Investification: Financialisation of housing markets and persistence of suburban socio-economic disadvantage

Abstract

Chinese

The relationship between urban housing markets and spatial patterns of socio-economic disadvantage has fascinated urban scholars for decades. The gentrification and subsequently suburbanisation of disadvantage literatures have explained how housing markets are both a driver, and outcome, of changes in the socio-economic composition of urban areas, albeit focusing mainly on owner occupation and social housing. In the 2000s, research into the financialisation of housing finds increased household-level investment in private rented housing as an important contemporary driver of housing markets. Based on a detailed study of Melbourne (Australia) in 2001–2011, the article identifies established suburbs of persistent population socio-economic disadvantage, which were characterised by sale prices and rents increasing above citywide rates in 2001–2011 and a disproportionate increase in private rented housing. The article offers a new concept of investification to explain a process whereby disproportionately high levels of household investor purchases in disadvantaged suburbs contribute to higher prices/rents and to the persistence of socio-economic disadvantage, as properties are rented on the private market to low socio-economic households, indicating replacement rather than displacement. Connecting with research on the financialisation of housing through the concept of ‘investification’ can provide a more nuanced understanding of the relationship between contemporary housing market change and the geography of suburban disadvantage in the Australian context. The concept is likely to be of broader significance given the recent increase in Buy-to-Let activity in countries such as the UK, opening up new research questions on the interrelationship between households as investors and consumers and the geography of urban disadvantage.

Keywords

gentrification investification private rented sector rental investment suburbanisation of disadvantage

Introduction

Mapping spatial patterns of urban settlement by the socio-economic characteristics of residents of local areas has fascinated urban scholars internationally for almost a century. From the days of the Chicago School human ecologists in the 1920s, there has been a strong desire to understand the processes that underlie these settlement patterns. Early explanations of spatial sorting by socio-demographic and economic status were based on biological metaphors and competition (Burgess, 1929; Hoyt, 1939; Park et al., 1925). Subsequently, social area analysis and urban ecology analyses deployed increasingly sophisticated statistical techniques for mapping the spatial settlement patterns of city residents whilst seeking to understand the ways in which economic, social and household factors underpinned change in urban areas (Burnley, 1980; Johnston, 1973; Kasarda, 1990; Shevky and Bell, 1955). More recently, sophisticated computer mapping and data visualisation techniques have enabled a more nuanced understanding of changing spatial patterns of settlement in cities with a greater degree of integration between changing population profiles, economic activity and changes in the built and natural environment (e.g. Bell et al., 2007; Randolph and Freestone, 2012).

Within a very broad field of research, there has been ongoing interest in the interplay between housing market dynamics and spatial patterns of socio-economic disadvantage over time using population-based measures.¹ This article seeks to contribute to a greater understanding of the role of housing markets, in particular household-level investment in rented housing, in reinforcing high levels of persistent population-based, socio-economic disadvantage in established middle/outer suburban areas, through a detailed examination of the relationship between persistent socio-economic disadvantage and housing market change in Melbourne, Australia, between 2001 and 2011.

The article builds on insights from three strands of literature: gentrification, the suburbanisation of disadvantage and the financialisation of housing markets. It proceeds by discussing the Australian urban context and changes in the scale and type of investment in rental housing over the study period, followed by a discussion of the research approach and methods. The article then presents findings on the persistence of socio-economic disadvantage in clusters of established middle/outer suburbs in Melbourne in 2001–2011, and explores housing market change in these suburbs, identifying growth in private rented housing as a key element of change. Finally, it discusses the implications of these findings, introducing the concept of ‘investification’ to explain how household investment in property for rental became a key driver of Australian housing markets during the decade, with a disproportionate effect on areas of persistent socio-economic disadvantage. The article concludes by considering whether these findings are specific to growth cities such as those in Australia, or could be applicable more broadly given the increase in private rental in many countries as noted by Crook and Kemp (2014).

Theorising the nexus between housing market dynamics and areas of socio-economic disadvantage

The gentrification and the suburbanisation of disadvantage literatures have offered considerable insights into the relationship between housing market dynamics and areas of population socio-economic disadvantage, focusing primarily on owner occupation and social rented housing. Newer research into the financialisation of housing markets has identified the growth in household level investment in private rental, although implications for the geography of urban disadvantage remain under theorised.

The extensive scholarship on gentrification over more than five decades (see Bridge and Watson, 2011; Lees et al., 2008; and Slater, 2006, 2011 for recent reviews) has contributed an understanding that housing market change could be both a driver of area socio-economic change and an outcome of such changes. Originally seen as opposing production and consumption approaches to gentrification, recent reviews suggest that a binary approach is outdated and unhelpful since any account of gentrification must consider both to some degree (e.g. Lees, 1994; Lees et al., 2008: Chapters 2 and 3; Slater, 2011: 575). The production approach positions housing/land markets as drivers of change, owing much to Smith’s (1979, 1982) work on the rent gap between current and best value use. When this gap becomes sufficiently large, a variety of institutional actors rehabilitate dwellings and/or redevelop these areas, in the process attracting higher socio-economic households and raising housing prices (Smith, 1979). The consumption approach, associated particularly with the work of Ley (1986, 1994, 1996), identifies socio-cultural reasons why middle class owner occupier households move into disadvantaged inner urban areas, upgrading the housing and in the process re-shaping the socio-economic, political and cultural dimensions of these areas. The outcome of this process is that the area becomes less disadvantaged, using population-based socio-economic measures, and house prices increase due to quality improvements and greater ‘desirability’ of the area.

