city and Gosford and Wyong to the far north of the metropolitan area. Altogether, these 10 LGAs accounted for 45% of the total PRLIHs in Sydney at this time, but only 36% of all households. This indicates the degree of concentration of the PRLIH sector in these 10 LGAs. In Melbourne, the largest concentrations of PRLIHs are found in the local government areas of Darebin and Moreland to the north of the CBD and Greater Dandenong to the south-east. Here, the top 10 LGAs accounted for almost half (49%) of the total PRLIHs in Melbourne, but only 38% of all households. In Adelaide there are significant concentrations of PRLIHs in West Torrens to the west of the CBD and Port Adelaide Enfield, to the north of the CBD. However, the concentration of PRLIH in the top 10 LGAs is noticeable lower than in Melbourne or Sydney, accounting for only a third (32%) of all PRLIHs and just 25% of total households. This finding might suggest a much lower level of spatial concentration of the low income rental market in Adelaide compared to the other two cities, although this is more likely to be a reflection of the larger number of SLAs in Adelaide, compared to LGAs in the other two cities. Overall, 10% of the households in these thirty areas were low income private renters, a proportion that remains remarkably uniform across the three cities. The individual percentage of all households in this group at the LGA/SLA level ranged around this average, from 15.7% in West Torrens East SLA in Adelaide to 7.3% in Blacktown LGA in Sydney. 7 A full listing of these data by LGA is given in Appendix 1. 29

35 Table 5.2: The 10 local government areas/statistical local areas in Sydney, Melbourne and Adelaide with the highest number of low income private rental households LGA/SLA Sydney Number of Low Income Private Renters Total Households Canterbury 6,575 43, % Wyong 6,155 49, % Fairfield 5,927 53, % Blacktown 5,728 79, % Gosford 5,557 57, % South Sydney 4,941 36, % Penrith 4,804 55, % Parramatta 4,783 48, % Randwick 4,600 44, % Marrickville 4,249 27, % Total 53, , % Melbourne Darebin 5,800 47, % Moreland 5,583 50, % Greater Dandenong 5,167 41, % Port Phillip 5,066 35, % Glen Eira 4,894 46, % Kingston 4,122 48, % Stonnington 3,913 36, % Frankston 3,716 40, % Boroondara 3,695 55, % Mornington Peninsula 3,623 47, % Total 45, , % Adelaide West Torrens - East 1,645 10, % West Torrens - West 1,196 11, % Holdfast Bay - North 1,105 8, % Port Adelaide Enfield - East 1,040 11, % Playford - Elizabeth , % Charles Sturt - North-East , % Charles Sturt - Coastal , % Port Adelaide Enfield - Port , % Marion - North , % Norwood Payneham St Peters - West 873 7, % Total 10, , % % 30

36 5.3.2 Concentrations at the suburb and postcode level The results presented above at the LGA/SLA level tend to either over-state or underestimate some of the differences in the number of PRLIHs at lower spatial levels (Table 5.3 and Figures 5.4 to 5.6). While generally the largest concentrations of PRLIHs at the LGA/SLA level can also be seen at the suburb level, a few other areas emerge at this level. For example, in Sydney the highest concentration of PRLIHs at the suburb level is in Liverpool, although Liverpool LGA was not on our top ten list. Conversely, an examination of the top ten suburbs in Sydney shows that no suburb in Gosford or Wyong emerges, despite these two LGAs having high concentrations of PRLIHs. This suggests that there are a number of suburbs in Gosford-Wyong with significant numbers of PRLIHs, whereas in Liverpool LGA the suburb of Liverpool has significantly more PRLIHs than other suburbs in that LGA. Similarly, in Melbourne, the highest concentrations of PRLIHs are located in Frankston to the south of the CBD, Reservoir to the north and in Dandenong and Noble Park to the south west. In Adelaide, the highest concentration of PRLIHs can be found in the suburb of Morphett Vale to the far south and in Salisbury to the north. Further, in 12 suburbs in Sydney (Kanwal, Lisarow, Parklea, The Entrance, Meadowbank, Gosford, Lakemba, Wiley Park, Cabramatta, Haymarket, Fairfield and Harris Park), 25% or more of the households in that suburb were PRLIHs. In Melbourne and Adelaide though, only 2 (Albion and Gardenvale) and 1 (Evanston South) suburb(s) contained 25% or more PRLIHs. The location of PRLIHs at the suburb level is also reflected at the postcode level. These findings are further amplified by the postcode level maps presented in Figures 5.7 to 5.8. Again, the local nature of the low income rental market is clearly shown. These maps have been included here as they are utilised later in chapter 8 to explore the relationship between the sector and rental assistance recipients and waiting list data. The suburb and post code level analyses highlight how locally concentrated low income working private renters are within each city. Previous research by the authors has pin pointed the local nature of the housing market segments this group are closely associated with, for example, the low rise suburban flat market in Sydney which is clustered around local town centres and transport nodes (Randolph and Holloway, 2005; Bunker, Holloway and Randolph, 2005). Analysis at the LGA/SLA level clearly masks important local variations in the housing markets which structure the locational choices of this group. 31

37 Table 5.3: The 10 suburbs in Sydney, Melbourne and Adelaide with the highest number of low income private rental households Suburb Sydney Number of Low Income Private Renters Total Households Liverpool 1,717 7, % Campsie 1,650 6, % Cabramatta 1,627 6, % Randwick 1,377 10, % Ashfield 1,358 7, % Marrickville 1,346 8, % Auburn 1,326 7, % Lakemba 1,275 4, % Bankstown 1,268 7, % Parramatta 1,265 5, % Melbourne Frankston 1,834 14, % Reservoir 1,793 17, % Dandenong 1,534 6, % Noble Park 1,515 11, % St Kilda 1,500 7, % Brunswick 1,359 8, % Thornbury 1,250 6, % St Albans 1,199 11, % Preston 1,197 10, % Hawthorn 1,175 7, % Adelaide Morphett Vale 647 9, % Salisbury 516 2, % Prospect 514 5, % Seaton 374 4, % Adelaide 373 3, % Brooklyn Park 372 1, % Plympton 369 2, % Glenelg North 365 2, % North Adelaide 330 2, % Norwood 327 2, % % 32

38 Figure 5.1: Location of low income private renters in Sydney by local government area in

39 Figure 5.2: Location of low income private renters in Melbourne by local government area in

40 Figure 5.3: Location of low income private renters in Adelaide by statistical local area in

41 Figure 5.4: Location of low income private renters in Sydney by suburb in

42 Figure 5.5: Location of low income private renters in Melbourne by suburb in

43 Figure 5.6: Location of low income private renters in Adelaide by suburb in

44 Figure 5.7: Location of low income private renters in Sydney by postcode in

45 Figure 5.8: Location of low income private renters in Melbourne by postcode in

46 Figure 5.9: Location of low income private renters in Adelaide by postcode in

47 5.4 Profile of low income households in the private rental market Having established where PRLIHs live, this section focuses on who these households are, in terms of their socio-demographic and housing characteristics. An analysis of a selected range of socio-economic variables from the 2001 census provides a profile of the social characteristics of households in the sector and how they differ from the population as a whole. Overall, in 2001 there were 117,545 PRLIHs in Sydney, 92,982 in Melbourne, 32,559 in Adelaide and 577,067 across Australia. Table 5.4. and 5.5 set out the profile of PRLIHs from the 2001 Census Dwelling type Confirming the note above from previous research by the authors, the data show that the proportion of PRLIHs who live in flats in our three cities is significantly greater than that for the rest of the population. In Sydney only 33% of PRLIHs live in separate houses whereas 53% reside in flats. In Melbourne, 42% of PRLIHs live in flats, while in Adelaide 36% live in flats. This is twice the proportion of all households who live in flats in Sydney and approximately three times the proportion in Melbourne and Adelaide. Further, the proportion of PRLIHs who reside in semi detached dwellings in Sydney, Melbourne and Adelaide is also higher than that recorded for the population as a whole in the three cities Rent paid Not surprisingly, the weekly rents of PRLIHs in Sydney, Melbourne and Adelaide are concentrated in certain bands. Only 3% of households in Sydney, 7% in Melbourne and 15% in Adelaide pay less than $100 per week in rent. This compares with 14% in Sydney and Melbourne, and 32% of all households in the three cities. The majority of these low rent payers will be paying rebated rents in public housing. Conversely, only 15% of PRLIHs in Sydney, 6% in Melbourne, and 2% in Adelaide pay more than $300 per week in rent. This compares with 25% of all households in Sydney, and 11% and 3% of all households in Melbourne and Adelaide. In fact, the great majority of PRLIHs pay between $100 and $250 per week in rent. In Sydney, 71% of PRLIHs, 83% in Melbourne and 82% in Adelaide pay rents in this band Household type A significant proportion of PRLIHs are single person households. In fact, across the three cities the proportion of single person PRLIHs is at least twice as high as that for the population as a whole. In Sydney, 40% of PRLIHs are lone person households while another 22% are one parent families. Similarly, in Melbourne 46% of PRLIHs are lone person households and a further 21% are one parent families. In Adelaide, 49% of PRLIHs are lone person households and 25% are one parent families. Conversely, 30% of PRLIHs in Sydney are couple families compared with 59% of all households. In Melbourne, 22% are couple families compared with 60% of all households. Interestingly in Adelaide only 17% of PRLIHs are couple families compared to 55% for all households Age Perhaps surprisingly, the proportion of children (aged 0-14) in PRLIHs in the three cities is higher than for all households. The result is that overall, the age profile of PRLIHs is younger than that for all households. In Sydney, 27% of persons in PRLIHs are aged 0-14, 21% are aged while 11% are aged over 55. For all households in Sydney the comparable figures are 20%, 16% and 21% respectively. A similar situation occurs in Melbourne and Adelaide. In Melbourne, 24% of persons in PRLIHs are aged 0-14 compared with 20% for the entire population. Some 12% of persons in PRLIHs in Melbourne are aged over 55 compared with 21% for the population as a whole. In Adelaide, 25% of persons in PRLIHS are aged 0-14 compared with 19% across the entire population, whereas, 10% of persons in PRLIHs in Adelaide are aged over 55 compared to 24% for the population as a whole. Further, there are significantly lower proportions of persons aged in PRLIHs in the three cities compared to their populations as a whole. 42

48 In other words, the PRLIH sector has a much younger profile than the overall population and contains above average proportions of younger children, the latter reflecting the high proportions of single parents in the sector Country of birth Overall, the PRLIH sector is biased towards non-australian born, although more so in Sydney. In Sydney, 52% of persons in PRLIHs were born in Australia compared with 62% for all Sydney households. In Melbourne, 60% of all persons in PRLIHs were born in Australia compared to 66% for the population as a whole. In Adelaide, 74% of persons in PRLIHs were born in Australia compared to 72% for the population as a whole. In Sydney, there were higher proportions of persons in PRLIHs who were born in Asia, the Middle East and Other Oceania countries compared to the entire Sydney population. Persons who were born in Asia, the Middle East and Other Oceania countries also comprise significant proportions of persons in PRLIHs in Melbourne and Adelaide. Conversely, in Sydney, Melbourne and Adelaide the proportion of persons in PRLIHs who were born in Europe is significantly lower than that for the entire populations of those cities Education Despite having larger overseas born populations, the level of educational qualifications for persons in PRLIHs is similar to that for the population as a whole. While there are slightly more persons with School Certificates and slightly fewer with a Bachelor Degree, the educational profile of persons in PRLIHs in Sydney, Melbourne and Adelaide does not differ significantly from that of the entire population Occupation Given their low incomes, PRLIHs tend to work in occupations with lower skill and educational requirements. In Sydney, Melbourne and Adelaide, 6% of PRLIHs are employed as Labourers and Related workers. This compares with 3% of all employed persons in Sydney and Melbourne and 4% in Adelaide. Conversely, 8% of employed persons in PRLIHs in Sydney and Melbourne and 5% in Adelaide are employed as Managers, Administrators or Professional workers. This compares with 14% of all employed persons in Sydney, 13% in Melbourne and 11% in Adelaide. Employed persons in PRLIHs in Sydney, Melbourne and Adelaide are over-represented in Intermediate Clerical, Sales and Service jobs as well as Intermediate Production and Transport workers, and Elementary Clerical, Sales and Service workers Industry The industries that employed PRLIHs work in are not very different to the population as a whole. PRLIHs are slightly over-represented in Sydney, Melbourne and Adelaide in Personal Services, Retail Trade, Transport and Storage, Accommodation, Cafes and Restaurants, and Cultural and Recreational Services. Conversely, persons in PRLIHs in the three cities are slightly under-represented in Finance and Insurance, Property and Business Services, Construction, Education, Electricity, Gas and Water Supply, Communication Services and Government Administration and Defence. This implies that low paid renters occupy low skilled positions across all industrial sectors, making up a substantial section of the workforce whose jobs support those in higher skilled or higher income occupations in their industries. 43

49 5.4.9 Income The weekly household income distributions of low income working households show a substantial variation between the three cities. As might be expected, incomes of this group are higher in Sydney, with three in ten (29%) receiving incomes over $600 per week (up to the income threshold of $773 in Sydney). This compares to 18% in Melbourne and just 4% in Adelaide. The income thresholds used for the latter two cities are lower, of course ($706 and $625 respectively), but nevertheless, these figures reflect clear income relativities between the three city s low income working populations. This is reflected at the other end of the income range, with 33% of PRLIHs earning under $300 per week in Adelaide, compared to 28% in Melbourne and 22% in Sydney. Higher rental levels in Sydney and Melbourne are therefore to some extent compensated for by higher incomes in these two cities. 44

