Projections of Housing Demand in Australia, 2008-2038: Housing Needs of Older Australians Narrative Report



This report was prepared for National Housing Supply Council. It is not a Treasury report.

The National Housing Supply Council commissioned latent demand projections for the 2008 and 2010 State of Supply Report from Professor Peter McDonald and Dr Jeromey Temple of the Australian Demographic and Social Research Institute (ADSRI), the Australian National University.

Related Content

• The National Housing Supply Council
• State of Supply Report 2008
• State of Supply Report 2010

Background

In addition to the underlying demographic process of fertility, mortality and migration, cohort effects ripple across time, influencing the propensity to marry, divorce and family formation more generally. This combination, in turn, influences living arrangements in the later life course. Until recently, projections of spatial variations in elderly living arrangements and households in Australia have been limited due to modelling complexity and data limitations (Rowland, 1997). The results herein, present projections of older persons households and demand for dwellings across 15 separate geographies using the net transition probability method.

This report provides a narrative description of results of the projection of future housing demand in the capital cities and balances of state for the eight States and Territories of Australia for the period, 2008-38.

This report focuses on the projection results for three groups of older Australians: (1.) aged 65-74, (2.) aged 75-84 and (3.) aged 85 years and over. The baseline housing data for the projections is obtained from the 2006 Census of Population and Housing and the baseline Estimated Resident Population data is from 30 June 2008.

Projection Methodology

The overview of the detailed projections methodology is given elsewhere (Temple and McDonald, 2009; McDonald, Kippen and Temple, 2007). However, in interpreting the following results regarding older persons demand for dwellings, several key points require reiteration:

• The projections are demand-side projections, with no account given to availability of land, the number of vacant dwellings, construction of new dwellings and affordability.
• The projections utilize the dwelling and tenure 'preferences' by age as estimated from the 2006 Australian Census of Population and Housing. Future cohorts, of course, may have very different preferences for types of dwellings, as well as facing very different pricing constraints etc.
• The usual caveat with demographic projections applies: results present a possible future based upon a restricted set of assumptions. There are exogenous policy shocks that may affect the utility of belonging to different living arrangements in the later life course. For example, American studies have shown that increases in income and social security payments as well as reforms to nursing home subsidies have given rise to a higher demand for independent living (McGarry and Schoeni, 2000; Costa, 1997; Hoerger, Picone and Sloan, 1996). Engelhardt, Gruber and Perry (2002) estimate that a 10% cut in social security in America would create a movement of 600,000 lone persons moving into shared living arrangements. Likewise, changes in Australian social and economic policy could hasten the transition probabilities toward living independently. But as data extracted from HaPPE shows, the proportion living in non private dwellings is relatively small in Australia, until advanced old age. This is consistent with the Australian government's policy of ageing in place, promoting independent living through community care services (Bishop, 1999). Although community care partially offsets government funded residential services, many carers and their employers bear a heavy financial and social cost given foregone time spent in the labour market, leisure and other activities (Rubin, 2002). Transition probabilities for independent living may be effected by the availability and public support for carers. Another factor that may effect the transition probabilities, particularly in the earlier life cycle, is housing prices (McDonald and Temple, 2004). However, in old age there is little evidence to suggest that living arrangement decisions are made on the basis of house prices, with the major determinants being demographic (Börsch-Supan, 1989).

Given these limitations, how should the results be interpreted and considered:

• Our approach is to project housing demand on the basis of current and recent trends in demand inputs. These demand projections should then be assessed in supply terms, that is, the results from the projections of demand for housing can be compared with existing and planned supply of housing and assessments made of what corrections for demand-supply discrepancies need to be made. Where meeting demand would create supply difficulties, consideration would need to be given to how this demand is re-directed. Do the projected households maintain their dwelling preference but change their location or do they change their dwelling preference within the location. The fact that supply cannot meet housing preferences could also conceivably lead to the household not being formed at all.
• Our method of projecting living arrangements and households has a number of advantages, particularly with reference to older persons. Firstly, this model has captured regional level heterogeneity in the compositional aspects of populations through using region specific estimates and projections of fertility, mortality, migration and living arrangement transitions. Second, providing strong support for this model, the net transition probabilities estimated from the 1991, 1996 and 2001 Australian Census of Population and Housing are highly comparable. Moreover, re-estimating the model using historical data shows the projections produced from the model to be highly reliable (McDonald and Temple, 2003). Additional comparisons show that in the majority of cases, the tempo of the age specific transition probabilities is highly consistent, although the quantum, as one would expect, differs considerably across the regions. Finally, by building upon the usual headship or propensity type models, the net transition approach specifically tracks cohort differences in living arrangement transitions.