The complex causal processes involved can be seen in the evolution of debates about displacement as the means by which the population socio-economic profile of an area changes. Whilst originally viewed as direct displacement (see Hamnett, 2000) of working class residents by middle class owner occupiers (e.g. by eviction from tenancy), subsequent work on gentrification suggests that displacement associated with gentrification is often more complex, encompassing a broad range of physical, economic, social and cultural processes (see Lees et al., 2008 and Slater, 2006 for overviews). This includes indirect displacement of working class residents through changing the cultural and social context of neighbourhoods (Davidson and Lees, 2010), including indirect economic displacement in which households are priced out of inner city areas (Davidson, 2008). A more recent focus on ‘new build gentrification’, rather than renovation of run-down existing properties, argues that this type of gentrification may not result in displacement at all since the additional housing supply is the result of conversions from office or industrial/warehouse buildings (Hamnett and Whitelegg, 2007: 122).

Whilst the gentrification literature focuses predominantly on inner urban areas, scholarship on the ‘suburbanisation of disadvantage’ has charted concentration of population-based socio-economic disadvantage in middle and outer suburban areas increasingly distant from the central core of cities. Viewed initially as the consequence of displacement due to inner urban gentrification, in the 2000s there has been a greater appreciation of the spatial effects of intensified urban economic restructuring. In this process, housing prices which factor in increasing concentration of jobs, infrastructure, facilities and services in inner urban areas have meant that households with fewer resources have to live further and further from the inner city (Randolph and Freestone, 2012; Randolph and Tice, 2014). Australian urban scholars have made a particular contribution to this literature (e.g. Baum and Gleeson, 2010; Baum et al., 2006; Gleeson, 2006; Randolph and Freestone, 2012; Randolph and Holloway, 2005; Randolph and Tice, 2014). This phenomenon is not unique to Australia; there is also a growing literature on the suburbanisation of disadvantage in the US (e.g. Holliday and Dwyer, 2009; Kneebone, 2014; Kneebone and Garr, 2010; Madden, 2003; Murphy, 2007) and England/Wales (e.g. Hunter, 2014).

A third strand of scholarship has highlighted increasing financialisation of urban housing markets, in particular, increased household-level investment in properties for private rental not owner occupation. This trend has been observed in a number of advanced economies internationally (Hulse and Yates, 2017; Kemp, 2015; Schwartz, 2015), albeit with variation according to local context (Crook and Kemp, 2014). It has been argued that this trend has accelerated since the Global Financial Crisis of 2008–2009, such that: ‘a reinvigorated private rented sector … is emerging from the “creative destruction” of post-crisis and post-bubble housing markets’ (Forrest and Hirayama, 2015: 239). This trend is important in that it involves household-level rather than institutional investment in which households can be both producers and consumers of housing, disentangling decisions on where to invest from where to live. Whilst some of the accounts of gentrification of inner city areas did include increased private rental (e.g. Engels, 1999; Logan, 1982; Maher, 1985; Rofe, 2003), a recent criticism of the gentrification literature, at least in the UK, is that it does not take into account the increasingly important role of households as Buy to Let (BTL) landlords (Paccoud, 2017). It is further argued that BTL landlords respond to the opening of a ‘value gap’ (drawing on Smith, 1979) to attract wealthy tenants to disadvantaged central areas 2001–2011, thereby contributing to further displacement of low income residents (Paccoud, 2017). The focus is still on ‘central areas’, however, and not on the effects of such investment on disadvantaged suburban areas which is of particular relevance in the Australian context, as discussed next.

Urban housing market restructuring: The Australian context

Australia is a highly urbanised country, with two thirds of Australians (66 per cent) living in just eight state/territory capital cities near the coast, and 35 per cent living in two of these, Sydney and Melbourne (ABS, 2014). Population growth of 15 per cent between 2001 and 2011 (ABS, 2012) was centred on a very small number of large cities which have very low population density by international standards. Consistent with many other cities internationally, Melbourne experienced its first wave of gentrification in the 1970s with deindustrialisation of inner suburbs and an influx of ‘culturally conspicuous, educated middle class’ (Wulff and Lobo, 2009) who bought, renovated and lived in often historic but poor quality properties, leading to direct displacement of low income residents (Kendig, 1979, 1984; Logan, 1985). Second wave gentrification took place following global economic restructuring in the later 1980s and was characterised by new high-rise housing, conversion of office blocks to apartment complexes and new medium density housing in inner suburbs, which resulted in more indirect replacement of low income and disadvantaged residents (Badcock, 1994; Maher, 1994). By 2001, the gentrification of the inner areas of Australia’s major cities was largely complete (Randolph and Tice, 2014: 385) other than small ‘islands’ of medium and high rise public housing (Shaw and Hagemans, 2015).