50 Table 5.4: Socio-economic profile of low income households in the private rental market Sydney Melbourne Adelaide Australia Dwelling type Separate House 39,126 37,886 14, ,476 Semi detached dwellings 14,094 13,693 5,830 82,016 Flat, unit or apartment 61,750 39,332 11, ,397 Other dwelling 2,575 2, ,178 Total 117,545 92,982 32, ,067 Household type Couple with children 21,920 11,852 3,004 74,382 Couple family without children 12,984 9,028 2,606 59,105 One parent family 25,723 19,580 8, ,158 Lone person households 47,159 42,809 15, ,985 Group Households 7,533 7,549 2,253 39,052 Other households 2,226 2, ,385 Total 117,545 92,982 32, ,067 Age of reference person years 16,020 15,182 6,645 99, years 34,028 26,610 9, , years 30,408 20,804 7, , years 17,112 12,870 4,184 80, years 9,887 7,850 2,252 50, years or more 10,073 9,652 2,771 61,136 Not Stated/Not Applicable Total 117,545 92,982 32, ,067 Household income $0 4,822 3, ,668 $1-$ ,205 $40-$ ,861 $80-$ ,351 $120-$159 1,722 1, ,861 $160-$199 4,052 4,414 2,160 32,758 $200-$299 13,842 14,955 6, ,999 $300-$399 18,007 16,330 7, ,892 $400-$499 18,930 16,614 7, ,086 $500-$599 20,339 17,393 6, ,415 $600 or more 34,392 16,904 1,224 50,888 Not Stated/Not Applicable Total 117,545 92,982 32, ,067 Birthplace of all persons Australia 137, ,626 46, ,849 Other Oceania and Antarctica 13,785 5, ,097 Northern and Western Europe 12,903 10,640 5,467 72,597 45

51 Sydney Melbourne Adelaide Australia Southern and Eastern Europe 10,328 10,855 2,166 31,022 Middle East 14,599 5, ,758 North Africa 2, ,292 Asia 52,015 28,270 3,684 96,527 North America 1, ,583 South and Central America 4,031 1, ,640 Carribbean Sub Saharan Africa 2,692 2, ,603 Other/Not Stated 13,183 8,438 2,488 56,494 Total 264, ,668 62,793 1,213,739 Level of post-school qualification Postgraduate Degree 2,781 1, ,098 Graduate Diploma and Graduate Certificate 986 1, ,902 Bachelor Degree 12,379 9,355 2,287 37,577 Advanced Diploma and Diploma 8,672 6,706 1,592 31,585 Certificate 18,228 12,944 5,303 91,858 Not Stated/Not Applicable 74,499 61,162 22, ,047 Total 117,545 92,982 32, ,067 Weekly Rent $0-$ ,582 $50-$99 2,698 5,816 4,670 72,319 $100-$149 15,241 32,319 15, ,220 $150-$199 38,616 34,256 9, ,066 $200-$249 29,249 10,562 1,330 53,880 $250-$299 14,251 3, ,916 $300-$349 7,555 1, ,272 $350-$399 3, ,661 $400-$449 1, ,393 $450-$499 1, ,214 $500 and over 2,652 1, ,461 Not Stated Total 117,545 92,982 32, ,067 Occupation Managers and Administrators 1,990 1, ,846 Professionals 7,540 5,887 1,413 24,226 Associate Professionals 6,202 4,523 1,244 23,227 Tradespersons and Related Workers 7,109 4,947 1,591 29,044 Advanced Clerical and Service Workers 1,812 1, ,889 Intermediate Clerical, Sales and Service Workers Intermediate Production and Transport Workers 13,381 10,218 3,431 53,951 7,223 4,893 1,448 25,803 46

52 Elementary Clerical, Sales and Service Workers Sydney Melbourne Adelaide Australia 7,646 6,068 1,942 34,595 Labourers and Related Workers 6,691 5,160 1,980 36,431 Not Stated/Not Applicable 57,951 48,724 18, ,055 Total 117,545 92,982 32, ,067 Industry Accommodation, Cafes and Restaurants 5,570 3,764 1,275 25,378 Agriculture, Forestry and Fishing ,146 Communication Services 1, ,333 Construction 3,462 1, ,428 Cultural and Recreational Service 2,184 1, ,516 Education 2,723 2, ,543 Electricity, Gas and Water Supply Finance and Insurance 2,184 1, ,209 Government Administration and Defence 1, ,774 Health and Community Services 5,713 4,710 1,837 26,466 Manufacturing 7,588 6,441 1,990 28,086 Mining Personal and Other Services 2,573 1, ,432 Property and Business Services 7,274 5,267 1,386 24,171 Retail Trade 9,228 7,639 2,491 44,534 Transport and Storage 3,602 2, ,889 Wholesale Trade 3,911 2, ,469 Not Stated/Not Applicable 58,369 48,931 18, ,574 Total 117,545 92,982 32, ,067 47

53 Table 5.5: Socio-economic profile of low income households in the private rental market (%) Low Income Private Rental Households All Households Sydney Melbourne Adelaide Australia Sydney Melbourne Adelaide Australia Dwelling type Separate House 33.3% 40.7% 45.2% 49.3% 63.1% 74.0% 75.1% 75.3% Semi detached dwellings 12.0% 14.7% 17.9% 14.2% 11.3% 10.3% 13.3% 8.9% Flat, unit or apartment 52.5% 42.3% 35.8% 34.0% 23.9% 14.3% 10.6% 13.0% Other dwelling 2.2% 2.2% 1.1% 2.5% 1.8% 1.5% 1.1% 2.7% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Household type Couple with children 18.6% 12.7% 9.2% 12.9% 35.9% 35.6% 29.8% 32.8% Couple family without children 11.0% 9.7% 8.0% 10.2% 23.1% 23.3% 25.3% 24.9% One parent family 21.9% 21.1% 25.4% 25.2% 10.7% 10.5% 11.2% 10.8% Lone person households 40.1% 46.0% 48.7% 43.3% 21.3% 22.3% 27.2% 22.9% Group Households 6.4% 8.1% 6.9% 6.8% 4.1% 4.1% 3.4% 3.7% Other households 1.9% 2.3% 1.7% 1.6% 4.9% 4.2% 3.0% 4.9% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Age of reference person years 13.6% 16.3% 20.4% 17.3% NA NA NA NA years 28.9% 28.6% 29.3% 26.9% NA NA NA NA years 25.9% 22.4% 22.0% 22.7% NA NA NA NA years 14.6% 13.8% 12.9% 13.9% NA NA NA NA years 8.4% 8.4% 6.9% 8.8% NA NA NA NA 65 years or more 8.6% 10.4% 8.5% 10.6% NA NA NA NA Not Stated/Not Applicable 0.0% 0.0% 0.0% 0.0% NA NA NA NA Total 100.0% 100.0% 100.0% 100.0% NA NA NA NA 48

54 Low Income Private Rental Households All Households Sydney Melbourne Adelaide Australia Sydney Melbourne Adelaide Australia Age of all persons 0-14 years 27.0% 23.5% 25.4% 27.3% 20.2% 19.8% 18.9% 20.8% years 16.5% 19.4% 21.3% 18.7% 14.0% 14.2% 13.8% 13.7% years 20.8% 21.1% 20.6% 18.7% 16.0% 16.0% 14.0% 14.5% years 16.2% 14.8% 14.2% 14.3% 15.7% 15.6% 15.1% 15.3% years 8.8% 8.8% 8.2% 8.7% 13.4% 13.4% 14.1% 13.7% years 5.3% 5.6% 4.7% 5.7% 8.8% 8.9% 9.4% 9.4% 65 years or more 5.5% 6.8% 5.7% 6.6% 11.9% 12.1% 14.6% 12.6% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Household income Less than $ % 11.6% 12.3% 11.7% 4.0% 4.1% 5.4% 4.6% $200-$ % 16.1% 20.2% 18.7% 5.7% 6.1% 8.9% 7.2% $300-$ % 17.6% 21.6% 20.9% 6.3% 7.2% 9.4% 8.4% $400-$ % 17.9% 21.6% 20.3% 5.4% 6.2% 8.3% 7.1% $500-$ % 18.7% 20.5% 19.5% 4.1% 4.8% 6.0% 5.2% $600 or more 29.3% 18.2% 3.8% 8.8% 58.3% 56.7% 50.8% 52.1% Not Stated 0.0% 0.0% 0.0% 0.0% 16.1% 14.9% 11.2% 15.4% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Birthplace of all persons Australia 51.9% 60.2% 73.7% 71.3% 62.2% 65.7% 72.2% 72.6% Other Oceania and Antarctica 5.2% 3.1% 1.5% 3.7% 3.2% 1.8% 0.9% 2.4% Northern and Western Europe 4.9% 5.6% 8.7% 6.0% 6.3% 6.6% 11.6% 7.2% Southern and Eastern Europe 3.9% 5.8% 3.4% 2.6% 5.4% 8.4% 5.9% 4.1% Middle East 5.5% 2.8% 1.1% 1.7% 2.6% 1.3% 0.4% 0.9% 49

55 Low Income Private Rental Households All Households Sydney Melbourne Adelaide Australia Sydney Melbourne Adelaide Australia North Africa 0.8% 0.5% 0.2% 0.3% 0.5% 0.4% 0.1% 0.2% Asia 19.7% 15.0% 5.9% 8.0% 10.6% 8.0% 3.5% 5.2% North America 0.5% 0.4% 0.5% 0.5% 0.5% 0.4% 0.4% 0.4% South and Central America 1.5% 0.8% 0.4% 0.5% 0.9% 0.5% 0.2% 0.4% Caribbean 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Sub Saharan Africa 1.0% 1.2% 0.6% 0.8% 1.0% 1.0% 0.4% 0.8% Other/Not Stated 5.0% 4.5% 4.0% 4.7% 6.7% 5.8% 4.3% 5.6% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Level of post-school qualification Postgraduate Degree 2.4% 1.7% 0.9% 1.1% 2.3% 1.7% 1.3% 1.4% Graduate Diploma and Graduate Certificate 0.8% 1.3% 0.7% 0.7% 1.0% 1.5% 1.1% 1.1% Bachelor Degree 10.5% 10.1% 7.0% 6.5% 9.8% 9.7% 7.5% 7.7% Advanced Diploma and Diploma 7.4% 7.2% 4.9% 5.5% 5.5% 5.2% 4.7% 4.8% Certificate 15.5% 13.9% 16.3% 15.9% 12.2% 11.1% 12.8% 12.5% Not Stated/Not Applicable 63.4% 65.8% 70.2% 70.4% 69.2% 70.8% 72.6% 72.5% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Weekly Rent $0-$49 0.3% 0.7% 0.9% 1.0% 5.6% 3.6% 12.2% 6.7% $50-$99 2.3% 6.3% 14.3% 12.5% 8.5% 10.0% 19.8% 15.0% $100-$ % 34.8% 47.6% 39.5% 8.1% 20.8% 30.7% 24.1% $150-$ % 36.8% 30.4% 29.8% 18.4% 29.7% 23.4% 23.9% $200-$ % 11.4% 4.1% 9.3% 18.4% 14.1% 5.7% 11.2% $250-$ % 4.3% 1.2% 3.5% 12.3% 7.1% 2.0% 5.6% $300-$ % 2.1% 0.4% 1.6% 8.9% 4.1% 0.9% 3.3% 50

56 Low Income Private Rental Households All Households Sydney Melbourne Adelaide Australia Sydney Melbourne Adelaide Australia $350-$ % 1.0% 0.2% 0.8% 5.7% 2.2% 0.4% 2.0% $400-$ % 0.9% 0.2% 0.6% 5.3% 1.9% 0.4% 1.8% $500 and over 2.3% 1.8% 0.7% 1.3% 4.9% 2.7% 0.8% 2.3% Not Stated 0.0% 0.0% 0.0% 0.0% 3.8% 4.0% 3.6% 3.9% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Occupation Managers and Administrators 1.7% 1.4% 0.9% 1.2% 4.1% 3.9% 3.1% 4.1% Professionals 6.4% 6.3% 4.3% 4.2% 9.7% 9.5% 8.2% 8.1% Associate Professionals 5.3% 4.9% 3.8% 4.0% 5.4% 5.3% 5.2% 5.2% Tradespersons and Related Workers 6.0% 5.3% 4.9% 5.0% 5.1% 5.4% 5.3% 5.4% Advanced Clerical and Service Workers 1.5% 1.4% 1.0% 1.0% 2.1% 1.8% 1.5% 1.7% Intermediate Clerical, Sales and Service Workers Intermediate Production and Transport Workers Elementary Clerical, Sales and Service Workers 11.4% 11.0% 10.5% 9.3% 7.9% 7.8% 7.8% 7.3% 6.1% 5.3% 4.4% 4.5% 3.4% 3.7% 3.5% 3.6% 6.5% 6.5% 6.0% 6.0% 4.2% 4.5% 4.3% 4.2% Labourers and Related Workers 5.7% 5.5% 6.1% 6.3% 3.1% 3.4% 4.0% 3.8% Not Stated/Not Applicable 49.3% 52.4% 58.1% 58.4% 54.9% 54.7% 57.0% 56.7% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Industry Accommodation, Cafes and Restaurants 4.7% 4.0% 3.9% 4.4% 2.2% 1.9% 1.9% 2.2% Agriculture, Forestry and Fishing 0.3% 0.4% 0.5% 1.2% 0.3% 0.4% 0.5% 1.8% Communication Services 0.9% 0.9% 0.6% 0.6% 1.1% 1.0% 0.8% 0.8% Construction 2.9% 2.0% 1.7% 2.0% 3.2% 3.0% 2.5% 3.0% 51