Household Growth

Tables 1 – 3 display the projected growth in households between 2008-2023 and 2023-38 for age groups 65-74, 75-84 and 85 years and over respectively. The projection results are for the medium scenario (Scenario number 9 in HaPPE). The two main household types in older age (couples with no children and lone persons) are shown in addition to other living arrangement types. The number of households is shown in addition to the household growth for two time periods, 2008-2023 and 2023-2028. The household growth measures are ratios expressing the differing speeds of growth across the time periods. For example, between 2008 and 2023, the number of couple only households in Sydney (NSWCC), grew by 1.58 times compared to 1.42 times in the second period.

This simple example underscores a key result across this, and other tables; that is, the differential speed of 'ageing of the aged'. Across the tables for the youngest age group, the growth in households is much greater between 2008 and 2028, when compared to the later time period. However, for the second age group (75-84), this difference is reduced and for the oldest age group, household growth is considerably greater in the second time period when compared to the first for many regions. This result, undoubtedly, is driven primarily by increases in longevity, but also by cohort flow and migration.

For the first time period, growth is quite strong for lone person house
holds in WAB and NT, all growing by about between 3 and 4 times during this time period. For couple only households, the fastest growers were in the Northern Territory (3.02), QLDB (1.92) and WAB (1.94). The list of top growth rates for this first time period is dominated predominately by lone person households. For the second time period, the strongest growing group is lone person households in WA, NT, QLDB and VICCC growing by at least 1.5 times in the period 2023-2038.

Again, the NT, WAB and QLDB populations figure strongly in the fastest growing household types in the two older age groups. The number of couple only households is projected to increase by 3.69 times in the NT between 2008-23 and by 1.93 times for the period 2023-2038 for 75-84 year olds.

As noted above, cohort flow has a significant effect on the projected growth rates. An excellent example is the projected number of lone persons aged 85 and over in QLDCC. In the first time period, this population grows by just 1.23 times (the second lowest growth rate). However, in the second time period, the number of lone persons grows by about 2.3 times (ranked within the top quarter of growth rates). This result underscores the importance of considering (1.) cohort flow, (2.) differential migration, and (3.) the importance of adopting regional level household classification type propensities and living arrangement transition probabilities.

Dwelling Demand Growth

With the underlying population and household projection complete, the demand for dwellings and tenure are projected based upon the preferences of the current cohort of aged persons (from the Census). A very detailed table which cross-references a household reference persons' household type X age X dwelling type X tenure is used to estimate the underlying region specific propensities. This same cross sectional table is applied forward to the projections from 2008 to 2038.

Tables 4 – 6 display the projected growth in dwelling demand between 2008-2023 and 2023-38 for age groups 65-74, 75-84 and 85 years and over respectively. As with the household growth data, QLDB, NT and WAB feature strongly among the fastest growers in the first period. For example, in WAB, the demand for flats among 65-74 year olds is projected to rise by 2.73 times in 2008-23 and a further 1.5 times in the second time period.

In interpreting the data from HaPPE, in addition to interpreting the growth rates and timing effects, it is also important to consider the quantum or raw numerical effects. A good example is comparing QLDB with NT. The demand for separate houses in NT headed by 75-84 year olds is projected to grow by 3.23 times in the first time period, being the fastest growing dwelling type across all regions. QLDB grows only by about 1.95 times over the same period. However, in terms of the real demand for additional dwellings built on the ground, the effect is obviously much stronger for QLDB. In 2008, there was just 856 households headed by persons 75-84 living in separate house dwellings compared to 45,846 such households in QLDB. This notion is again supported looking at some of the lowest growth areas. For example, demand for separate houses in NSWCC is projected to grow by 1.42 times over the first time period, however this is based on a baseline figure of 77,177 households headed by a person 75-84.