Over the ensuing decade 2001–2011, the subject of this article, housing markets increasingly priced-in access to the CBD and inner city jobs, transport, facilities and services, such that house prices in Australian cities now follow a classic bid rent curve, with significantly higher prices at the centre and much lower prices on the rapidly developing fringe (Yates, 2010). The reasons are complex and include a move from manufacturing industries located in middle and outer suburbs to knowledge and service industries located in and around CBDs (Randolph and Holloway, 2005), and poor access to public transport in some middle and many outer suburban areas relative to inner cities as infrastructure has lagged behind urban growth (Dodson and Snipe, 2007). High rates of in-migration (including from international students) throughout the decade contributed to sustained demand for housing, particularly in the inner and well-located middle suburbs (Fincher and Shaw, 2009). The result was further intensive development on high-priced urban land, with an increase in high-rise apartments in the central core and some inner suburbs, and more medium density dwellings elsewhere in established middle and outer suburbs,² as well as new developments in growth areas at the urban fringe.

Whilst there had been prior periods when the activities of rental investors were important, such as in the 1960s and 1970s apartment building boom (Archer, 1978), a defining feature of Australian urban housing markets in the 2000s is the financialisation of housing, in particular households buying investment properties in established areas of the city. One of the key indicators of housing market activity in Australia is the percentage of housing loans for investment purposes and owner occupation, the latter having traditionally dominated lending practices. Loans to investors increased from around 25 per cent in the mid-1990s to just over 40 per cent of new housing loans (excluding re-financing) in 2011, as shown in Figure 1. Although direct comparisons are difficult, this appears to be well in excess of the percentage of loans to Buy to Let landlords in the UK in 2011, which totalled around 10 per cent of gross mortgage lending in that year (CML, 2013).

Figure 1.

Investor housing finance as a percentage of all housing finance* and the proportion of investor finance used for the construction of new dwellings: Australia, 1992–2017.

Favourable tax settings have facilitated this increase in household rental investment in Australia, including a 50 per cent discount on capital gains tax for rental properties after 12 months’ ownership, introduced in 1999, and greater take up of longstanding ‘negative gearing’ of losses on rental properties against other income to reduce personal income tax. These tax advantages for household investors are particularly attractive within the context of broader economic factors, such as a decline in the cost of borrowing due to lower interest rates and greater capacity to borrow against investors’ owner occupied housing as house prices increased. Importantly, much of this investment was in established dwellings, rather than new dwellings, as shown in Figure 1. In Australian cities, very few dwellings are built specifically for rental, and established dwellings can move seamlessly between ownership and rental and vice versa (Hulse and Yates, 2017). During the decade 2001–2011, household investors became a more important driver of urban housing markets but the geography of this type of rental investment is little understood, a gap that this article addresses through a detailed empirical exploration of disadvantaged areas in Melbourne, Australia, in 2001–2011.³

Research approach

The research approach was to investigate at a suburb level (see below) the relationship between changes in key socio-economic variables and indicators of housing market change over time. The first step was to identify suburbs in Melbourne that in 2001 were relatively disadvantaged, using population-based measures of socio-economic disadvantage, and to reapply this method in 2011. This process enabled identification of suburbs where relative socio-economic disadvantage declined, persisted or emerged during this period. The second stage involved detailed investigation of housing market change in suburbs where disadvantage persisted between 2001 and 2011.

Choice of spatial unit is critical in analysing changes in relative socio-economic disadvantage and housing markets (Darcy, 2010). The research focused on the suburb level, specifically the 2011 State Suburb (SSC) as defined by the national statistics agency Australian Bureau of Statistics (ABS). SSCs (henceforth suburbs) were chosen because they are large enough to enable analysis of population characteristics and housing markets but small enough not to ‘dilute’ any spatial concentrations of socio-economic disadvantage.⁴

Within Melbourne, suburbs that were disadvantaged in 2001 were selected using an established index of disadvantage: the ABS Socio-Economic Index for Areas (SEIFA), Index of Relative Socio-Economic Disadvantage (IRSD). The IRSD is an ABS product developed to calibrate the socio-economic status of localities based on attributes of residents, using data from the five-yearly Census of Population and Housing, and is widely used by academics and policy makers for this purpose. The approach aggregated up from the smallest ABS Census units – Collection Districts (CDs) in 2001 and Statistical Area Level 1s (SA1s) in 2011 – to suburbs (SSCs using 2011 boundaries).⁵ ‘Disadvantaged suburbs’ were identified where at least 50 per cent of the constituent CDs had lowest quintile IRSD Australian rankings in 2001, with some adjustment to allow for a minimum population threshold and inclusion of areas with a large number of people living in disadvantaged CDs (Hulse et al., 2014). Subsequently, the same set of rules was applied using SEIFA IRSD for 2011 to identify: persistent disadvantage where suburbs were identified using this approach in both 2001 and 2011; declining disadvantage where suburbs were identified in 2001 but not 2011, and; emerging disadvantage where suburbs were not identified as disadvantaged in 2001 but were so in 2011.

Using the SEIFA IRSD in this way is not without its limitations. In particular, the composition and weighting of variables in the IRSD is reviewed in each Census. The ABS warns against comparing IRSD scores across Censuses, due to changes in the variables and relative weighting, and suggests instead the comparison of spatial unit deciles and quintiles, an approach taken in this research. As in any research of this type, the method identified areas with a concentration of people with characteristics of relative socio-economic disadvantage but not all people living in these areas had such characteristics.