57 Low Income Private Rental Households All Households Sydney Melbourne Adelaide Australia Sydney Melbourne Adelaide Australia Cultural and Recreational Service 1.9% 1.9% 1.4% 1.5% 1.3% 1.3% 1.0% 1.1% Education 2.3% 2.8% 2.2% 2.2% 2.9% 3.2% 3.2% 3.2% Electricity, Gas and Water Supply 0.1% 0.1% 0.1% 0.1% 0.3% 0.2% 0.3% 0.3% Finance and Insurance 1.9% 1.6% 1.0% 0.9% 2.8% 2.1% 1.6% 1.7% Government Administration and Defence 1.3% 0.9% 0.8% 1.0% 1.6% 1.3% 2.0% 2.0% Health and Community Services 4.9% 5.1% 5.6% 4.6% 4.1% 4.3% 5.4% 4.3% Manufacturing 6.5% 6.9% 6.1% 4.9% 5.6% 7.4% 6.7% 5.4% Mining 0.0% 0.0% 0.0% 0.1% 0.1% 0.1% 0.1% 0.4% Personal and Other Services 2.2% 1.9% 1.9% 1.8% 1.6% 1.5% 1.8% 1.6% Property and Business Services 6.2% 5.7% 4.3% 4.2% 6.6% 6.0% 4.7% 4.9% Retail Trade 7.9% 8.2% 7.7% 7.7% 6.1% 6.7% 6.5% 6.5% Transport and Storage 3.1% 2.2% 1.7% 1.9% 2.3% 1.8% 1.7% 1.9% Wholesale Trade 3.3% 2.8% 2.2% 2.3% 2.8% 2.8% 2.2% 2.3% Not Stated/Not Applicable 49.7% 52.6% 58.2% 58.7% 57.3% 56.8% 58.9% 59.0% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 52

58 6 CHANGES IN THE DISTRIBUTION OF LOW INCOME PRIVATE RENTERS IN Introduction One of the central research questions of this study was to establish whether the private renter low income households had changed locations between 1991 to Having established in Chapter 5 where the PRLIH sector was concentrated in 2001, this chapter presents an analysis of the changes in the number and location of low income private renters in the three case study cities over that decade. The data presented here comes from the 1991 census and the 2001 census value added file although the point at which a household is defined as a low income household differs from previous analyses (see Section 4.4 for a more detailed explanation). The data for Sydney and Melbourne are for local government areas (LGAs) while in Adelaide statistical local areas (SLAs) are used. Analysis at a finer spatial scale is not possible as the 1991 data are not available on a comparable post code or suburb basis. 6.2 Changes in the number and location of private renter low income households, Under the definition of low income used for this section of the research (see above), there were 106,937 PRLIHs in Sydney in 2001, up from 85,229 in 1991 (Table 6.1 and Figure 6.1). This represents an increase of 25%. In 2001, the highest number of PRLIHs were found in Canterbury, Wyong and Fairfield LGAs, followed by Blacktown and Gosford. Ten years earlier in 1991 Canterbury was also the LGA with the highest number of PRLIHs. However, the position in 1991 for other LGAs was rather different with a number of inner city areas recording substantial losses of PRLIHs between 1991 and By 2001, the numbers of PRLIHs in these LGAs had declined significantly. Over this period LGAs with the largest increase in PRLIHs included Blacktown (2,761), Wyong (2,252) and Penrith (1,800). LGAs with the largest decrease in PRLIHs included Waverley (-706), Leichhardt (-527) and Marrickville (-528). Proportionally, the largest increases in PRLIHs between 1991 and 2001 was recorded in Blacktown and Wyong, while the areas which recorded the largest proportional decrease were the inner and eastern LGAs of Marrickville, Randwick, Waverley and South Sydney. However, in a potentially significant counter trend, the numbers of PRLIH tripled in the City of Sydney over the decade. The impact of higher density residential urban renewal here has clearly led to greater opportunities for this group in the CBD area despite their low incomes. In Melbourne there were 93,487 PRLIHs in 2001, an increase of 14% from the figure of 81,694 in 1991 (Table 6.2 and Figure 6.2). In 2001, the LGAs with highest number of PRLIHs were Darebin (5,835), Moreland (5,612) and Greater Dandenong (5,193). In 1991, however, the two LGAs with the largest number of PRLIHs were Port Phillip and Glen Eira. The numbers in Glen Eira have changed little since 1991, but the inner city LGA of Port Phillip recorded the largest loss of PRLIHs over the period (-1,904). Proportionally, the loss in Port Phillip was also the highest. In 1991 the LGAs of Darebin and Moreland also had significant numbers of PRLIHs. Between 1991 and 2001, the areas with the largest increase in the number of PRLIHs in Melbourne were Casey, Frankston and the City of Melbourne itself, the latter again paralleling the trend in Sydney towards an increase in housing opportunities in higher density central city residential development. These three LGAs also had the highest proportional increases. Conversely, other than Port Phillip, the LGAs with significant declines in the number of PRLIHs were recorded in Stonnington and Yarra. These LGAs along with Wyndham also had the largest proportional losses of this group. Figure 6.2 highlights the relative shift of this group away from inner city areas immediately surrounding the CBD, especially in the gentrifying suburbs to the east of the city. In Adelaide there were 35,803 PRLIHs in 2001, up from 29,558 in 1991 (Table 6.3 and Figure 6.3), an increase of 21%. In 1991, the SLAs with the largest number of PRLIHs were West Torrens East (1,899), Holdfast Bay North (1,472) and Norwood Payneham St Peters West (1,267). These were closely followed by West Torrens West, Campbelltown West and Charles Sturt Coastal. In 2001, West Torrens East also had the largest number of 53

59 PRLIHs (1,817) and this was followed by Holdfast Bay North (1,226). In 2001 West Torrens West had the third largest number of PRLIHs. Between 1991 and 2001 though the largest increase in Adelaide were in SLAs with lower numbers of PRLIHs in This is slightly different from the situation in Sydney and Melbourne. For example, Playford Elizabeth, Onkaparinga South Coast and Playford West Central all had the largest increase in the number of PRLIHs between 1991 and These areas had relatively low numbers of PRLIHs in Conversely, it was the areas with higher numbers of PRLIHs in 1991 that recorded the largest declines. The areas which recorded the largest declines between 1991 and 2001 were Norwood Payneham St Peters West and Holdfast Bay North and Unley East. The changes in absolute numbers in Adelaide between 1991 and 2001 also hold true proportionally except in the case of West Torrens East. West Torrens East had the highest number of PRLIHs in Adelaide in both 1991 and 2001, however, the number has remained relatively stable over the ten year period. Due to a 21% increase across the Adelaide metropolitan area, particularly in areas around Onkaparinga and Playford, the proportion of PRLIHs in West Torrens East has actually declined. Figure 6.3 highlights the relative shift away from the central and inner band of Adelaide, with relative increases in the suburbs to the north and south of the metropolitan area. The extent to which these changes are, in part, a product of sales of public housing in some of these areas, such as Salisbury and Elizabeth, is worthy of further research. Most notably, there is no evidence of a counter trend of a higher density led increase in PRLIH in Adelaide s CBD. 6.3 Summary Overall, then, the numbers of households defined as PRLIH increased in all three cities between 1991 and 2001, by between 14% and 25%, with the greatest increase in Sydney. Within this overall increase in numbers, the distributions of PRLIH within the three cities have shown broadly similar trends. The greatest losses of this group were found in inner city and gentrified areas, some of which had significant concentrations of this group in The greatest increases were found in middle or outer city LGAs. On the other hand, several LGAs retained their position of high concentrations in both years: for example, Canterbury and Fairfield in Sydney, Darebin and Moreland in Melbourne, and West Torrens West in Adelaide. There can be little doubt that the escalation of house prices in the inner and high value areas of all three cities have acted to push this group out of the central areas over this period, while the boom in investor activity across the city into lower value locations has acted as the key driver in this relocation of the PRLIH population. The most notable exceptions to this trend have been the increases in PRLIHs in the CBDs of Sydney City and Melbourne. Here, the significant increase in higher density residential construction during the under urban consolidation policies decade has clearly resulted in an increase in rental opportunities for this group, despite their relatively low incomes. No such trend was in evident in the case of Adelaide s CBD. 54

60 Table 6.1 The number of PRLIHs by LGA in Sydney in Number of PRLIHs 1991 Number of PRLIHs 2001 Absolute Change Ashfield 1,935 1, % Auburn 1,436 2, % Bankstown 1,874 3,214 1, % Baulkham Hills % Blacktown 2,448 5,209 2, % Blue Mountains 1,528 2, % Botany Bay 1,037 1, % Burwood 886 1, % Camden % Campbelltown 1,480 3,084 1, % Canterbury 4,788 6,105 1, % Concord % Drummoyne % Fairfield 4,156 5,543 1, % Gosford 3,647 5,134 1, % Hawkesbury 1,046 1, % Holroyd 1,725 2,759 1, % Hornsby 1,919 2, % Hunter's Hill % Hurstville 1,310 1, % Kogarah 890 1, % Ku-ring-gai % Lane Cove % Leichhardt 2,635 2, % Liverpool 2,234 3,840 1, % Manly 1, % Marrickville 4,391 3, % Mosman % North Sydney 2,095 1, % Parramatta 3,041 4,344 1, % Penrith 2,559 4,359 1, % Pittwater % Randwick 4,529 4, % Rockdale 2,110 2, % Ryde 2,424 2, % South Sydney 4,749 4, % Strathfield % Sutherland Shire 2,409 3, % Sydney 419 1, % Warringah 2,617 2, % Waverley 3,149 2, % Willoughby 972 1, % Wollondilly % Woollahra 2,058 1, % Wyong 3,495 5,747 2, % Total 85, ,937 21,708 Percentage Point Change 55

61 Table 6.2: The number of PRLIHs by LGA in Melbourne in Number of PRLIHs 1991 Number of PRLIHs 2001 Absolute Change Banyule 1,828 2, % Bayside 1,557 1, % Boroondara 4,288 3, % Brimbank 2,437 3,612 1, % Cardinia % Casey 1,578 3,139 1, % Darebin 5,103 5, % Frankston 2,398 3,740 1, % Glen Eira 5,148 4, % Greater Dandenong 4,123 5,193 1, % Hobsons Bay 1,743 2, % Hume 1,199 2, % Kingston 3,530 4, % Knox 1,664 2, % Manningham 1,101 1, % Maribyrnong 2,740 2, % Maroondah 1,935 2, % Melbourne 1,831 3,149 1, % Melton % Monash 2,745 3, % Moonee Valley 2,998 2, % Moreland 5,014 5, % Mornington Peninsula 2,681 3, % Nillumbik % Port Phillip 7,001 5,097-1, % Stonnington 4,559 3, % Whitehorse 2,786 3, % Whittlesea 1,362 2, % Wyndham 918 1, % Yarra 3,614 2, % Yarra Ranges 2,243 2, % Total 81,694 94,469 11,793 Percentage Point Change 56

62 Table 6.3: The number of PRLIHs by SLAs in Adelaide in Number of PRLIHs 1991 Number of PRLIHs 2001 Absolute Change Adelaide (C) % Adelaide Hills (DC) - Central % Adelaide Hills (DC) - Ranges % Burnside (C) - North-East % Burnside (C) - South-West % Campbelltown (C) - East % Campbelltown (C) - West 1, % Charles Sturt (C) - Coastal 1, % Charles Sturt (C) - Inner East % Charles Sturt (C) - Inner West % Charles Sturt (C) - North-East 945 1, % Gawler (M) % Holdfast Bay (C) - North 1,472 1, % Holdfast Bay (C) - South % Marion (C) - Central % Marion (C) - North % Marion (C) - South % Mitcham (C) - Hills % Mitcham (C) - North-East % Mitcham (C) - West % Norw. P'ham St Ptrs (C) - East % Norw. P'ham St Ptrs (C) - West 1, % Onkaparinga (C) - Hackham % Onkaparinga (C) - Hills % Onkaparinga (C) - Morphett % Onkaparinga (C) - North Coast % Onkaparinga (C) - Reservoir % Onkaparinga (C) - South Coast % Onkaparinga (C) - Woodcroft % Playford (C) - East Central % Playford (C) - Elizabeth 292 1, % Playford (C) - Hills % Playford (C) - West % Playford (C) - West Central % Port Adel. Enfield (C) - Coast % Port Adel. Enfield (C) - East 835 1, % Port Adel. Enfield (C) - Inner % Port Adel. Enfield (C) - Port % Prospect (C) % Salisbury (C) - Central % Salisbury (C) - Inner North % Salisbury (C) - North-East % Salisbury (C) - South-East % Salisbury (C) Bal % Tea Tree Gully (C) - Central % Tea Tree Gully (C) - Hills % Tea Tree Gully (C) - North % Tea Tree Gully (C) - South % Unley (C) - East % Unley (C) - West % Walkerville (M) % West Torrens (C) - East 1,899 1, % West Torrens (C) - West 1,140 1, % Total 29,558 35,803 6,245 Percentage Point Change 57

