Australia's private rental market: the supply of, and demand for, affordable dwellings

AIMS In 2006 only 37 per cent of private renters with household incomes in the lowest 40 per cent of the national income distribution accessed affordable housing (Wulff et al. 2009). This Final Report aims to provide relevant information to assist policy-makers in addressing this situation. Specifically, the report addresses the following: What are the characteristics of low and very low income households in the private rental sector? 'Low income' refers to households with incomes in the second lowest income quintile (21% to 40% of all incomes nationally) and 'very low' refers to households with incomes in the bottom income quintile (20% or lower). Which low and very low income private renter households gain access to the affordable stock and which miss out? Is there a systematic pattern that might help to explain this outcome? For low and very low income private renter households, what is the shortage or surplus of affordable stock and the true shortage (based on available stock) for each capital city and large regional centre? Lower income households face two main difficulties in their efforts to access affordable rental dwellings: The first relates to the actual supply of affordable dwellings; in other words, whether the stock of affordable dwellings, irrespective of who occupies these dwellings, is sufficient to meet the demand from low income renters. The second relates to the availability of the stock. In the competitive private market, not all affordable stock necessarily goes to households who need it the most. For a number of reasons, middle to high income households may occupy this stock, thereby effectively removing dwellings from the affordable supply for low income renters. METHODS This analysis draws upon a special request ABS census matrix related to all occupied private dwellings in Australia. All missing values for dwelling rents and household incomes have been imputed by the Australian Bureau of Statistics in order to ensure the reliability and quality of the results. Five gross unequivalised household income ranges, which represent household income quintiles, are used. Weekly rent categories are matched against the household income quintiles to represent no more than 30 per cent of the corresponding income category. Although 'low income' households are defined generally for policy purposes as falling into the bottom 40 per cent of the national household income distribution, this analysis disaggregates between Q1 and Q2 private renter households in presenting results. This study measures the actual supply of affordable dwellings by calculating the number of private rental dwellings that can be rented to households in the bottom two income quintiles for an amount no greater than 30 per cent of their household income. The available supply is measured by subtracting from the affordable supply any dwellings that are occupied by higher income households and are therefore not available to low income households. For further insight into the rental affordability problems faced by Q1 and Q2 private renter households, this chapter has focused not only on those who can pay an affordable (30% of household income), but also divide the rest into two broad groups: unaffordable rent (31% to 50% of household income) and severely unaffordable rent (over 50% of household income). FINDINGS There was a shortage of affordable housing stock available to those on very low and low incomes in 2006. Very low income renters (Q1) face affordability problems both because of an absolute shortage of housing that is affordable and because a significant proportion of affordable stock is not available to them because it is occupied by higher income households. Low income renters (Q2), on the other hand, do not face an absolute shortage of affordable housing. This group, however, experience affordability problems because the affordable stock is occupied by other income groups and is, therefore, unavailable to them. For the 267 819 very low income private renter households (Q1), the results indicate that, in 2006, there was an absolute shortfall of 138 000 affordable dwellings. The shortage increased to 211 000 dwellings that were both affordable and available once use by higher income households was taken into account. Seventy-nine per cent of very low income renters miss out on affordable rental housing. For the 360 467 low income private renter households (Q2), the results show a surplus of 527 533 affordable dwellings. Importantly, however, use of these dwellings by other households turns this surplus into a shortage of 87 000 dwellings that are affordable and available. Overall, 24 per cent of low income private renters face affordability problems that are a result of a shortage of affordable and available housing. Affordable and available private rental supply varied by capital cities and large regional centres. For Q1 private renter households, the greatest shortages, both in absolute and relative terms, in affordable and available private rental dwellings were observed in Sydney (-44,500; 93% miss out), Melbourne (-40,200; 87% miss out) and Brisbane (-19,100; 87% miss out). All eight large regional centres studied had high levels of Q1 private renter households missing out on affordable rental housing. It was particularly pronounced in the Gold Coast (93%) and the Sunshine Coast (88%) in Queensland. However, Geelong and Launceston had relative lower levels of shortages (64% and 63% respectively) than the others. This was associated with a larger affordable supply in these two regional centres compared with the other regional centres. For Q2 private renter households, the geographic pattern is similar to that of Q1. More severe shortages in affordable and available housing were observed in Sydney, Melbourne and Brisbane than the smaller cities (with the exception of Canberra). The pattern among the regional centres also parallels that documented for Q1 private renters. Relatively fewer Q2 households miss out on affordable dwellings in Geelong (7%) and Launceston (6%) compared to the rest. Notably, a much greater proportion missed out in the regional centres of the Gold Coast (54%) and the Sunshine Coast (43%) compared with 31 per cent in the capital city of Brisbane. Only 21 per cent of very low income private renter households pay an affordable rent; 19 per cent pay a severely unaffordable rent. Of the 267 819 Q1 private renters, 57 313 households pay an affordable rent that was less than 30 per cent of their gross household income; an almost equal number of households (51 421) pay a severely unaffordable rent of over 50 per cent of their gross household income. Age, living arrangements and number of children play an important role in determining access to affordable rental housing. In general, elderly renters, people living alone, and households without children present are more likely to find an affordable rental dwelling. In contrast, very few younger renters or families with two or more children access affordable rental housing. About 33 per cent of young childless couples; 36 per cent of couples with children and 21 per cent of single parents pay over 50 per cent of their gross household income to access a rental dwelling. The vast majority (79%) of Q2 private renter households can find affordable rental housing. Severe unaffordability is not a major concern for this group (3% pay over 50 per cent of their gross household income on rent). The social demographic differential in terms of accessing affordable housing was not noticeable among Q2 private renters. POLICY IMPLICATIONS The results of this study highlight the ongoing difficulties faced by very low income households (Q1) for whom there are just not enough affordable private rental dwellings. The shortage of 211 000 dwellings that are both affordable and available for very low income households highlights the importance of ensuring that the affordable stock that does exist is somehow targeted in national affordability schemes, such as NRAS, to those most in need of it. Policies aimed at increasing the supply of affordable dwellings could use the findings of this study to determine the affordability variations among different household types. For example, the need for small dwellings for singles should be balanced gainst the severe housing needs of the larger families with children. This study provides the information (on relative need and absolute numbers of households) that is required to develop appropriate housing targets. In this light, the Canadian concept of 'suitability', which assesses household size and composition against dwelling type, may be worth considering in Australia. Shortages in the supply of affordable and available dwellings are felt most keenly in the major capital cities. This study found, however, that Geelong and Launceston have better affordability outcomes than either the other regional centres or major capital cities. It is worth studying these two locations in greater detail to help identify housing market or economic factors that reduce the affordability problems in these places. Methodological policy implications are provided in Appendix 2. In brief, future research should consider disaggregating the second household income quintile (Q2) into two deciles and possibly supplementing this census-based approach between census years with other national survey data on the private rental sector. This may improve the measures of affordability and bring Australia's approach in line with what is international best practice. Further, some consideration may be given to investigating the use of area median income levels in determining the affordable supply compared with the number of households in need in particular locations.