The second stage of the research involved detailed original analysis of Melbourne housing market data. In addition to analysis of customised Census data relating to households, dwellings and housing tenure in 2001 and 2011, this involved assembling, cleaning and analysing two administrative data sets at two capture points (2001 and 2011). House sales price data are collated by the Valuer-General Victoria at the time of all dwelling sale transactions (for stamp duty purposes). Data on rents are collated by the state government authority that collects bonds and associated information on new lettings. Rent levels in this context refer to current market ‘entry rents’, that is, the amount paid at the beginning of the tenancy, which differs from rent levels for all new and ongoing tenancies. To facilitate the analysis, house sale price and entry rent data were geocoded to street address and then allocated and aggregated to match 2011 suburb boundaries to enable spatial comparison with variables sourced from the ABS Census.

Spatial patterns of socio-economic disadvantage in Melbourne, 2001–2011

Using the method described above, 47 Melbourne suburbs were identified as being disadvantaged in 2001, comprising 9.5 per cent of the city’s suburbs (with a resident population of 555,350 people or 16.5 per cent of the 2001 Melbourne population). Importantly, disadvantaged suburbs were mainly contiguous with other disadvantaged suburbs and clustered spatially in established, built-up, middle and outer suburban areas in Melbourne’s west, northwest and southeast (Figure 2). Reapplying the method in 2011, 83 per cent (39 of 47) of these suburbs were still disadvantaged, indicating persisting disadvantage. These suburbs were mainly in established areas in the three clusters and had experienced lower population and household growth than Melbourne generally. Some of these suburbs were well located in terms of transport and access to the CBD, particularly in the west and northwest of the city.

Figure 2.

Change in socio-economic disadvantage in Melbourne suburbs 2001–2011.

Eight suburbs which were identified as disadvantaged in 2001 were no longer so in 2011. The spatial location of these suburbs of declining disadvantage is shown in Figure 2 and is consistent with gentrification in the case of Melbourne’s inner western suburbs but not so obviously elsewhere. In particular, there was some indication of improving socio-economic status in parts of the south-eastern cluster of disadvantaged suburbs, which are at considerable distance from the CBD, indicating that factors other than conventional gentrification processes based on proximity to the CBD are involved.

To complete the picture, six additional suburbs of emerging disadvantage were newly identified in 2011. These were not randomly distributed; they were either contiguous with but at a greater distance from the CBD than clusters of suburbs of persistent disadvantage and/or at the end of Melbourne’s radial railway system. The emergence of these areas is consistent with a thesis of the suburbanisation of disadvantage in areas that already had concentrated disadvantage using population based measures. In other words, the suburbanisation of disadvantage does not apply evenly across the city.

The focus of the remainder of this article is on those suburbs where disadvantage persisted over the 2001–2011 decade. The persistence of socio-economic disadvantage in most (39) suburbs in the three main spatial clusters occurred in the context of a decade of economic growth with real growth in household incomes at the bottom of the income distribution as well as at the top end (Greenville et al., 2013). Whilst growth slowed after the Global Financial Crisis of 2008–2009, there was no Great Recession as in many other countries and there was some improvement in economic conditions across Melbourne’s disadvantaged suburbs. Socio-economic disadvantage is relative, however, and the suburbs of persistent disadvantage still rated highly on indicators of socio-economic disadvantage. These included significantly higher levels of unemployment; lower levels of educational achievement; higher rates of young people not in employment, education or training; a higher percentage of people in low status occupations; and a higher rate of low-income households, compared to Melbourne in both 2001 and 2011 (see Table 1).

Table 1.

Change in key socio-economic variables: Suburbs of persistent disadvantage compared to Melbourne, 2001–2011.

	Suburbs of persistent disadvantage	Melbourne
Unemployment
Rate (%) 2001	12	7
Rate (%) 2011	10	6
Highest level of education is Year 10 (aged 25–44 yrs)
2001 %	26	14
2011 %	14	7
Not in emp, educ, training or the labour force (aged 15–24 yrs)
2001 %	15	9
2011 %	13	7
Low skilled/status occupations*
2001 %	56	38
2011 %	48	30
Low income households ^{^}
2001 %	40	29
2011 %	47	33
Overall household growth rate
2001 to 2011 (%)	7	18

Notes: *Low skilled/status occupations were defined for 2011 using the occupational status scale, ‘the Australian Socioeconomic Index 2006 (AUSEI06)’; and for 2001 using the previous version of the scale, ‘ANU4’ (see McMillan et al., 2009). According to the authors, these scales are comparable over time.

Low income households are those households in the bottom two quintiles of the Australia-wide (non-equivalised) household income distribution.

Source: Derived from ABS Census of Population and Housing, 2001 and 2011.

Disadvantaged suburbs in Melbourne had lower rates of household growth than the city as a whole in 2001–2011. Opportunities for household growth in such areas depends on gradual intensification of housing stock, rather than more rapid development of new housing on ‘greenfield’ sites on the urban fringe or more intense development of the inner suburbs. The household growth rate in suburbs of persistent disadvantage was less than half the Melbourne growth rate (Table 1). This is important context for considering housing market changes, as is discussed next.

Housing market dynamics in suburbs of persistent disadvantage, Melbourne 2001–2011

Not surprisingly, in view of continuous economic growth allied with household growth, the latter due in part to high levels of in-migration, house prices in Melbourne grew strongly between 2001 and 2011. In suburbs of persistent disadvantage, however, real (inflation-adjusted) prices for both detached and other (non-detached) dwelling types increased by 95 per cent: greater than the respective Melbourne medians. As a result, median sale prices for both dwelling types were considerably closer to city medians in 2011 than in 2001 (Table 2). Consequently, the percentage of ‘affordable’ sales in these suburbs (those sale prices in the lowest quartile of all Melbourne sale prices) declined markedly, although the majority of sales of ‘other’ dwellings were still ‘affordable’ in 2011 (59 per cent).

