The outcomes of the projections are described by focusing upon Scenario 3, the scenario that is equivalent to the ABS population projections. In each result section, the outcomes from Scenario 3 will be described and then variations from Scenario 3 will be discussed. The tables shown in the text will relate to outcomes from Scenario 3 while the tables showing outcomes for other scenarios are shown in the Appendix Tables. All tables show results for the capital city and the balance of the state. However, because of its particular interest, all tables also show results for South East Queensland (defined in the note to Table 4).
Demographic inputs and outcomes
Table 5 shows the variations across states and territories in the total fertility rate and expectations of life for men and women. These are both assumed inputs to the projections.
Fertility rates range from 2.24 children per woman in the Balance of Tasmania down to a low of 1.68 in the ACT. Rates are somewhat lower in the capital cities than in the balances of state. It is important to note that variations in the fertility rate have only a marginal impact on housing projections over a 15-year period as the additional child does not generally lead to an additional dwelling. However, the birth of the first child may have a bearing upon the type of dwelling that the parents occupy.
Expectations of life for men range from a high of 79.9 years in the ACT to a low of 72.8 years in the Northern Territory. Expectations of life in the capital cities are generally a little higher that those in the balances of state. Similar differences apply to expectations of life for women. Mortality affects housing demand primarily at the older ages. Besides the extension of life in general, the most important effect in relation to housing is the rise in joint survival for couples at older ages.
In Scenario 3, population growth is highest in Balance of Queensland, Brisbane and Perth all of which experience increases in population of 32-35 per cent from 2006 to 2021. In Scenario 2, with a higher level of migration to Australia, the growth in these three regions is 36-40 per cent. With increased shares of international migration going to Queensland and Western Australia (Scenario 4), the impact on the population growth of these three high growth regions is small compared to Scenario 3. Capital city growth rates are lowest for Sydney, Adelaide and Hobart with 14-18 per cent increases from 2006 to 2021 (Scenario 3), about half the growth for Brisbane and Perth. Melbourne’s growth is intermediate at 25 per cent. South East Queensland (36%) grows a little more rapidly than Brisbane and more rapidly than any other region shown in the table.
In general, population growth rates are lower with Scenario 1 because international migration to Australia is lower in this scenario than in any of the other scenarios. However, in relation to Queensland, Scenario 1 provides the highest population growth rates reflecting high levels of interstate migration to Queensland in the 2001-06 period.
Based on the proportion of the population aged 65 and over, the oldest region in Australia in 2006 was the Balance of New South Wales with 16.1 per cent aged 65 and over. The movement of young people from the country to the city affects the age distributions of all state balances but, in NSW, there is probably an additional effect due to retirement migration out of Sydney to coastal areas. Hobart is the oldest of the cities with 14.2 per cent aged 65 and over. Except for Tasmania, the capital cities are noticeably younger than their balances of state. The Northern Territory and the ACT are much younger than all of the other geographical regions. South East Queensland’s population is noticeably older than that of Brisbane in both 2006 and 2021.
The extent of further ageing of populations by 2021 is related to the level of ageing in 2006; the older places in 2006 get older faster. This means that the balances of state are much older in 2021 than in 2006 with around 22-23 per cent of their populations aged 65 and over. With the exceptions of Hobart and Adelaide, the capital cities remain relatively young in 2021 with the proportion aged 65 and over being 14-16 per cent.