63 Figure 6.1: The percentage point change in PRLIHs in Sydney by LGA in

64 Figure 6.2: The percentage point change in PRLIHs in Melbourne by LGA in

65 Figure 6.3: The percentage point change in PRLIHs in Adelaide by SLA in

66 7 AN EXAMINATION OF THE LOW COST PRIVATE RENTAL STOCK 7.1 Introduction This chapter examines the locational characteristics of the lower cost private rental stock in Sydney, Melbourne and Adelaide. This in part addresses the question of the nature of the relationship between the location of private renter low income households and the location of the low cost rental stock (see also Section 8.3). The chapter is in four parts. Section 7.2 below supplements the work done by Yates et al (2003) and examines the spatial concentration of the very low cost rental stock (defined as being privately rented dwellings with rents in the bottom 15% of all privately rented dwellings for each city) for local government areas (LGAs) in the three cities. Section 7.3 then analyses the spatial concentration of the low cost rent stock at the LGA level. The low cost rental stock is defined here as being in the bottom 40% of all rental stock (in terms of weekly rent) in each of the three cities. The 40 th percentile was taken as it approximates the same percentage of all households that is equivalent to the bottom income quintile of employed low income private renters that has been used throughout this report to define a low income working household (see Tables 4.4 to 4.6) 8. Finally, Sections 7.4 and 7.5 examine both the very low cost private rental stock and the low cost rental stock at the postcode level. The total numbers of very low cost and low cost dwellings identified for each of the three cities is shown in Table 7.1, together with the total number of private rented dwellings. Table 7.1: The number of very low cost and low cost private rental stock in Sydney, Melbourne and Adelaide, 2001 Very low cost rental dwellings Low cost rental dwellings All private rental dwellings Sydney 47, , ,331 Melbourne 33,936 94, ,212 Adelaide 9,846 29,091 65, An analysis of very low cost private rental stock LGA level Of the 312,331 private rented dwellings in Sydney in 2001, 47,907 were identified as very low cost for the purposes of this study (i.e. where the rent was approximately at or below the 15 th percentile of all rents). Three local government areas had significantly higher numbers of very low cost private rental dwellings than other LGAs (Table 7.2 and Figure 7.1). These three included Fairfield (4,399), Wyong (4,266) and Canterbury (4,073). As a proportion of all the private rental stock in an LGA, Wyong and Fairfield had the highest proportions of very low cost stock (45% and 43% respectively) (Figure 7.2). Canterbury LGA had the fourth highest proportion. Although Wollondilly on Sydney s south-western fringe has low numbers of private rental stock, a large proportion is of very low cost (39%). All LGAs in Sydney with high numbers of very low cost private rental stock were located in the inner west and western parts of the metropolitan area. In Melbourne, 33,936 privately rented dwellings out of a total of 224,212 were identified as being very low cost. Greater Dandenong had the highest number of very low cost private rental dwellings (3,103) (Table 7.2 and Figure 7.3). This was followed by Darebin (2,642), Moreland (2,346) and Maribyrnong (1,966). Maribyrnong and Greater Dandenong also had the highest proportion of their stock in the very low cost range (35% and 34% respectively) (Figure 7.4), and these two were the only LGAs in Melbourne, which had more than 30% of their stock in the very low cost range. Conversely, Sydney had six LGAs where more than 30% of their stock that was classified as very low cost stock. In Melbourne the LGAs with the 8 In practice, the actual figures used to define the low income stock varied slightly above the 40% threshold due to the data boundaries used by the ABS. The actual figures were 43.7% for Sydney, 42.1% for Melbourne and 44.3% for Adelaide. 61

67 largest number of very low cost private rental stock were located in the south-east and western parts of the city. Of the 65,538 private rental dwellings in Adelaide, 9,846 were classified as having very low rents. West Torrens East had the largest number of very low cost private rental dwellings in Adelaide (780) (Table 7.2 and Figure 7.5). This was followed by West Torrens West (597) and Port Adelaide Enfield Port (543). Port Adelaide Enfield Port had the highest proportion of very low rent stock (39%) (Figure 7.6). This was the only SLA in which 30% or more of the private rental stock was very low cost. This was followed by West Torrens- West (27%) and Port Adelaide Enfield Inner (25%). Generally speaking the very low cost private rental stock in Adelaide is concentrated in the west and north of the metropolitan area. Table 7.2: The location of the very low cost private rental stock in Sydney, Melbourne and Adelaide (LGAs/SLAs with the top ten numbers and proportion), 2001 Number of Dwellings Classified as Very Low Cost Private Rental Stock Sydney Fairfield 4,399 Wyong 45% Wyong 4,266 Fairfield 43% Canterbury 4,073 Wollondilly 39% Gosford 2,949 Canterbury 34% Penrith 2,873 Blue Mountains 33% Blacktown 2,600 Campbelltown 32% Liverpool 2,506 Liverpool 29% Campbelltown 2,235 Hawkesbury 29% Marrickville 2,083 Penrith 28% Parramatta 1,618 Gosford 28% Melbourne Greater Dandenong 3,103 Maribyrnong 35% Darebin 2,642 Greater Dandenong 34% Moreland 2,346 Mornington Peninsula 28% Maribyrnong 1,966 Brimbank 25% Glen Eira 1,850 Frankston 24% Mornington Peninsula 1,799 Yarra Ranges 24% Kingston 1,765 Cardinia 23% Port Phillip 1,662 Darebin 23% Frankston 1,651 Kingston 20% Brimbank 1,614 Moreland 19% Adelaide West Torrens - East 780 Port Adel. Enfield - Port 39% West Torrens - West 597 West Torrens - West 27% Port Adel. Enfield - Port 543 Port Adel. Enfield - Inner 25% Holdfast Bay - North 368 West Torrens - East 25% Port Adel. Enfield - East 353 Salisbury - Central 23% Charles Sturt - Coastal 333 Playford - West 22% Port Adel. Enfield - Inner 332 Charles Sturt - Inner West 21% Charles Sturt - North-East 319 Adelaide Hills - Ranges 20% Salisbury - Central 317 Port Adel. Enfield - Coast 20% Port Adel. Enfield - Coast 306 Charles Sturt - Inner East 19% Proportion of an area s stock classified as very low cost private rental stock 62

68 Figure 7.1: The location of very low cost private rental stock in Sydney LGAs in

69 Figure 7.2: The proportion of private rental stock in each LGA which is very low cost in Sydney in

70 Figure 7.3: The location of very low cost private rental stock in Melbourne LGAs in

71 Figure 7.4: The proportion of private rental stock in each LGA which is very low cost in Melbourne in

72 Figure 7.5: The location of very low cost private rental stock in Adelaide SLAs in

73 Figure 7.6: The proportion of the private rental stock in each SLA which is very low cost in Adelaide in

74 7.3 An analysis of low cost private rental stock LGA level This section presents an analysis of the low cost private rental stock in Sydney, Melbourne and Adelaide, i.e. dwellings with in lowest 40% in terms of cost. The figures presented here are generally reflective of the results presented above (section 7.2). Overall, Sydney (136,498) had the greatest number of low cost private rental dwellings, followed by Melbourne (94,469) and Adelaide (29,091). However, there are significant concentrations of this stock. In fact, 13 LGAs in Sydney recorded over 60% of the private rental stock as being low or very low cost. This compares to nine LGAs in Melbourne and six SLAs in Adelaide. This suggests a higher degree of polarisation in the lower cost rental market in Sydney and Melbourne compared with Adelaide. In Sydney, Blacktown LGA has the largest number of low cost private rental dwellings (9,772) (Table 7.3 and Figure 7.7). This is followed by Canterbury (8,551), Wyong (8,385) and Fairfield (8,351), all of which have a large number of very low cost private rental stock. Interestingly, Canterbury, Fairfield and Wyong all have a significant proportion of low cost stock as a well as very low cost stock whereas Blacktown has much less very low cost private rental dwellings (see above) (Figure 7.8). Further, although Campbelltown has 6,259 low cost private rental dwellings, this is 89% of all the private rental stock in Campbelltown. This is the highest proportion of any LGA in Sydney. This is followed by Wyong (87%), Wollondilly (87%) and Fairfield (82%). Most of the low cost private rental stock in Sydney is concentrated in the western suburbs of Sydney and in the Central Coast to the north (Gosford and Wyong LGAs). There are few low cost private rental dwellings in the inner city and northern suburbs of Sydney. In Melbourne, Greater Dandenong has the largest number of low cost private rental dwellings (7,265) (Table 7.3 and Figure 7.9). There is also a large gap to the next largest LGAs Darebin (6,230) and Moreland (5,996). Greater Dandenong also has the highest proportion of low cost private rental stock in Melbourne (80%) (Figure 7.10). This is followed by Cardinia (73%) and Melton (72%). Generally speaking the largest numbers of low cost private rental stock in Melbourne are located in the south east and western suburbs of the city and to the immediate north of the centre (e.g. Darebin). A significant amount of the low cost private rental stock in Adelaide is concentrated in the north and western parts of the metropolitan area, similar to the very low cost stock (Table 7.3 and Figure 7.11). West Torrens East and West Torrens West have the highest number of these dwellings (1,709 and 1,197 respectively), followed by Playford Elizabeth (1,154) and Holdfast Bay North (1,115). However, as a proportion of total private rental stock, Playford Elizabeth and Playford West Central have the largest proportions of low cost stock (83% and 80% respectively) (Figure 7.12). There is then a large gap to Port Adelaide Enfield Port (68%) and Salisbury Central (62%). 7.4 Summary This above analysis indicates the following: Sydney had the by far the largest low and very low cost rental stock of the three cities. The proportion of the total rental stock that was classified as low cost reached a higher percentage in some Sydney LGAs compared to Melbourne or Adelaide. The very low cost stock is likely to be numerically concentrated in a more limited number of districts in all three cities. These tend to be either nearer the city centre compared to the low cost stock, or in more distant suburban locations on the metropolitan periphery, such as Wyong in Sydney, the Mornington Peninsula in Melbourne or Playford in Adelaide. The low cost stock is more likely to be concentrated in the outer metropolitan areas than the very low cost stock, and forms a higher proportion of the rental stock in outer districts. 69

75 Sydney Table 7.3: The location of the low cost private rental stock in Sydney, Melbourne and Adelaide (LGAs/SLAs with the top ten numbers and proportion), 2001 Number of Dwellings Classified as Low Cost Private Rental Stock Blacktown 9,772 Campbelltown 89% Canterbury 8,551 Wyong 87% Wyong 8,385 Wollondilly 87% Fairfield 8,351 Fairfield 82% Penrith 8,135 Hawkesbury 80% Gosford 7,820 Penrith 80% Parramatta 6,468 Blue Mountains 79% Campbelltown 6,259 Gosford 74% Liverpool 5,862 Blacktown 74% Marrickville 5,309 Canterbury 71% Melbourne Greater Dandenong 7,265 Greater Dandenong 80% Darebin 6,230 Cardinia 73% Moreland 5,996 Melton 72% Frankston 4,683 Frankston 69% Glen Eira 4,615 Mornington Peninsula 69% Port Phillip 4,592 Brimbank 65% Kingston 4,472 Yarra Ranges 62% Mornington Peninsula 4,418 Maribyrnong 62% Brimbank 4,242 Wyndham 60% Casey 3,627 Darebin 53% Adelaide West Torrens - East 1,709 Playford - Elizabeth 83% West Torrens - West 1,197 Playford - West Central 80% Playford - Elizabeth 1,154 Port Adel. Enfield - Port 68% Holdfast Bay - North 1,115 Salisbury - Central 62% Port Adel. Enfield - East 1,001 Onkaparinga - North Coast 61% Port Adel. Enfield - Port 954 Playford - West 60% Norw. P'ham St Ptrs - West 915 Gawler 56% Charles Sturt - Coastal 894 Port Adel. Enfield - Inner 56% Salisbury - Central 861 West Torrens - East 54% Charles Sturt - North-East 860 West Torrens - West 54% Proportion of an area s stock classified as low cost private rental stock 70