Table 2.

Change in sale prices for detached and other dwellings in suburbs of persisting disadvantage and Melbourne, 2001 and 2011.

	Detached dwelling sales		‘Other’ dwelling sales
	Suburbs of persistent disadvantage	Melbourne	Suburbs of persistent disadvantage	Melbourne
Real median sale price (AU$2011)
2001	AU$195,650	AU$292,500	AU$163,800	AU$304,200
2011	AU$382,000	AU$490,000	AU$320,000	AU$432,500
AU$ change 2001–2011	AU$186,350	AU$197,500	AU$156,200	AU$128,300
% change 2001–2011	95	68	95	42
% of Melbourne median 2001	67	100	54	100
% of Melbourne median 2011	78	100	74	100
Avg. ann. change 2001–2011	6.9	5.3	6.9	3.6
Affordable sales (in Melbourne’s lower quartile)
% affordable 2001	59	25	76	25
% affordable 2011	47	25	59	25

Source: Calculated by authors from Valuer-General Victoria property sales data, 2001 and 2011.

Increases in real (inflation-adjusted) median entry rents were considerably less than for increases in median sale prices in suburbs of persistent disadvantage (Table 3). Median entry rents for detached dwellings, for example, increased by 28 per cent, considerably less than sale prices for detached dwellings (Table 2). Melbourne rents were less dispersed than dwelling sale prices in both 2001 and 2011, with entry rents in disadvantaged suburbs being closer to city medians in both 2001 and 2011 than sale prices. The implication is that there was a greater ‘affordability discount’ for households buying in disadvantaged suburbs than for those renting. There was also a greater ‘affordability discount’ for both purchasers and renters of ‘other dwellings’ (i.e. not detached) in disadvantaged suburbs than there was for detached houses.

Table 3.

Change in entry rents for detached and other dwellings in suburbs of persistent disadvantage and Melbourne, 2001 and 2011.

	Detached dwelling entry rents		‘Other’ dwelling entry rents
	Suburbs of persistent disadvantage	Melbourne	Suburbs of persistent disadvantage	Melbourne
Real median entry rent (AU$2011)
2001	AU$221	AU$273	AU$181	AU$247
2011	AU$300	AU$350	AU$277	AU$350
AU$ change 2001–2011	AU$79	AU$77	AU$96	AU$103
% change 2001–2011	36	28	53	42
% of Melbourne median 2001	81	100	73	100
% of Melbourne median 2011	86	100	79	100
Avg. ann. change 2001–2011	3.1	2.5	4.4	3.5
Affordable entry rents (in Melbourne’s lower quartile)
% affordable 2001	67	25	69	25
% affordable 2011	52	25	56	25

Source: Calculated by authors from Victorian Residential Tenancies Bond Authority data, 2001 and 2011.

In terms of expected growth, the number of private renter households in Melbourne increased by 22.6 per cent more than might be expected over the 2001–2011 period (Table 4, Melbourne index score minus 100). Change would be ‘as expected’ if the number of households in each tenure group increased in line with the overall household growth rate of Melbourne (2001–2011), resulting in an unchanged tenure structure in 2011. Table 4 shows, however, that change within Melbourne tenure groups was not ‘as expected’, with the absolute decline in the number of owner households reflected in a very low index score and the number of purchaser households increasing by nearly 26 per cent more than expected. As such, the tenure structure of Melbourne did change over the decade, at the expense of owner households.

Table 4.

Indexed growth^a in tenure groups in suburbs of persistent disadvantage (PDS), relative to expected growth and Melbourne growth, 2001–2011.

	PDS growth index	Melbourne growth index	Difference (to Melbourne)
Private renter households	133.1	122.6	10.5
Owner households	73.4	75.6	–2.6
Purchaser households	126.3	125.8	0.5

Notes: ^aThe indexed growth figures presented here are derived from the difference between two figures: the expected number of households in the tenure group in 2011 if it increased by the overall household growth rate of the suburb or city; and the actual number of households in that tenure group in 2011. This score varies around 100: with 100 representing expected growth (in line with household growth); a score below 100 represents less than expected growth or absolute decline; and scores greater than 100 represent higher than expected growth.

Source: Derived from ABS Census of Population and Housing, 2001 and 2011.

Table 4 also shows how change unfolded in the main tenure groups in suburbs of persistent disadvantage compared with Melbourne. Change in the number of both owner households and purchaser households was essentially in line with that which occurred across Melbourne over the decade (see final column). Private renter households, however, not only increased at a greater rate than the overall household growth rate of persistently disadvantaged suburbs (by 33 per cent) but, importantly, also at a greater rate than that which occurred across Melbourne as a whole (by 10.5 percentage points).

These findings indicate that there was a disproportionate increase in rental investment and private rental in persistently disadvantaged suburbs than for the city as a whole, raising questions about ways in which increased financialisation of housing through rental investment may relate to the persistence of population socio-economic disadvantage, which we address next.