Tenure Demand Growth

Tables 7 – 9 display the projected growth in dwellings by tenure between 2008-2023 and 2023-38 for age groups 65-74, 75-84 and 85 years and over respectively.

Of interesting note, are the regions with high demand for public housing. For 65-74 year olds, the top 5 areas of demand between 2008-2023 are NT (2.8), WAB (2.68), QLDB (2.32), WACC (2.32) and SEQ (2.2). For the oldest age group, the top five comparative regions are NT (2.55), WAB (2.34), QLDB (1.79), WACC (1.67). For this older age group, demand is much higher for public housing (as detailed previously) in the second period, but again it is important to note that some of these populations are relatively small in size.

Tenure Demand Growth, Selected Household Types

Finally, using HaPPE, it is possible to extract data on household demand by household type by composition (dwelling type or tenure type). Appendix Tables 1-3 display projected dwelling demand for couples without children and Appendix Tables 4-6 provide the estimates for persons living alone. These tables simply provide an example of the detail that can be extracted using HaPPE to enable users to look at the important interaction effects that occur between living arrangements, dwelling structure and tenure. In interpreting the tenure interaction tables, it is also important to keep in mind that the underlying propensities are kept constant. This is why the within household tenure specific growth rates are the same between tenure types.

Concluding Comment

From the council's perspective, projections of older persons households provide an important insight into the levels and composition of housing demand, and the importance of these dynamics within the context of population ageing. However, the data provided in this report and included in far greater detail in HaPPE (Household and Population Projection Evaluation) Resource, are important for a range of policy considerations beyond housing. For example, understanding indicative futures of elderly living arrangements is important as living arrangements are key indicators of need and wellbeing in retirement (Rowland, 1982; Rowland, 1986). Where there are shortfalls of familial support, particularly within the household, aged persons require additional, mostly economic resources to fulfil this unmet need (Rowland, 2003). The substitution of economic for familial support often occurs at the public level, placing greater strain on government resources.

References

Börsch-Supan, A. 1989. Household dissolution and the choice of alternative living arrangements among elderly Americans. Pp. 119–150 in D. Wise (ed.),The Economics of Aging. Chicago: University of Chicago Press.

Bishop, B. 1999.The National Strategy for an Ageing Australia. Healthy Ageing Discussion Paper. Canberra: Commonwealth of Australia. Costa, D. 1997. Displacing the family: Union army pensions and elderly living arrangements.Journal of Political Economy 105(6): 1269–1292.

Englehardt, G., J. Gruber and C. Perry. 2002. Social security and elderly living arrangements.NBER Working Paper No. 8911, April. Massachusetts: National Bureau of Economic Research.

Hoerger, T., G. Picone and F. Sloan. 1996. Public subsidies, private provision of care and living arrangements of the elderly. Review of Economics and Statistics 78(3): 428–440.

McDonald P. and Temple J. 2004. Evaluation of medium-term projections of housing needs in Australia. Melbourne: Australian Housing and Urban Research Institute.

McDonald P., Kippen R. and Temple J. 2006. Net Transition Probabilities: An Approach to Sub-National Level Projections of Households and Housing Demand Based on Census Data. Population, Space and Place Forthcoming.

McGarry, K. and R. Schoeni. 2000. Social security, economic growth, and the rise in elderly widows independence in the twentieth century.Demography 37(2): 221–236.

Rowland D. 1997. The Demography of Ageing and Families in Australia. Australian Journal on Ageing 16: 99-104.

Rowland D. 1986. Family Structure, In Ageing and Families: A Social Networks Perspective, H. Kendig (ed); Sydney: Allen and Unwin, 17-37.

Rowland D. 1982. Living Arrangements and the Later Family Life Cycle in Australia. Australian Journal on Ageing 1: 3-6.

Rowland D. 2003. An Ageing Population: Emergence of a New Stage of Life? In The Transformation of Australia's Population: 1970-2030, S. Khoo and P. McDonald (eds); Sydney: UNSW Press, 238-65.

Rubin, R. 2002. The economic costs of informal elder caregiving.Business Perspectives 14(1): 22–27.