76 Figure 7.7: The location of low cost private rental stock in Sydney in

77 Figure 7.8: The proportion of private rental stock in each LGA which is low cost stock in Sydney in

78 Figure 7.9 The location of low cost private rental stock in Melbourne in

79 Figure 7.10: The proportion of private rental stock in each LGA which is low cost stock in Melbourne in

80 Figure 7.11: The location of low cost private rental stock in Adelaide in

81 Figure 7.12: The proportion of private rental stock in each SLA which is low cost stock in Adelaide in

82 7.5 An analysis of very low cost private rental stock postcode level The postcode analysis of the low cost rental stock reflects that of the LGA analysis presented above, but again confirms the more localised nature of the low cost rental sub-markets revealed in the analysis in chapter 5. In Sydney in 2001 the postcodes with the highest number of very low cost private rental dwellings were located in the west (particularly the central west), south-west and the Central Coast to the north (see Appendix 2). The postcodes with the highest number of very low cost private rental dwellings were 2166 (Cabramatta/Canley Vale) with 2,426 dwellings, postcode 2170 (Liverpool) with 2,144 dwellings and postcode 2195 (Wiley Park/Lakemba) with 1,731 dwellings (Table 7.3). There were few very low cost private rental dwellings in the inner city and northern parts of the Sydney metropolitan area. In Melbourne the very low cost private rental housing was concentrated in postcodes in the south-east, west and inner north of the metropolitan area. The postcodes with the highest concentrations of very low cost private rental dwellings were 3175 (Dandenong) with 1,621 dwellings, postcode 3163 (Carnegie/Glen Huntly/Murrumbeena) with 1,078 dwellings and postcode 3199 (Frankston) with 1,029 dwellings. There are relatively few very low cost private rental dwellings in Melbourne in the east and middle-outer northern suburbs of the city. In Adelaide the very low cost private rental housing is concentrated to the west of the CBD and in the north of the city. In 2001 the postcodes with the highest number of very low cost private rental dwellings were 5038 (Camden Park/Plympton) with 433 dwellings, 5108 (Salisbury) with 351 dwellings and postcode 5032 (Brooklyn Park/Lockleys/Underdale) with 314 dwellings. The concentration of very low cost private rental housing in Adelaide seems to be more dispersed than that of Melbourne or Sydney, although there are few very low cost private rental dwellings in the outer east and southern areas of the city. This section has not detailed the proportions of very low cost rental stock in each of the postcodes in Sydney, Melbourne and Adelaide. This is because the proportions become skewed in the outer fringe postcode areas in each of the three cities (see Appendix 2). That is, in the outer fringe areas there are numerically few private rental dwellings and therefore that stock which is classified as very low cost tends to be a significant proportion of the private rental stock, even though they are few in number. Thus any proportional analysis shows that there are higher concentrations of very low cost private rental housing on the fringes of the three cities, however, there is nowhere as much stock in absolute terms as in other postcode areas. This also applies to section 7.6 below. 77

83 Table 7.4: The location of the very low cost private rental stock in Sydney, Melbourne and Adelaide (postcodes with the top ten numbers), 2001 Number of Dwellings Suburb(s) Classified as Very Low Cost Private Rental Stock Sydney 2166 Cabramatta, Cabramatta West, Canley Heights, Canley Vale, 2, Casula, Chipping Norton, Hammondville, Liverpool, Liverpool South, 2,144 Lurnea, Moorebank, Mount Pritchard, Prestons, Warwick Farm 2195 Lakemba, Wiley Park 1, Bateau Bay, Berkeley Vale, Blue Bay, Chittaway Bay, Chittaway Point, Gleening Valley, Killarney Vale, Long Jetty, Shelly Beach, The 1,456 Entrance, Tumbi Umbi Airds, Ambarvale, Appin, Blair Athol, Bradbury, Campbelltown, Englorie Park, Gilead, Glen Alpine, Leumeah, Rosemeadow, Ruse, 1,188 St Helens Park, Woodbine 2165 Fairfield, Fairfield East, Fairfield Heights, Fairfield West 1, Bidwill, Blackett, Dharruk, Emerton, Lethbridge Park, Minchinbury, 1,110 Mount Druitt, Shalvey, Tregear, Whalan, Willmot 2263 Charmhaven, Gorokan Lake, Lake Haven, Norah Head, Toukley 1, Gwandalan, Kanwal, Lake Munmorah, Mannering Park, Rocky 1,028 Point, South Tacoma, Summerland Point, Tacoma, Tuggerah 2750 Emu Heights, Emu Plains, Jamisontown, Leonay, Penrith, Penrith 1,028 Melbourne 3175 Bangholme, Dandenong, Dandenong East, Dandenong North, 1,621 Dandenong South, Dunearn 3163 Carnegie, Glen Huntly, Murrumbeena 1, Frankston, Frankston East, Frankston Heights, Frankston South, 1, Noble Park, Noble Park East, Noble Park North Footscray, Seddon, Seddon West Albion, Glengala, Sunshine, Sunshine North, Sunshine West Brooklyn, Kingsville, Kingsville West, Maidstone, Tottenham, West 764 Footscray 3071 Thornbury Brunswick, Brunswick Lower, Brunswick North St Kilda, St Kilda South, St Kilda West 619 Adelaide 5038 Camden Park, Plympton, Plympton Park, South Plympton Paralowie, Salisbury, Salisbury Downs, Salisbury North Brooklyn Park, Lockleys, Underdale Glenelg, Glenelg East, Glenelg North, Glenelg South Findon, Seaton, Seaton North Glandore, Kurralta Park, Netley, North Plympton Albert Park, Alberton, Cheltenham, Hendon, Queenstown, Royal Felixstow, Firle, Glynde, Joslin, Marden, Payneham, Payneham 229 South, Royston Park 5082 Fitzroy, Ovingham, Prospect, Prospect East, Prospect West, Cowandilla, Hilton Plaza, Marleston, Richmond, West Richmond

84 7.6 An analysis of low cost private rental stock postcode level The results presented above in section 7.5 are similar to those presented in this section. This is not surprising given that the low cost stock also includes the very low cost private rental stock. In 2001, the low cost rental stock in Sydney is concentrated in the west, south west and Central Coast of the metropolitan area. There is a particularly high concentration of low cost private rental dwellings in the central west (see Appendix 2). In 2001, the postcode areas with the highest number of low cost private rental dwellings were 2170 (Liverpool) with 4,611 dwellings, 2166 (Cabramatta/Canley Vale) with 3,428 dwellings and postcode 2560 (Campbelltown) with 3,156 dwellings (Table 7.4). This was closely followed by two postcode areas on the Central Coast (2261 and 2250). In Melbourne, as noted above, the largest concentrations of low cost private rental housing are in the south east, west and inner northern areas of the city. In 2001 the postcode areas with the largest number of low cost private rental housing were 3175 (Dandenong) with 3,240 dwellings, 3163 (Carnegie/Glen Huntly/Murrumbeena) with 2,531 dwellings, and postcode 3199 (Frankston) with 2,513 dwellings. This is closely followed by postcode 3174 (Noble Park) in the south east and 3073 (Reservoir) in the inner north of the metropolitan area. In Adelaide, the largest number of low cost private rental housing is located in the north and western parts of the metropolitan area. In 2001 postcode 5108 (Salisbury) had the highest number of low cost private rental housing with 1215 dwellings. This was significantly higher than the next two highest areas 5113 (Elizabeth Downs/Elizabeth North/Elizabeth West) with 971 dwellings and postcode 5045 (Glenelg) with 942 low cost private rental dwellings. This was followed by 5038 (Camden Park/Plympton) with 828 dwellings and 5070 (Payneham) with 679 dwellings. Table 7.5: The location of the low cost private rental stock in Sydney, Melbourne and Adelaide (postcodes with the top ten numbers), 2001 Sydney Suburb(s) Casula, Chipping Norton, Hammondville, Liverpool, Liverpool South, 4,611 Lurnea, Moorebank, Mount Pritchard, Prestons, Warwick Farm Cabramatta, Cabramatta West, Canley Heights, Canley Vale, 3,428 Lansvale Number of Dwellings Classified as Low Cost Private Rental Stock Airds, Ambarvale, Appin, Blair Athol, Bradbury, Campbelltown, Englorie Park, Gilead, Glen Alpine, Leumeah, Rosemeadow, Ruse, 3,156 St Helens Park, Woodbine Bateau Bay, Berkeley Vale, Blue Bay, Chittaway Bay, Chittaway Point, Gleening Valley, Killarney Vale, Long Jetty, Shelly Beach, The 3,010 Entrance, Tumbi Umbi Calga, East Gosford, Erina, Gosford, Gosford South, Gosford West, Lisarow, Mangrove Mountain, Mooney Mooney, Mount White, 2,991 Narara, North Gosford 2195 Lakemba, Wiley Park 2, Blacktown, Huntingwood, Kings Park, Marayong, Prospect 2, Harris Park, Parramatta 2, Emu Heights, Emu Plains, Jamisontown, Leonay, Penrith, Penrith 2,617 South Girraween, Greystanes, Mays Hill, Pendle Hill, Wentworthville, 2,583 Westmead 79

85 Melbourne 3175 Suburb(s) Bangholme, Dandenong, Dandenong East, Dandenong North, 3,240 Dandenong South, Dunearn 3163 Carnegie, Glen Huntly, Murrumbeena 2, Frankston, Frankston East, Frankston Heights, Frankston South, 2,513 Karingal 3174 Noble Park, Noble Park East, Noble Park North 2, Keon Park, Reservoir, Reservoir East, Reservoir North, Reservoir 2,044 South 3021 Albanvale, Keabla, Kings Park, St Albans 1, Albion, Glengala, Sunshine, Sunshine North, Sunshine West 1, St Kilda, St Kilda South, St Kilda West 1, Brunswick, Brunswick Lower, Brunswick North 1, Thornbury 1,371 Adelaide 5108 Paralowie, Salisbury, Salisbury Downs, Salisbury North 1, Davoren Park, Elizabeth Downs, Elizabeth North, Elizabeth Park, 971 Elizabeth West 5045 Glenelg, Glenelg East, Glenelg North, Glenelg South Camden Park, Plympton, Plympton Park, South Plympton Felixstow, Firle, Glynde, Joslin, Marden, Payneham, Payneham 679 South, Royston Park 5032 Brooklyn Park, Lockleys, Underdale Elizabeth, Elizabeth East, Elizabeth South, Hillbank Morphett Vale, Woodcroft Fitzroy, Ovingham, Prospect, Prospect East, Prospect West, 553 Thorngate 5022 Grange, Henley Beach, Henley Beach South, Kirkcaldy, Tennyson 541 Number of Dwellings Classified as Low Cost Private Rental Stock 80

86 8 THE RELATIONSHIP BETWEEN LOW INCOME PRIVATE RENTERS AND HOUSING ASSISTANCE In this section the relationship between Commonwealth Rent Assistance, public housing waiting lists and low income private renters in 2001 is examined. In particular, the correlation between these three variables is analysed, together with the location of the low cost private rental stock. This chapter therefore addresses the third substantive research question, namely whether the spatial concentration of Commonwealth Rent Assistance (RA) recipients and households on the public housing waiting list are more or less concentrated than the spatial distribution of lower income private renters who are in employment. The primary form of housing assistance for tenants in the private rented sector is Commonwealth Rent Assistance. This form of assistance is paid to benefits and pensions recipients only. While this includes some lower income working households with children, the majority of working households who rent privately are unlikely to qualify for this form of assistance. It is also well understood that the majority of waiting list applicants live in private rental housing. Eligibility rules now more or less exclude all low income working households from accessing this second form of housing assistance. It is instructive to see how far low income households who rent privately are associated with RA recipients and those on the waiting list for public housing assistance in order to judge how far these three groups share similar housing markets, or whether the market in effect segregates these into distinctive demand segments. The final component of this section also examines whether the concentration of Commonwealth Rent Assistance recipients has changed between March 2000 and March 2005 (the longest time series available). Before this we begin with an analysis of Commonwealth Rent Assistance payments in Commonwealth Rent Assistance in 2001 In 2001, there were just over 54,000 recipients of Commonwealth Rent Assistance payments in Adelaide, 155,000 in Melbourne and 189,000 in Sydney 9 (Tables 8.1 to 8.3). In Adelaide, 38 per cent of recipients were from 20 postcode areas. In Melbourne and Sydney, 29 per cent and 34 per cent of Commonwealth Rent Assistance recipients lived in 20 postcode areas. In total just over $26 million was paid in Commonwealth Rent Assistance payments per fortnight to recipients in Adelaide, Melbourne and Sydney in This $26 million was made up of $3 million in Adelaide, $10 million in Melbourne and $13 million in Sydney. In Adelaide, there were 7 postcode areas in 2001 that contained over 1,000 Commonwealth Rent Assistance recipients (Figure 8.5). Postcode 5108 (around Salisbury) contained the largest number of recipients at 2,190. Of these seven postcode areas, 4 were concentrated in the far north of Adelaide CBD, while the other 3 were to the immediate north, south west and far south of the CBD. In comparison with Adelaide, Melbourne contained 45 postcode areas in which there were over 1,000 Commonwealth Rent Assistance recipients in 2001 (Figure 8.4). In fact, there were 9 postcode areas in which there were over 2,000 recipients. The highest number of recipients was concentrated in postcode 3175 (around Dandenong) of which there were 4,520 recipients in The postcode areas of which there were over 1,000 recipients were geographically spread across the Melbourne metropolitan area. Those areas with over 2,000 recipients were concentrated in Brimbank to the west of Melbourne CBD, north of the CBD around Darebin, and Dandenong and Mornington Peninsula in the south-east. In contrast to both Melbourne and Adelaide, Commonwealth Rent Assistance recipients in Sydney are concentrated in a large number of postcode areas. While this may be related to the populations of postcode areas in Sydney, in 2001, 57 postcode areas in Sydney each contained over 1,000 Commonwealth Rent Assistance recipients (Figure 8.3). However, there were also 24 postcode areas in Sydney with over 2,000 recipients and 2 areas in which there were over 5,000 recipients. Postcode 2166 (around Cabramatta) had the highest 9 These figures exclude postcode areas for which there are less than 20 recipients. This included 11 postcode areas in Adelaide, 18 in Melbourne and 5 in Sydney. 81