Discussion

The gentrification literature suggests that increased prices and rents result in direct or indirect displacement of low socio-economic households, whilst the suburbanisation of disadvantage literature would also anticipate displacement of low socio-economic residents to lower priced areas further from the CBD as house prices and rents increased. Whilst both of these are evident in a small number of suburbs of declining disadvantage and emerging disadvantage respectively, there is little evidence of these processes in suburbs of persistent disadvantage in Melbourne, or at least, if households have been displaced, people with similar socio-economic characteristics have replaced them. In interpreting these findings, it appears that the financialisation of housing, evident in increased household investment, has played an important role. Although the growth in private rental was experienced across Melbourne in 2001–2011, there was a disproportionate increase in persistently disadvantaged suburbs. This raises two important questions: i) why did households invest in suburbs of persistent disadvantage?; and ii) why did displacement of low socio-economic residents not occur as entry rents increased relative to city medians?

Whilst the research method deployed in this study does not enable a direct answer to these questions, we draw on other research which might provide some answers. The rather limited research into the profile, motivations and strategies of rental investors in Australia suggests that household investors are motivated by prospects of capital gain allied, where possible, with good rental yields. Between 2001 and 2011, real dwelling prices in these disadvantaged suburbs increased at higher annual rates than for Melbourne, providing significant opportunities for capital gain. At the same time, median rents moved closer to city medians, indicating the potential for good revenue streams. The ‘package’ of higher percentage capital gains relative to purchase price and higher rental yields for investors in private rental housing are likely to have been greater in disadvantaged suburbs than elsewhere in the city.

It could be argued that this explanation is a contemporary reworking of Smith’s ‘rent gap’ discussed earlier, with investors moving in because housing/land prices are under capitalised by their current use, namely low price home ownership. However, rental investors in Australia⁶ are predominantly ‘mums and dads’ who are relatively unsophisticated in their investment strategies (Berry, 2000; Seelig et al., 2009) and who often prefer to invest in single detached dwellings or small medium-density developments at a price they can afford and with which they can exercise control over their investment. Suburbs of persistent disadvantage in Melbourne comprise detached dwellings and small apartment complexes which enable such control as well as the possibility of resale in ‘thick’ markets in terms of prices (Hulse and Yates, 2017). The re/investment that Smith (1979) identified in inner cities as a driver of gentrification was by a much wider range of institutional actors including financial institutions, developers, builders, governments and corporate landlords (Smith, 1979) who could be expected to be more financially savvy.

On the second question, there may well be non-financial reasons why low socio-economic households stay or move into in these areas notwithstanding increased prices and rents. Recent research suggests, for example, that suburbs of persistent disadvantage play an important role as gateway suburbs for migrants and refugees, providing valuable social connections and cultural infrastructure (Easthope et al., 2017). Investigating these factors in more depth requires different methodologies.

Reflecting on our research findings, we suggest a new concept, investification, referring to the process by which disproportionately high levels of household investor purchases in disadvantaged suburbs contribute to higher prices and rents and to the persistence of socio-economic disadvantage as properties are rented on the private market to low socio-economic households, indicating replacement rather than displacement. Investification refers to a process in which households are both consumers and investors, buying into areas that they may not want to live in. Understanding this process requires investigation of their consumption and investment practices.⁷

Conclusion

The contemporary story of disadvantaged suburbs in Melbourne, as in Australia’s other major cities, is not primarily about a cycle of decline, abandonment and reinvestment in inner cities, which is well covered in the international literature on gentrification. It is about suburban disadvantage in established areas further from city centres, as indicated in the suburbanisation of disadvantage literature. There is some evidence for gentrification in the small number of suburbs which were disadvantaged in 2001 but not in 2011, particularly those in the inner north and western suburbs, which were relatively close to Melbourne’s CBD. There is also some evidence of further suburbanisation of disadvantage in the handful of suburbs of emerging disadvantage which were continguous to, but further from, the CBD than other disadvantaged suburbs. This article has focused on the trajectory of the majority of suburbs, which indicated persistent socio-economic disadvantage between 2001 and 2011 using population based measures but ‘improving’ housing markets as indicated by median sale prices and entry rents that moved closer to city medians in 2001–2011. In the context of increased financialisation of housing during the decade, housing markets in persistently disadvantaged suburbs were increasingly driven by the activities of household rental investors.

The increase in household level rental investment is a contemporary phenomenon not well covered by the existing literature on disadvantaged areas, which focuses mainly on owner occupation and social rental. We have coined the term ‘investification’ to capture this process, drawing attention not only to the increasing importance of the scale of private rental investment to urban housing market dynamics but also to the effect of spatial patterns of rental investment in contributing to the geography of disadvantage (and potentially advantage) in contemporary urban settings. The concept of investification raises questions about the persistence of socio-economic disadvantage in some established middle/outer suburbs, which open up areas for future research. What are the motivations, preferences and decisions of rental investors in respect of different urban geographies? Why do low socio-economic status residents not respond to increased prices/rents by moving to cheaper areas further from the CBD as indicated by the suburbanisation of disadvantage literature?

The contribution of the current article is in drawing together scholarship on housing market dynamics and explanations of spatial patterns of socio-economic disadvantage. Introducing the concept of investification, and starting to explore the ways in which this plays out spatially, enables a more nuanced explanation of contemporary urban change in Australia, adding to prior work on gentrification and the suburbanisation of disadvantage. This approach may well be relevant to other contexts in view of a general increase in investment in housing by household investors rather than owner occupiers and increasing rental in countries such as the UK. It has the potential to open up new ways of understanding the housing market and other processes, which drive spatial patterns of urban socio-economic disadvantage and advantage.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This research was funded by the Australian Research Council, DP150102582.