| Region | TFR 2006 |
% aged 65+ 2006 |
% aged 65+ 2021 |
Ratio of Population 2021-2006 |
e0m | e0f |
|---|---|---|---|---|---|---|
| 1. NSW CC | 1.74 | 11.98 | 14.79 | 1.18 | 79.31 | 83.66 |
| 2. NSW B | 1.97 | 16.08 | 22.75 | 1.14 | 78.39 | 83.12 |
| 3. VIC CC | 1.69 | 12.55 | 15.35 | 1.25 | 79.61 | 83.75 |
| 4. VIC B | 1.99 | 15.59 | 22.32 | 1.12 | 78.55 | 83.23 |
| 5. QLD CC | 1.75 | 10.87 | 14.05 | 1.32 | 78.97 | 83.40 |
| 6. QLD B | 1.91 | 13.06 | 18.79 | 1.35 | 78.83 | 83.74 |
| 7. SA CC | 1.70 | 15.00 | 19.03 | 1.15 | 79.02 | 83.54 |
| 8. SA B | 2.16 | 15.32 | 22.67 | 1.14 | 78.97 | 83.56 |
| 9. WA CC | 1.85 | 11.74 | 16.06 | 1.35 | 79.36 | 84.06 |
| 10. WA B | 2.28 | 11.84 | 20.29 | 1.26 | 79.11 | 83.26 |
| 11. TAS CC | 2.01 | 14.24 | 19.47 | 1.14 | 77.58 | 82.80 |
| 12. TAS B | 2.24 | 14.83 | 22.74 | 1.07 | 78.28 | 82.86 |
| 13. NT | 2.19 | 4.63 | 10.11 | 1.25 | 72.82 | 78.08 |
| 14. ACT | 1.68 | 9.53 | 14.99 | 1.21 | 79.90 | 84.02 |
| 15. SEQ | 1.76 | 12.14 | 16.00 | 1.36 | 79.31 | 83.75 |
Outcomes by household type
Table 6 shows the ratio of the number households in 2021 compared to the number in 2006, for each region and each household type. The patterns of changes in household type broadly reflect both population growth and the extent of ageing of the population in a region.
Couple families with children grow most in Brisbane and the Balance of Queensland (both 28%) closely followed by Perth (26%). Then there is a drop to the next band of cities, Sydney, Melbourne and Canberra, where the number of couple families with children grows by 15-16 per cent. In Adelaide and Hobart, there is little growth in this household type and in the balances of state (with the exception of Queensland), the number of families with children tends to fall over the projection period. There are no regions in which the number of couples with children grows faster than the growth of all households. Households consisting of families with children grow more rapidly in South East Queensland than any region shown in the table (31-33%).
The growth of one parent families is broadly similar to the growth of two-parent families with children in most regions, sometimes a little lower, sometimes a little higher.
Reflecting the national trends towards living as a couple described above, households consisting of couple families without children grow much more rapidly than those with children in all regions. In all regions (except Balance of Victoria), the growth of couple families without children is greater than the growth of all households. In the balance of Queensland and in its overlapping South East Queensland region, households consisting of couples without children grow in number by about 50 per cent in the 15-year period of the projection.
Reflecting the ageing of the population, the number of lone person households grows faster than any other household type in all the regions shown in the table. Growth of this household type is highest in the balance of Western Australia (70%), in the Northern Territory (65%) and in the balance of Queensland (64%). There is quite a difference between Sydney (31%) and Melbourne (51%) in the growth of this household type.
Consistent with the national trend away from group household living, the number of group households grows more slowly than total households in all regions except Melbourne. Group households grow more slowly than any other household type in South East Queensland, the only region for which this was the case.
| Ratio of Households in 2021 to Households in 2006 | |||||||
|---|---|---|---|---|---|---|---|
| 2 parent families |
1 parent families |
Coupled without children |
Lone Person |
Group Households |
Total Households |
Persons in NPDS |
|
| 1. NSW CC | 1.15 | 1.13 | 1.26 | 1.31 | 1.20 | 1.22 | 1.29 |
| 2. NSW B | 1.03 | 1.04 | 1.28 | 1.49 | 1.22 | 1.23 | 1.38 |
| 3. VIC CC | 1.16 | 1.28 | 1.29 | 1.51 | 1.32 | 1.30 | 1.43 |
| 4. VIC B | 0.99 | 1.06 | 1.25 | 1.46 | 1.16 | 1.21 | 1.34 |
| 5. QLD CC | 1.28 | 1.28 | 1.40 | 1.48 | 1.28 | 1.36 | 1.38 |
| 6. QLD B | 1.28 | 1.24 | 1.52 | 1.64 | 1.29 | 1.44 | 1.43 |
| 7. SA CC | 1.07 | 1.09 | 1.21 | 1.32 | 1.14 | 1.19 | 1.32 |
| 8. SA B | 0.98 | 1.07 | 1.25 | 1.49 | 1.20 | 1.22 | 1.50 |
| 9. WA CC | 1.26 | 1.32 | 1.44 | 1.57 | 1.30 | 1.40 | 1.53 |
| 10. WA B | 1.11 | 1.14 | 1.47 | 1.70 | 1.25 | 1.38 | 1.39 |
| 11. TAS CC | 1.04 | 1.07 | 1.27 | 1.38 | 1.07 | 1.20 | 1.06 |
| 12. TAS B | 0.92 | 1.01 | 1.21 | 1.40 | 1.10 | 1.16 | 1.41 |
| 13. NT | 1.14 | 1.19 | 1.36 | 1.65 | 1.31 | 1.33 | 1.27 |
| 14. ACT | 1.16 | 1.16 | 1.31 | 1.53 | 1.19 | 1.29 | 1.18 |
| 15. SEQ | 1.33 | 1.31 | 1.47 | 1.55 | 1.30 | 1.42 | 1.44 |
The varying levels of migration shown in Scenarios 1, 2 and 4 change the numeric growth of each household type but do not change the distribution of household types within each region.