87 number of Commonwealth Rent Assistance recipients in 2001 with no less than 5,859 recipients. The postcode areas with over 1,000 recipients were geographically spread across Sydney, although there were none to the immediate north of the Sydney CBD. The highest number of recipients in Sydney were located in the south-west (Fairfield-Liverpool- Campbelltown), west (Parramatta to Penrith) and Gosford and Wyong in the north. The distribution of the amount of Commonwealth Rent Assistance paid to recipients on a fortnightly basis was also different in Sydney compared to both Melbourne and Adelaide (Figures 8.6 to 8.8). To begin with, average RA in Sydney was generally higher, reflecting the generally higher rent levels here. However, the highest amount of assistance per recipient in Melbourne and Adelaide was found in areas located on the fringe postcodes. In contrast, while there were fringe areas in Sydney where per recipient payments were high, areas of high average payments were found across the city in both the middle and inner suburbs. Nevertheless, it is worth noting that the average amounts paid do not vary greatly between the three cities. For example, average payments per suburb range from $62 to $76 in Sydney, $59 to $69 in Melbourne and $54 to $72 in Adelaide. This finding also holds within each city, as shown in Figures 8.6 to 8.8. The flat rate nature of Commonwealth Rent Assistance means that recipients receive broadly similar amount regardless of location. This detailed spatial analysis therefore might reflect what some have regarded to be a shortcoming of the Commonwealth Rent Assistance system, namely the failure to compensate for rent differentials across and between regions. 82

88 Table 8.1: Number of Commonwealth Rent AssistanceCommonwealth Commonwealth Rent Assistance Recipients, 2001 Top 20 Postcodes in Sydney Postcode Suburb(s) Total Fortnightly Amount of CRA Total Number of CRA Recipients Total Amount of CRA per Fortnight per Recipient Cabramatta, Cabramatta West, Canley Heights, Canley $364,164 5,859 $62.15 Vale, Lansvale Casula, Chipping Norton, Hammondville, Liverpool, Liverpool South, Lurnea, Moorebank, Mount Pritchard, $368,433 5,388 $68.38 Prestons, Warwick Farm Bateau Bay, Berkeley Vale, Blue Bay, Chittaway Bay, Chittaway Point, Gleening Valley, Killarney Vale, Long $273,045 3,907 $69.89 Jetty, Shelly Beach, The Entrance, Tumbi Umbi 2165 Fairfield, Fairfield East, Fairfield Heights, Fairfield West $253,030 3,789 $ Gwandalan, Kanwal, Lake Munmorah, Mannering Park, Rocky Point, South Tacoma, Summerland Point, Tacoma, $233,383 3,776 $61.81 Tuggerah Calga, Central Mangrove, East Gosford, Erina, Gosford, Gosford South, Gosford West, Lisarow, Mangrove $235,278 Mountain, Mooney Mooney, Mount White, Narara, Niagara 3,584 $65.65 Park, North Gosford Girraween, Greystanes, Mays Hill, Pendle Hill, South $224,511 3,202 $70.12 Wentworthville, Wentworthville, Westmead Bidwill, Blackett, Dharruk, Emerton, Lethbridge Park, Minchinbury, Mount Druitt, Shalvey, Tregear, Whalan, $215,491 3,189 $67.57 Willmot Airds, Ambarvale, Appin, Blair Athol, Bradbury, Campbelltown, Englorie Park, Gilead, Glen Alpine, $217,878 3,181 $68.49 Leumeah, Rosemeadow, Ruse, St Helens Park, Woodbine 2200 Bankstown, Bankstown North, Condell Park, Mount Lewis $211,757 2,970 $ Arndell Park, Blacktown, Huntingwood, Kings Park, $199,704 2,933 $68.09 Marayong, Prospect 2195 Lakemba, Wiley Park $221,581 2,930 $ Auburn, Auburn South $199,377 2,831 $ Box Head, Daleys Point, Empire Bay, Ettalong Beach, Hardys Bay, Killcare, Pretty Beach, Umina Beach, $173,046 2,486 $69.61 Wagstaffe Charmhaven, Gorokan Lake, Lake Haven, Norah Head, $162,073 2,383 $68.01 Toukley Emu Heights, Emu Plains, Jamisontown, Leonay, Penrith, $154,484 2,353 $65.65 Penrith South 2204 Marrickville, Marrickville South $153,474 2,257 $ Darlinghurst, Surry Hills, Taylor Square $151,064 2,226 $ Campsie $159,707 2,214 $ Abbotsbury, Bossley Park, Edensor Park, Greenfield Park, $138,923 2,182 $63.67 Prairiewood, St Johns Park, Wakeley Total $4,310,403 63,640 $68.06 Sydney $12,765, ,707 $67.09 Note: Excludes 5 postcode areas where the number of Commonwealth Rent Assistance recipients was less than 20. Fortnightly amounts of Commonwealth Rent Assistance have been rounded. Source: Centrelink Data

89 Table 8.2: Number of Commonwealth Rent Assistance Recipients, 2001 Top 20 Postcodes in Melbourne Postcode Suburb(s) 3175 Total Fortnightly Amount of CRA Total Number of CRA Recipients Total Amount of CRA per Fortnight per Recipient Bangholme, Dandenong, Dandenong East, Dandenong North, Dandenong South, $289,007 4,520 $63.94 Dunearn 3021 Albanvale, Kealba, Kings Park, St Albans $211,276 3,324 $ Frankston, Frankston East, Frankston $219,435 3,300 $66.50 Heights, Frankston South, Karingal Keon Park, Reservoir, Reservoir East, $208,876 3,113 $67.10 Reservoir North, Reservoir South Noble Park, Noble Park East, Noble Park $181,545 2,769 $65.56 North 3182 St Kilda, St Kilda South, St Kilda West $175,314 2,671 $ Albion, Glengala, Sunshine, Sunshine $165,419 2,638 $62.71 North, Sunshine West Brunswick, Brunswick Lower, Brunswick $135,296 2,124 $63.70 North Gilberton, Preston, Preston Lower, Preston South, Preston West, Regent West, $133,919 2,106 $63.59 Sylvester 3171 Sandown Village, Springvale $114,425 1,922 $ Footscray, Seddon, Seddon West $115,314 1,865 $ Chartwell, Cocoroc, Derrimut, Point Cook, $123,938 1,842 $67.28 Quandong, Werribee, Werribee South Batman, Coburg, Coburg North, Merlynston, $118,583 1,832 $64.73 Moreland 3046 Glenroy, Hadfield, Oak Park $119,272 1,768 $ Cranbourne, Cranbourne East, Cranbourne North, Cranbourne South, Cranbourne $119,784 1,765 $67.87 West, Devon Meadows, Five Ways, Skye 3163 Carnegie, Glen Huntly, Murrumbeena $118,924 1,732 $ Balaclava, St Kilda East $112,534 1,653 $ Brooklyn, Kingsville, Kingsville West, $101,140 1,602 $63.13 Maidstone, Tottenham, West Footscray Croydon, Croydon Hills, Croydon North, $109,721 1,585 $69.22 Croydon South Bonbeach, Chelsea, Chelsea Heights, $94,263 1,450 $65.01 Edithvale Total $2,967,985 45,581 $65.26 Melbourne $10,030, ,033 $64.65 Note: Excludes 18 postcode areas where the number of Commonwealth Rent Assistance recipients was less than 20. Fortnightly amounts of Commonwealth Rent Assistance have been rounded. Source: Centrelink data

90 Table 8.3: Number of Commonwealth Rent Assistance Recipients, 2001 Top 20 Postcodes in Adelaide Postcode Suburb(s) 5108 Total Fortnightly Amount of CRA Total Number of CRA Recipients Total Amount of CRA per Fortnight per Recipient Paralowie, Salisbury, Salisbury Downs, $100,000 2,190 $67.67 Salisbury North 5162 Morphett Vale, Woodcroft $100,000 1,599 $ Davoren Park, Davoren Park North, Davoren Park South, Elizabeth Downs, $98,004 Elizabeth North, Elizabeth Park, Elizabeth 1,476 $66.40 West Elizabeth, Elizabeth East, Elizabeth Grove, $74,027 1,182 $62.63 Elizabeth South, Elizabeth Vale, Hillbank Glenelg, Glenelg East, Glenelg North, $74,648 1,155 $64.63 Glenelg South Gilles Plains, Greenacres, Hampstead $77,528 1,113 $69.66 Gardens, Hillcrest, Manningham, Oakden Andrews Farm, Blakeview, Craigmore, Gould Creek, Humbug Scrub, One Tree Hill, $70,952 Sampson Flat, Smithfield, Smithfield Plains, 1,024 $69.29 Smithfield West, Uleybury, Yattalunga Brahma Lodge, Salisbury East, Salisbury $63, $65.34 Heights, Salisbury Park, Salisbury Plain Camden Park, Plympton, Plympton Park, $62, $66.61 South Plympton Ascot Park, Marion, Mitchell Park, $57, $61.58 Morphettville, Park Holme Adelaide, Adelaide BC, Halifax St, Hutt St, $48, $54.05 Sturt St Fitzroy, Ovingham, Prospect, Prospect East, $59, $66.91 Prospect West, Thorngate Brighton, Dover Gardens, Hove, North $62, $71.84 Brighton, South Brighton Felixstow, Firle, Glynde, Joslin, Marden, $49, $57.26 Payneham, Payneham South, Royston Park 5032 Brooklyn Park, Lockleys, Underdale $49, $ Aberfoyle Park, Chandlers Hill, Flagstaff Hill, $55, $69.42 Fountain Valley, Happy Valley, Woodlea 5023 Findon, Seaton, Seaton North $52, $ Mile End, Mile End South, Thebarton, $45, $57.99 Torrensville 5074 Campbelltown, Newtown $47, $ Eastwood, Frewville, Fullarton, Highgate, $48, $63.47 Parkside Total $1,297,947 20,819 $64.58 Adelaide $3,434,826 54,232 $64.20 Note: Excludes 11 postcode areas where the number of Commonwealth Rent Assistance recipients was less than 20. Fortnightly amounts of Commonwealth Rent Assistance have been rounded. Source: Centrelink data

91 Figure 8.1: The proportion of postcode areas by the number of Commonwealth Rent Assistance recipients 60% 50% 40% 30% Sydney Melbourne Adelaide 20% 10% 0% Less than ,000 1,001-2,000 2,001-3,000 3,001 or more Note: Excludes postcode areas with less than 20 recipients receiving Commonwealth Rent Assistance Figure 8.2: The proportion of postcode areas by the amount of Commonwealth Rent Assistance per recipient 90% 80% 70% 60% 50% 40% Sydney Melbourne Adelaide 30% 20% 10% 0% $1-$39 $40-$49 $50-$59 $60-$69 $70-$79 $80 or more Note: Excludes postcode areas with less than 20 recipients receiving Commonwealth Rent Assistance 86

92 Figure 8.3: Number of Commonwealth Rent Assistance recipients by postcode in Sydney in

93 Figure 8.4: Number of Commonwealth Rent Assistance recipients by postcode in Melbourne in

94 Figure 8.5: Number of Commonwealth Rent Assistance recipients by postcode in Adelaide in

95 Figure 8.6: Amount of Commonwealth Rent Assistance per recipient in Sydney in

96 Figure 8.7: Amount of Commonwealth Rent Assistance per recipient in Melbourne in

97 Figure 8.8: Amount of Commonwealth Rent Assistance per recipient in Adelaide in

98 8.2 Public housing waiting lists in 2001 The second major form of housing assistance available to lower income households is public housing. The majority of households who access public housing come from the private rental market (Burke et al 2005). There is, therefore, likely to be some relationship between those on public housing waiting lists and the lower income segments of the rental market. However, again, whether the low income working segment is closely spatially associated with those on the waiting lists, who are likely to have become increasingly dominated by nonworking and benefit dependent households as a result of increased targeting of public housing over the last twenty years, has not been explored in detail. In effect this form of housing assistance has been withdrawn from the low income working population who now are expected to meet their housing needs in the private rental sector, or to buy. As we have seen, the latter option seems to be increasingly constrained, leading to greater numbers of low income working households living in private rental accommodation for long periods (Wulff and Maher 1998). A close spatial coincidence between wait list applicants and PRLIH would confirm that this group is again being excluded from the only other form of housing assistance available to low income households while others in essentially the same housing market, and who are also likely to be in receipt of RA while they wait for public housing allocation, may be assisted. The waiting list data used in this section was obtained from each of the respective State Housing Authorities for the three cities in mid The data presented in Figures 8.9 to 8.11 suggest that the largest numbers of applicants on public housing waiting lists are in areas in which there are also high numbers of PRLIHs and PRLIWHs. In Sydney, high numbers of public housing waiting list applicants can be found in the areas in the south west and western parts of the city where there were also high numbers of PRLIHs and Commonwealth Rent Assistance recipients (Figure 8.9). One significant exception is in Gosford-Wyong, where despite large numbers of Commonwealth Rent Assistance recipients, there were relatively low numbers of waiting list applicants. This may be a function of both the more limited supply of public housing here as well as the more prevalent supply of low cost rental housing. In Melbourne the largest concentrations of public housing waiting list applicants can be found around Brimbank, Darebin, Dandenong and Mornington Peninsula (Figure 8.10). These areas also have large numbers PRLIHs and Commonwealth Rent Assistance recipients. In Adelaide, the largest numbers of applicants on the public housing waiting list are located in and around Salisbury in the north and Onkaparinga in the south (Figure 8.11). Both these areas also have large numbers of Commonwealth Rent Assistance recipients. 8.3 The relationship between public housing waiting lists, Commonwealth Rent Assistance, low cost rental stock and PRLIHs in 2001 In order to better understand the relationship between the four sets of data analysed above (numbers of applicants on public housing waiting lists, number of Commonwealth Rent Assistance recipients, the location of PRLIHs and the low cost private rental stock), a simple correlation analysis was undertaken for the postcode level data. Table 8.4 confirms the significant association between these four sets of data across the three cities. In Sydney, Melbourne and Adelaide the correlation coefficient between PRLIHs and Commonwealth Rent Assistance recipients is greater than 0.9. The correlation between the number of PRLIHs and the number of public housing applicants is 0.9 in Adelaide and 0.8 in Sydney and Melbourne. Thirdly, the correlation between the number of Commonwealth Rent Assistance recipients and the number of public housing applicants is 0.9 in Adelaide and Melbourne and 0.8 in Sydney. Finally, the correlation between the location of the low cost private rental stock and the number of PRLIHs is greater than 0.9 in the three cities. This is also similar to the correlation coefficients between the low cost private rental stock and the number of Commonwealth Rent Assistance recipients. The correlation between the number of public housing applicants and the location of the low cost private rental stock is also significant at approximately 0.8 for the three cities. It is important to understand that these results do not 93