Notes

References

ABS (2012) Regional Population Growth, Australia, 2011. Cat No 3218.0. Canberra: Australian Bureau of Statistics.

ABS (2014) Regional Population Growth, Australia, 2012–13. Cat No 3218.0. Canberra: Australian Bureau of Statistics.

Archer

(1978) The market for new multi-unit housing in Sydney and Melbourne. In: Department of Housing and Construction Housing Economics. Canberra: Australian Government Publishing Service, pp. 172–187.

Badcock

(1994) Snakes or ladders? The housing market and wealth distribution in Australia. International Journal of Urban and Regional Research 18(4): 609–627.

Baum

Gleeson

(2010) Space and place: Social exclusion in Australia’s suburban heartlands. Urban Policy and Research 28(2): 135–159.

Baum

Haynes

Gellecum

et al . (2006) Advantage and disadvantage across Australia’s extended metropolitan regions: A typology of socioeconomic outcomes. Urban Studies 43(9): 1549–1579.

Bell

Schuurman

Hayes

(2007) Using GIS-based methods of multicriteria analysis to construct socioeconomic deprivation indices. International Journal of Health Geographics 6(17): 1–19.

Berry

(2000) Investment in rental housing in Australia: Small landlords and institutional investors. Housing Studies 15(5): 661–681.

Bridge

Watson

(eds) (2011) A Companion to the City. Oxford: Blackwell Publishing.

10.

Burgess

(1929) Urban areas. In: Smith

White

(eds) Chicago: An Experiment in Social Science Research. Chicago, IL: University of Chicago Press, pp. 113–138.

11.

Burnley

(1980) The Australian Urban System: Growth, Change, and Differentiation. Melbourne: Longman Cheshire.

12.

Council of Mortgage Lenders (CML) (2013) Buy-to-let lending £16.4 billion in 2012. Available at: https://www.cml.org.uk/news/press-releases/3423/ (accessed 8 May 2017).

13.

Crook

Kemp

(2014) Private Rental Housing: Comparative Perspectives. Cheltenham: Edward Elgar Publishing.

14.

Darcy

(2010) De-concentration of disadvantage and mixed income housing: A critical discourse approach. Housing, Theory and Society 27(1): 1–22.

15.

Davidson

(2008) Spoiled mixture: Where does state-led gentrification end? Urban Studies 45(12): 2385–2405.

16.

Davidson

Lees

(2010) New-build gentrification: Its histories, trajectories, and critical geographies. Population, Space and Place 16(5): 395–411.

17.

Dodson

Snipe

(2007) Oil vulnerability in the Australian city: Assessing socioeconomic risks from higher urban fuel prices. Urban Studies 44(1): 37–62.

18.

Easthope

Stone

Cheshire

(2017) The decline of ‘advantageous disadvantage’ in gateway suburbs in Australia: The challenge of private housing market settlement for newly arrived migrants. Urban Studies. Epub ahead of print 26 April 2017. DOI: 10.1177/0042098017700791.

19.

Engels

(1999) Property ownership, tenure and displacement: In search of the process of displacement. Environment and Planning A 31: 1473–1495.

20.

Fincher

Shaw

(2009) The unintended segregation of transnational students in central Melbourne. Environment and Planning A 41(8): 1884–1902.

21.

Forrest

Hirayama

(2015) The financialisation of the social project: Embedded liberalism, neoliberalism and home ownership. Urban Studies 52(2): 233–244.

22.

Gleeson

(2006) Australian Heartlands: Making Space for Hope in the Suburbs. Crows Nest: Allen & Unwin.

23.

Greenville

Pobke

Rogers

(2013) Trends in the Distribution of Income in Australia. Productivity Commission Staff Working Paper. Canberra: Productivity Commission.

24.

Groenhart

(2014) Understanding the spatial impacts of direct and indirect government housing expenditure. Final Report 234. Melbourne: Australian Housing and Urban Research Institute.

25.

Hamnett

(2000) Gentrification, postindustrialism, and industrial and occupational restructuring in global cities. In: Bridge

Watson

(eds) A Companion to the City. Oxford: Blackwell Publishing, pp. 331–341

26.

Hamnett

Whitelegg

(2007) Loft conversion and gentrification in London: From industrial to postindustrial land use. Environment and Planning A 39(1): 106–124.

27.

Holliday

Dwyer

(2009) Suburban neighborhood poverty in U.S. metropolitan areas in 2000. City & Community 8(2): 155–176.

28.

Hoyt

(1939) The Structure and Growth of Residential Neighbourhoods in American Cities. Washington, DC: US Government Printing Office.

29.

Hulse

Yates

(2017) A private rental sector paradox: Unpacking the effects of urban restructuring on housing market dynamics. Housing Studies 32(3): 253–270.

30.

Hulse

Pawson

Reynolds

et al . (2014) Disadvantaged places in urban Australia: analysing socio-economic diversity and housing market performance. Final Report Series No 225, Australian Housing and Urban Research Institute, Melbourne, pp.1–96.

31.

Hunter

(2014) Poverty in suburbia: A Smith Institute study into the growth of poverty in the suburbs of England and Wales. London: Smith Institute.