On the other hand, changing the assumption relating to the projection of household transition probabilities does alter the distribution of household types with the extent of the alteration varying by region. This can be seen by comparing the results of Scenario 3 with those of Scenario 7. The largest changes in the growth rates of different household types due to changes in the household transition assumptions apply in Queensland and Western Australia as shown in Table 7. The table shows that the growth in total households is hardly affected at all by the change in household transition assumptions but the growth of each household type is affected relatively significantly. The direction of change is that couple families grow more rapidly under Scenario 7 than under Scenario 3 while lone person and group households grow more slowly. This occurs because the trend towards coupling described above is projected to continue under Scenario 7 while Scenario 3 holds all transitions constant at the 2001-06 levels. It seems that this difference is more accentuated in Queensland and Western Australia.
| Region | Scenario |
Ratio of numbers of households in 2021 to numbers in 2006 |
|||||
|---|---|---|---|---|---|---|---|
| 2 parent families |
1 parent families |
Couples without children |
Lone persons |
Group ouseholds |
Total | ||
| Brisbane | 3 | 1.28 | 1.28 | 1.40 | 1.48 | 1.28 | 1.36 |
| Brisbane | 7 | 1.38 | 1.18 | 1.45 | 1.30 | 1.18 | 1.35 |
| Bal. of Qld | 3 | 1.28 | 1.24 | 1.52 | 1.64 | 1.29 | 1.44 |
| Bal. of Qld. | 7 | 1.38 | 1.18 | 1.58 | 1.44 | 1.08 | 1.42 |
| SEQ | 3 | 1.33 | 1.31 | 1.47 | 1.55 | 1.30 | 1.42 |
| SEQ | 7 | 1.35 | 1.32 | 1.47 | 1.53 | 1.21 | 1.42 |
| Perth | 3 | 1.26 | 1.32 | 1.44 | 1.57 | 1.30 | 1.40 |
| Perth | 7 | 1.30 | 1.26 | 1.48 | 1.51 | 1.26 | 1.40 |
| Bal. of WA | 3 | 1.11 | 1.14 | 1.47 | 1.70 | 1.25 | 1.38 |
| Bal. of WA | 7 | 1.13 | 1.15 | 1.47 | 1.65 | 1.15 | 1.37 |
Demand for additional dwellings
Table 8 shows the numbers of additional dwellings for the whole of the 15-year period that would be required to meet projected demand under Scenarios 1, 2, 3 and 4. Under Scenario 3 (consistent with the 2008 ABS projections), the total new dwellings required for Australia as a whole over the 15-year period from 2006 to 2021 would be 2,309,000 or 154,000 per annum. The total number would be 2,583,000 (172,000 per annum) if net overseas migration was 230,000 per annum (Scenario 2). This means that an additional 50,000 immigrants (net) per annum leads to an additional 18,000 dwellings per annum. Under Scenario 3, the largest numbers of additional dwellings would be required in South East Queensland (31,000 per annum) and in Melbourne (29,000 per annum). An additional 23,000 dwellings per annum would be required in Sydney and 16,000 per annum in Perth.