99 show causality. However, it is clear that there is a very strong association between the four variables at the postcode level and that the associations are statistically significant. What does this mean? Put simply, it means that all four elements of the low cost and low income housing market largely coexist in the same spatial housing sub-markets. While there are some variations, waiting list applicants, Commonwealth Rent Assistance recipients and low paid working private renters all compete for housing in the same low cost rental housing market. For those in highest needs (defined increasingly by lone parenthood, medical vulnerability or disability) and for those eligible for RA, some form of housing assistance is available. For the working poor, no such assistance is available or likely to be available. Yet as noted above, the financial benefits for working for this group may be marginal once their full housing costs are taken into account, and of course there is no prospect of gaining access to subsidised public housing to lessen the rent burden. For many, this form of tenure will remain their only lifetime option as will be the locations in which their housing is concentrated. Table 8.4: Correlation coefficients of Commonwealth Rent Assistance recipients, public housing waiting lists, PRLIHs and low cost private rental dwellings in Adelaide, Melbourne and Sydney Sydney Number of PRLIHs Number of Commonweal th Rent Assistance Recipients Number of Public Housing Applicants Number of Low Cost Private Rental Dwellings Number of PRLIHs * 0.78* 0.94* Number of Commonwealth Rent Assistance Recipients 0.93* * 0.94* Number of Public Housing Applicants 0.78* 0.83* * Number of Low Cost Private Rental Dwellings 0.94* 0.94* 0.79* - Melbourne Number of PRLIHs * 0.82* 0.96* Number of Commonwealth Rent Assistance Recipients 0.92* * 0.93* Number of Public Housing Applicants 0.82* 0.91* * Number of Low Cost Private Rental Dwellings Adelaide 0.96* 0.93* 0.85* - Number of PRLIHs * 0.87* 0.98* Number of Commonwealth Rent Assistance Recipients 0.97* * 0.89* Number of Public Housing Applicants 0.87* 0.94* * Number of Low Cost Private Rental Dwellings * Correlation is significant at 0.01 level 0.98* 0.89* 0.79* - Note: Public housing waiting list data for Adelaide is at June 2003, while Commonwealth Rent Assistance data is for Public housing waiting list data for both Melbourne and Sydney is for Number of PRLIHs and low cost private rental dwellings is for

100 Figure 8.9: Number of applicants on the public housing waiting list by postcode in Sydney in June

101 Figure 8.10: Number of applicants on the public housing waiting list by postcode in Melbourne in

102 Figure 8.11: Number of applicants on the public housing waiting list by postcode in Adelaide in

103 8.4 The changing location of Commonwealth Rent Assistance recipients in We have seen that there has been a marked spatial shift in the relative and absolute location of the PRLIH population in the three cities between 1991 and 2001, with the principle outcome being a suburbanisation of the sector. And the close association between this population and Commonwealth Rent Assistance (RA) recipients has also been established. Has the shift in the PRLIH sector been associated with comparable changes in the location of RA recipients? This section reviews RA data for 2000 and 2005, the only time scale available for the analysis of these data, to see if any similar locational changes can be detected. While obviously not strictly comparable to the 1991 to 2001 period used in chapter 6, if there have been changes of a similar nature then they are likely to be reflected in trends during the 2000 to 2005 period. Indeed, given the state of the residential property market during this period, such trends may even have accelerated. The changes are to some extent related to the overall decline in RA recipients in Sydney and Melbourne, following the general fall in unemployment over this period, although it is noticeable that numbers of RA recipients rose in Adelaide. The trends will also be influenced by an increase in the number of families receiving CRA since Nevertheless, within each city, there has been a marked and consistent shift in the location of recipients Absolute changes in the distribution of Commonwealth Rent Assistance recipients in In 2000, there were 37,229 Commonwealth Rent Assistance (CRA) recipients in the ten postcode areas containing the largest number of recipients in Sydney (Table 8.5). This represented 20% of all CRA recipients in Sydney in The five postcode areas with the largest number of CRA recipients were 2166 (Cabramatta/Canley Vale), 2170 (Liverpool), 2261 (The Entrance/Bateau Bay), 2165 (Fairfield) and 2259 (Tuggerah/Lake Munmorah). In 2005 these same postcode areas also had the largest number of CRA recipients, although the order had changed (Table 8.5). In 2005 the five postcode areas with the largest number of CRA recipients were 2170, 2166, 2259, 2165 and Interestingly though, the ten postcode areas with the largest number of CRA recipients in 2005 in Sydney contained 38,315 recipients. This represents 21% of all CRA recipients in Sydney at this time, suggesting a marginally increased concentration in these ten areas. At the same time, there has been a noticeable shift in the location of CRA recipients (Table 8.8 and Figure 8.12). The postcode areas with the largest increases in the number of CRA recipients between 2000 and 2005 were located in south west Sydney, western Sydney and in Gosford and Wyong on the Central Coast. These areas included middle ring areas of disadvantage as well as newer housing areas on the fringe of Sydney where new residential development has been associated with the investment market. For example, Randolph and Holloway (2003) found that some 17% of dwellings in the most recent new suburbs in Western Sydney were privately rented in At the same time, there were significant decreases in the number of CRA recipients in the gentrifying inner city, Cabramatta in Sydney s west, and in the rural residential area on the edge of the new release areas in Liverpool (postcode 2171). The latter may be an outcome of cheap rental properties being vacated prior to urban development. The loss of RA recipients in Cabramatta, is harder to explain, but may simply reflect general economic improvement over this period. In Melbourne, there were 28,652 CRA recipients in the ten postcode areas containing the largest number of recipients (Table 8.6). This represented 19% of all CRA recipients in Melbourne in The four postcode areas with the largest number of CRA recipients were 3175 (Dandenong), 3199 (Frankston), 3021 (Albanvale/St Albans), and 3073 (Reservoir). In 2005 these same four postcode areas also had the largest number of CRA recipients and in the same order as in 2000 (Table 8.6). Interestingly though, the ten postcode areas with the 10 Unpublished data, Department of Families, Community Services and Indigenous Affairs. Between 2000 and 2005 the numbers of families with children among CRA recipients increased by 21,231 or 17.3%, in the three cities. 98

104 largest number of CRA recipients in 2005 in Melbourne contained 27,860 recipients, slightly less than the number of recipients in Overall this still represents 19% of all Commonwealth Rent Assistance recipients in Melbourne. Similar to Sydney, there has been a significant decrease in the number of CRA recipients between 2000 and 2005 in the gentrifying inner city and inner east of Melbourne (Table 8.8 and Figure 8.13). Conversely, there has also been an increase in the number of CRA recipients in areas on the western and south-eastern fringe areas of Melbourne. In 2000, there were 11,743 CRA recipients in the ten postcode areas containing the largest number of recipients in Adelaide (Table 8.7). This represented 22% of all CRA recipients in Adelaide in The five postcode areas with the largest number of CRA recipients in 2000 were 5108 (Salisbury), 5162 (Morphett Vale/Woodcroft), 5113 (Davoren Park/Elizabeth Park), 5045 (Glenelg) and 5086 (Greenacres/Manningham). The three postcode areas with the largest number of recipients in 2000 were again the largest in 2005, and in the same order (Table 8.7). In fact, the ten postcode areas with the largest number of CRA recipients in 2000 were the largest ten in 2005, although with different rankings. Interestingly, though unlike Sydney and Melbourne, there was an increase in the number of CRA recipients in Adelaide between 2000 and As such, the ten postcode areas with the largest number of recipients represented 24% of all CRA recipients in Adelaide in Overall there was a decrease in the number of CRA recipients in parts of inner Adelaide and the inner north-western coast, although not to the same extent as the changes in inner Sydney and Melbourne (Table 8.8 and Figure 8.14). Similar to Sydney though there was an increase in the number of CRA recipients in some of the more disadvantaged northern suburbs of Adelaide (Salisbury and Elizabeth) as well as Morphett Vale/Woodcroft in the south. 99

105 Table 8.5: Number of Commonwealth Rent Assistance recipients in 2000 and 2005 top 10 postcodes in Sydney Top 10 Postcodes Suburb(s) Number of CRA Recipients 2000 Top 10 Postcodes 2005 Cabramatta, Cabramatta West, Canley Heights, Canley Vale, Lansvale Casula, Chipping Norton, Hammondville, Liverpool, Liverpool South, Lurnea, Moorebank, Mount Pritchard, Prestons, Warwick Farm Bateau Bay, Berkeley Vale, Blue Bay, Chittaway Bay, Chittaway Point, Gleening Valley, Killarney Vale, Long Jetty, Shelly Beach, The Entrance, Tumbi Umbi Suburb(s) Casula, Chipping Norton, Hammondville, Liverpool, Liverpool South, Lurnea, Moorebank, Mount Pritchard, 5,398 Prestons, Warwick Farm Cabramatta, Cabramatta West, Canley Heights, 5,057 Canley Vale, Lansvale Gwandalan, Kanwal, Lake Munmorah, Mannering Park, Rocky Point, South Tacoma, Summerland Point, 4,339 Tacoma, Tuggerah 2165 Fairfield, Fairfield East, Fairfield Heights, Fairfield West Fairfield, Fairfield East, Fairfield Heights, Fairfield West 3, Gwandalan, Kanwal, Lake Munmorah, Mannering Park, Rocky Point, South Tacoma, Summerland Point, Tacoma, Tuggerah Calga, Central Mangrove, East Gosford, Erina, Gosford, Gosford South, Gosford West, Lisarow, Mangrove Mountain, Mooney Mooney, Mount White, Narara, Niagara Park, North Gosford Bidwill, Blackett, Dharruk, Emerton, Lethbridge Park, Minchinbury, Mount Druitt, Shalvey, Tregear, Whalan, Willmot Girraween, Greystanes, Mays Hill, Pendle Hill, South Wentworthville, Wentworthville, Westmead Airds, Ambarvale, Appin, Blair Athol, Bradbury, Campbelltown, Englorie Park, Gilead, Glen Alpine, Leumeah, Rosemeadow, Ruse, St Helens Park, Woodbine 2195 Lakemba, Wiley Park Bateau Bay, Berkeley Vale, Blue Bay, Chittaway Bay, Chittaway Point, Gleening Valley, Killarney Vale, Long 3,660 Jetty, Shelly Beach, The Entrance, Tumbi Umbi Calga, Central Mangrove, East Gosford, Erina, Gosford, Gosford South, Gosford West, Lisarow, 3,509 Mangrove Mountain, Mooney Mooney, Mount White, Narara, Niagara Park, North Gosford Girraween, Greystanes, Mays Hill, Pendle Hill, South 3,230 Wentworthville, Wentworthville, Westmead Bidwill, Blackett, Dharruk, Emerton, Lethbridge Park, Minchinbury, Mount Druitt, Shalvey, Tregear, Whalan, 3,175 Willmot Airds, Ambarvale, Appin, Blair Athol, Bradbury, Campbelltown, Englorie Park, Gilead, Glen Alpine, 3,171 Leumeah, Rosemeadow, Ruse, St Helens Park, Woodbine Arndell Park, Blacktown, Huntingwood, Kings Park, 3,037 Marayong, Prospect Total 37,229 Total 38,315 Number of CRA Recipients 2005 Sydney 182,590 Sydney 181,443 Note: Excludes 6 postcode areas in 2000 and 4 in 2005 where the number of Commonwealth Rent Assistance recipients was less than 20. Source: Australian Government Department of Family and Community Services and Indigenous Affairs 100