32.

Johnston

(1973) Urban Residential Patterns. London: Bell.

33.

Kasarda

(1990) Structural factors affecting the location and timing of urban underclass growth. Urban Geography 11(3): 234–264.

34.

Kemp

(2015) Private renting after the global financial crisis. Housing Studies 30(4): 601–620.

35.

Kendig

(1979) New Life for Old Suburbs: Postwar Land Use and housing in the Australian Inner City. Sydney: George Allen and Unwin.

36.

Kendig

(1984) Gentrification in Australia. In: Palen

London

(eds) Gentrification, Displacement and Neighbourhood Revitalisation. Albany, NY: State University of New York Press, pp. 235–255.

37.

Kneebone

(2014) The Growth and Spread of Concentrated Poverty, 2000 to 2008–2012. Washington, DC: Brookings Institution Press.

38.

Kneebone

Garr

(2010) The suburbanization of poverty: Trends in metropolitan areas, 2000 to 2008. Washington, DC: Brookings Institution Press.

39.

Lees

(1994) Rethinking gentrification: Beyond the positions of economics and culture. Progress in Human Geography 18(2): 137–150.

40.

Lees

Slater

Wyly

(2008) Gentrification. New York: Routledge.

41.

Ley

(1986) Alternative explanations for inner city gentrification: A Canadian assessment. Annals of the Association of American Geographers 76(4): 521–535.

42.

Ley

(1994) Gentrification and the politics of the new middle class. Environment and Planning D: Society and Space 12: 53–54.

43.

Ley

(1996) The New Middle Class and the Remaking of the Central City. Oxford: Oxford University Press.

44.

Logan

(1982) Gentrification in inner Melbourne: Problems of analysis. Australian Geographical Studies 20: 65–95.

45.

Logan

(1985) The Gentrification of Inner Melbourne. Brisbane: University of Queensland Press.

46.

McMillan

Beavis

Jones

(2009) The AUSEI06. A new socioeconomic index for Australia. Journal of Sociology 45(2): 123–149.

47.

Madden

(2003) The changing spatial concentration of income and poverty among suburbs of large US metropolitan areas. Urban Studies 40(3): 481–503.

48.

Maher

(1985) Building activity and socio-economic change in inner Melbourne: 1961–1981. Urban Policy and Research 3(1): 3–12.

49.

Maher

(1994) Residential mobility, locational disadvantage and spatial inequality in Australian cities. Urban Policy and Research 12(3): 185–191.

50.

Murphy

(2007) The suburban ghetto: The legacy of Herbert Gans in understanding the experience of poverty in recently impoverished American suburbs. City and Community 6(1): 21–37.

51.

Paccoud

(2017) Buy to let gentrification: Extending social change through tenure shifts. Environment and Planning A 49(4): 839–856.

52.

Park

Burgess

McKenzie

(1925) The City. Chicago, IL: University of Chicago Press.

53.

Pawson

Herath

(2015) Dissecting and tracking socio-spatial disadvantage in urban Australia. Cities 44(April): 73–85.

54.

Randolph

Freestone

(2012) Housing differentiation and renewal in middle-ring suburbs: The experience of Sydney, Australia. Urban Studies 49(12): 2557–2575.

55.

Randolph

Holloway

(2005) The suburbanization of disadvantage in Sydney. New problems, new policies. Opolis 1(1): 49–65.

56.

Randolph

Tice

(2014) Suburbanising disadvantage in Australian cities: Socio-spatial change in an era of neo-liberalism. Journal of Urban Affairs 36(S1): 1–16.

57.

Rofe

(2003) ‘I want to be a global’: Theorising the gentrifying class as an emergent elite global community. Urban Studies 40(12): 2511–2526.

58.

Schwartz

(2015) Housing Policy in the United States. 3rd edn. New York: Routledge.

59.

Seelig

Thompson

Burke

et al . (2009) Understanding what motivates households to become and remain investors in the private rental market. Final Report no 130. Melbourne: Australian Housing and Urban Research Institute.

60.

Shaw

Hagemans

(2015) ‘Gentrification without displacement’ and the consequent loss of place: The effects of class transition on low-income residents of secure housing in gentrifying areas. International Journal of Urban and Regional Research 39(2): 323–341.

61.

Shevky

Bell

(1955) Social Area Analysis. Stanford Sociological Series No 1. Stanford, CA: Stanford University Press.

62.

Slater

(2006) The eviction of critical perspectives from gentrification research. International Journal of Urban and Regional Research 30(4): 737–757.

63.

Slater

(2011) Gentrification of the city. In: Bridge

Watson

(eds) The New Blackwell Companion to the City. Oxford: Blackwell Publishing, pp. 571–585.

64.

Smith

(1979) Toward a theory of gentrification: A back to the city movement by capital, not people. Journal of the American Planning Association 45(4): 538–548.

65.

Smith

(1982) Gentrification and uneven development. Economic Geography 58: 139–155.

66.

Wulff

Lobo

(2009) The new gentrifiers: The role of households and migration in reshaping Melbourne’s core and inner suburbs. Urban Policy and Research 27(3): 315–331.

67.

Yates

(2010) Housing policies and wealth inequality. In: Tomlinson

(ed.) Australia’s Unintended Cities, The Impact of Housing on Urban Development. Melbourne: CSIRO Publishing, pp. 83–102.