106 Table 8.6: Number of Commonwealth Rent Assistance recipients in 2000 and 2005 top 10 postcodes in Melbourne Top 10 Postcodes Suburb(s) Number of Commonwealth Rent Assistance Recipients 2000 Top 10 Postcodes 2005 Bangholme, Dandenong, Dandenong East, Dandenong 4, North, Dandenong South, Dunearn Frankston, Frankston East, Frankston Heights, Frankston 3, South, Karingal Suburb(s) Bangholme, Dandenong, Dandenong East, 4,493 Dandenong North, Dandenong South, Dunearn Frankston, Frankston East, Frankston Heights, 3,540 Frankston South, Karingal 3021 Albanvale, Kealba, Kings Park, St Albans 3, Albanvale, Kealba, Kings Park, St Albans 3, Keon Park, Reservoir, Reservoir East, Reservoir North, 3, Reservoir South Keon Park, Reservoir, Reservoir East, Reservoir 2,911 North, Reservoir South 3182 St Kilda, St Kilda South, St Kilda West 2, Noble Park, Noble Park East, Noble Park North 2, Noble Park, Noble Park East, Noble Park North 2, Albion, Glengala, Sunshine, Sunshine North, Sunshine 2, West Chartwell, Cocoroc, Derrimut, Point Cook, Quandong, 2,374 Werribee, Werribee South Albion, Glengala, Sunshine, Sunshine North, 2,306 Sunshine West 3056 Brunswick, Brunswick Lower, Brunswick North 2, St Kilda, St Kilda South, St Kilda West 2, Gilberton, Preston, Preston Lower, Preston South, Preston 2, West, Regent West, Sylvester 3011 Footscray, Seddon, Seddon West 2, Cranbourne, Cranbourne East, Cranbourne North, Cranbourne South, Cranbourne West, Devon 2,074 Meadows, Five Ways, Skye Gilberton, Preston, Preston Lower, Preston South, 1,873 Preston West, Regent West, Sylvester Total 28,652 Total 27,860 Number of Commonwealth Rent Assistance Recipients 2005 Melbourne 153,243 Melbourne 148,450 Note: Excludes 18 postcode areas in 2000 and 14 in 2005 where the number of Commonwealth Rent Assistance recipients was less than

107 Table 8.7: Number of Commonwealth Rent Assistance recipients in 2000 and 2005 top 10 postcodes in Adelaide Top 10 Postcodes Suburb(s) 2000 Number of Commonwealth Rent Assistance Recipients Paralowie, Salisbury, Salisbury Downs, Salisbury North 2, Top 10 Postcodes 2005 Suburb(s) Paralowie, Salisbury, Salisbury Downs, Salisbury 2,275 North 5162 Morphett Vale, Woodcroft 1, Morphett Vale, Woodcroft 1, Davoren Park, Davoren Park North, Davoren Park South, Elizabeth Downs, Elizabeth North, Elizabeth Park, 1, Elizabeth West Glenelg, Glenelg East, Glenelg North, Glenelg South 1, Gilles Plains, Greenacres, Hampstead Gardens, Hillcrest, 1, Manningham, Oakden Elizabeth, Elizabeth East, Elizabeth Grove, Elizabeth 1, South, Elizabeth Vale, Hillbank Davoren Park, Davoren Park North, Davoren Park South, Elizabeth Downs, Elizabeth North, Elizabeth 1,675 Park, Elizabeth West Elizabeth, Elizabeth East, Elizabeth Grove, Elizabeth 1,295 South, Elizabeth Vale, Hillbank Andrews Farm, Blakeview, Craigmore, Gould Creek, Humbug Scrub, One Tree Hill, Sampson Flat, 1,132 Smithfield, Smithfield Plains, Smithfield West, Uleybury, Yattalunga Gilles Plains, Greenacres, Hampstead Gardens, 1,117 Hillcrest, Manningham, Oakden Andrews Farm, Blakeview, Craigmore, Gould Creek, Humbug Scrub, One Tree Hill, Sampson Flat, Smithfield, Glenelg, Glenelg East, Glenelg North, Glenelg South 1,040 Smithfield Plains, Smithfield West, Uleybury, Yattalunga Brahma Lodge, Salisbury East, Salisbury Heights, Salisbury Park, Salisbury Plain 5038 Camden Park, Plympton, Plympton Park, South Plympton Ascot Park, Marion, Mitchell Park, Morphettville, Park Holme Ascot Park, Marion, Mitchell Park, Morphettville, Park 1,006 Holme Brahma Lodge, Salisbury East, Salisbury Heights, 928 Salisbury Park, Salisbury Plain Camden Park, Plympton, Plympton Park, South 904 Plympton Total 11,743 Total 13,126 Adelaide 52,545 Adelaide 54,875 Note: Excludes 11 postcode areas in 2000 and 13 in 2005 where the number of Commonwealth Rent Assistance recipients was less than 20. Number of Commonwealth Rent Assistance Recipients

108 Postcode Sydney Table 8.8: Change in the number of Commonwealth Rent Assistance recipients in Sydney, Melbourne and Adelaide in top 10 postcodes with the largest increases and decreases Suburb(s) Change in Number of Commonw ealth Rent Assistance Recipients Alison, Callungra, Cedar Brush Creek, Chain Valley Bay, Crangan Bay, Dooralong, Durren Durren, Frazer Park, Gwandalan, Halloran, Hamlyn Terrace, Jilliby, Kanwal, Lake Munmorah, Lemon Tree, Little Jilliby, Mannering Park, Mardi, Point Wolstoncroft, Ravensdale, Rocky Point, South Tacoma, Summerland Point, Tacoma, Tuggerah, Tuggerawong Arndell Park, Blacktown, Huntingwood, Kings Park, Marayong, Prospect Postcode Suburb(s) Change in Number of Commonwe alth Rent Assistance Recipients Cabramatta, Cabramatta West, Canley Heights, Canley Vale, Lansvale Bondi, Bondi Beach, North -719 Bondi, Tamarama Kellyville, Kellyville Ridge, Marrickville, Marrickville South -528 Rouse Hill 2160 Merrylands, Merrylands West Currans Hill, Harrington Park, Mount Annan, Narellan, Narellan Vale 2220 Hurstville, Hurstville Grove Darlinghurst, Surry Hills, Taylor -460 Square Austral, Badgerys Creek, Bringelly, Catherine Field, Cecil Hills, Cecil Park, Edmondson Park, Horningsea Park, Hoxton -408 Park, Kemps Creek, Leppington, Rossmore, Voyager Point, West Hoxton Elizabeth Bay, Garden Island, Kings Cross, Potts Point, -338 Rushcutters Bay, Woolloomooloo Bankstown, Bankstown North, Condell Park, Lakemba, Wiley Park -316 Manahan, Mount Lewis Airds, Ambarvale, Appin, Blair Athol, Bradbury, Campbelltown, Englorie Park, Gilead, Glen Alpine, Leumeah, Rosemeadow, Ruse, St Helens Park, Woodbine Casula, Chipping Norton, Hammondville, Liverpool, Liverpool South, Lurnea, Moorebank, Mount Pritchard, Prestons, Warwick Farm Clovelly, Clovelly West, -302 Randwick Lewisham, Petersham North Petersham,

109 Postcode 2145 Melbourne Suburb(s) Change in Number of Commonw ealth Rent Assistance Recipients Postcode Suburb(s) Girraween, Greystanes, Mays Hill, Pendle Hill, South 286 Wentworthville, 2203 Dulwich Hill -241 Wentworthville, Westmead 3045 Melbourne Airport Calder Park, Delahey, Hillside, Plumpton, Sydenham, Taylors Hill Chartwell, Cocoroc, Derrimut, Point Cook, Quandong, Werribee, Werribee South Craigieburn, Kalkallo, Roxburgh Park St Kilda, St Kilda South, St -626 Kilda West Footscray, Seddon, Seddon -619 West Albion, Glengala, Sunshine, Sunshine North, Sunshine West -390 Donnybrook, Mickleham, Brighton Road, Elwood -385 Fountain Gate, Narrewarren, Narrewarren South Brunswick, Brunswick Lower, -328 Brunswick North Hoppers Crossing, Tarneit, Clayton, Notting Hill -319 Truganina Burnside, Cairnlea, Caroline Springs, Deer Park, Deer Park North, Ravenhall Batman, Coburg, Coburg North, -300 Merlynston, Moreland Pakenham, Pakenham South, Pakenham Upper, Cotham, Kew -283 Rythdale Cannons Creek, Cranbourne, Cranbourne East, Cranbourne North, Cranbourne South, Fitzroy -282 Cranbourne West, Devon Meadows, Five Ways, Junction Village, Skye 3201 Carrum Downs Balaclava, St Kilda East -279 Adelaide Davoren Park, Davoren Park North, Davoren Park South, Elizabeth Downs, Elizabeth Edinburgh -240 North, Elizabeth Park, Elizabeth West Elizabeth, Elizabeth East, Elizabeth Grove, Elizabeth South, Elizabeth Vale, Hillbank Evanston, Evanston Gardens, Evanston Park, Evanston South, Hillier Paralowie, Salisbury, Salisbury Downs, Salisbury North Eastwood, Frewville, Fullarton, -173 Highgate, Parkside Glenelg, Glenelg East, Glenelg Jetty Road, Glenelg North, -119 Glenelg South Mile End, Mile End South, -90 Thebarton, Torrensville Change in Number of Commonwe alth Rent Assistance Recipients

110 Postcode Suburb(s) Change in Number of Commonw ealth Rent Assistance Recipients Postcode Suburb(s) 5162 Morphett Vale, Woodcroft Holden Hill Andrews Farm, Blakeview, Craigmore, Gould Creek, Humbug Scrub, One Tree Hill, Sampson Flat, Smithfield, Smithfield Plains, Smithfield West, Uleybury, Yattalunga Clearview, Enfield, Northfield, Northgate Moana, Seaford, Seaford Heights, Seaford Meadows, Seaford Rise Bibaringa, Buchfelde, Concordia, Gawler, Gawler Belt, Gawler East, Gawler River, Gawler South, Gawler West, Hewett, Kalbeeba, Kangaroo Flat, Kingsford, Ward Belt, Willaston Ascot Park, Marion, Mitchell Park, Morphettville, Parkholme Modbury, Modbury Heights, -65 Modbury North Felixstow, Firle, Glynde, Joslin, Marden, Payneham, Payneham -64 South, Royston Park Cowandilla, Hilton, Marleston, -61 Richmond, West Richmond Exeter, Semaphore, Semaphore Park, Semaphore -58 South Grange, Henley Beach, Henley Beach South, Kirkcaldy, -57 Tennyson Note: No account is taken of changes in postcode boundaries between 2000 and Change in Number of Commonwe alth Rent Assistance Recipients

111 Figure 8.12: The change in number of Commonwealth Rent Assistance recipients by postcode in Sydney in March March

112 Figure 8.13: The change in number of Commonwealth Rent Assistance recipients by postcode in Melbourne in March March

113 Figure 8.14: The change in number of Commonwealth Rent Assistance recipients by postcode in Adelaide in March March

114 8.4.2 Relative changes in the distribution of Commonwealth Rent Assistance recipients in This section examines the relative change in the proportion of Commonwealth Rent Assistance (CRA) recipients between 2000 and To do this we have calculated the location quotients for the number of CRA recipients in each postcode area as a proportion of the metropolitan wide total in 2000 and We have then calculated the change between the two years. This method allows us to take account of the changing number of CRA recipients in each city over time. The analysis here confirms the data presented above in that, generally speaking, there has been a decrease in the proportion of CRA recipients in the inner cities of Sydney, Melbourne and Adelaide. Conversely there has been an increase in the relative proportion of CRA recipients in certain outer suburban areas in each of the three cities. In Sydney, greatest relative losses were recorded in the inner Sydney, the inner east and northern Sydney (Figure 8.15). There have also been decreases in the outer south-west and north west of Sydney in the rural fringe and Blue Mountains. At the same time, there were increases in the proportion of Sydney s CRA recipients in the western suburbs (along the Parramatta to Penrith corridor), in south-western Sydney (Fairfield to Campbelltown) and to the north in Gosford-Wyong. In Melbourne, there has also been a decrease in the relative proportion of Melbourne s RA population in the inner city, around Mornington in the south and in the rural fringe areas to the outer east and north-east. (Figure 8.16). Conversely, there were relative increases in the proportion of recipients in eastern Melbourne (Croydon-Noble Park), the south-east (Dandenong-Frankston), around Rosebud in the far south, and in the outer west of Melbourne. In Adelaide a similar picture emerges. There were decreases in the relative proportion of Adelaide s RA recipients in the inner city and in the inner north west (up to Port Adelaide/Prospect), and in the outer south-east between 2000 and 2005 (Figure 8.17). In contrast, relative increase was recorded mainly in the outer south (around Morphett Vale) and in the north of Adelaide (Salisbury-Elizabeth-Gawler). Although there were also a few postcode areas in the north (e.g. Edinburgh) who recorded decreases. The picture of the relative shift in RA recipients therefore appears remarkably similar in all three cities: Losses in inner and higher value central suburbs as well as in the life style and rural fringe areas, contrasting to increases in middle suburban and some outer suburban and coastal areas. In Adelaide, the relative shift to the north of the city and the increases in the south along the coast are notable features. These shifts may reflect the increasing numbers of families on rent assistance noted above, which might be expected to augment the increases recorded in more suburban areas. 109

115 Figure 8.15: The change in location quotient of Commonwealth Rent Assistance recipients in Sydney in