Scenario 1 assumes that age and sex specific migration rates will remain the same as they were on average over the years, 2001-06. The level of net overseas migration for this period was lower than the levels assumed in the other scenarios however rates of migration to South East Queensland were higher in that period than is assumed in the other scenarios. The result is that Scenario 1 implies a smaller number of additional dwellings for Australia as a whole when compared with other scenarios but a higher number in South East Queensland. The main offset is a much smaller number of dwellings required in New South Wales under Scenario 1.
Under Scenario 4, the total number of dwellings required for Australia is the same as the requirement under Scenario 3, but the distribution across the states changes because Scenario 4 assumes a higher level of movement to Queensland and Western Australia. However, the assumptions made under Scenario 4 do not have as much impact on the dwelling requirement in South East Queensland as a 50,000 annual increase in net overseas migration to Australia (Scenario 2).
Table 8 clearly shows the sensitivity of the projections to assumed levels of overseas and inter-regional migration. The authors consider that new research needs to be conducted on methods to project future levels of migration.
| Region | Additional dwellings required, 2006-2021 (thousands) | |||
|---|---|---|---|---|
| Scenario 1 | Scenario 2 | Scenario 3 | Scenario 4 | |
| Sydney | 228 | 428 | 348 | 334 |
| Balance of NSW | 168 | 241 | 237 | 237 |
| Melbourne | 431 | 498 | 430 | 417 |
| Balance of Victoria | 103 | 119 | 115 | 114 |
| Brisbane | 294 | 280 | 250 | 262 |
| Balance of Queensland | 463 | 409 | 388 | 397 |
| Adelaide | 68 | 106 | 88 | 83 |
| Balance of SA | 36 | 40 | 38 | 37 |
| Perth | 204 | 275 | 239 | 252 |
| Balance of WA | 61 | 84 | 80 | 81 |
| Hobart | 15 | 18 | 17 | 17 |
| Balance of Tasmania | 22 | 19 | 18 | 18 |
| Northern Territory | 18 | 26 | 24 | 24 |
| ACT | 26 | 39 | 38 | 37 |
| AUSTRALIA | 2137 | 2583 | 2309 | 2310 |
| South East Queensland | 563 | 520 | 458 | 484 |
Pair-wise comparisons of Scenarios 1 and 5, 2 and 6, 3 and 7 and 4 and 8 indicate the impact on demand for additional dwellings of changes in the assumptions about household transition probabilities. All the comparisons produce a similar result, that is, that about 3,000 fewer dwellings per annum would be required under Scenarios 5-8 than under Scenarios 1-4. This means that when the trends in household formation and dissolution from 1996-2001 to 2001-06 are projected forward slightly more large households are formed and the dwelling requirement is a little lower. As shown in Table 7, Scenarios 5-8 produce fewer lone person households than is the case with Scenarios 1-4.
Overall, the projections show that future housing demand is quite sensitive to changes in assumptions about migration both overseas and internal but not very sensitive to changes in assumptions about rates of household formation and dissolution.
Demand by dwelling type
Table 9 shows the ratio of dwellings required in 2021 to the number of dwellings in 2006 according to dwelling type. The main conclusion here seems to be additional demand for flats compared to other dwelling types outside of the cities. This is particularly the case in South Australia, Western Australia and Tasmania. For example, in the balance of Western Australia, the demand for additional separate houses is projected to be 36 per cent but the demand for additional flats is 52 per cent. The equivalent comparison in the balance of South Australia is 21 per cent compared to 37 per cent. There is also a projected higher demand for flats in the balance of Victoria compared to other dwelling types. In Western Australia, the projected demand for both semi-detached dwellings and flats is much higher than that for separate dwellings in both Perth and the balance of Western Australia.
Variations in the assumptions relating to household formation and dissolution have an effect on demand for dwelling type in some regions more than others. The largest variation is in the future demand for flats. The projected percentage increase in demand for flats based on Scenarios 3 and 7 were: for Brisbane, 37 per cent and 29 per cent; for Perth, 48 per cent and 45 per cent; for Balance of Western Australia, 52 per cent and 49 per cent; and for Balance of South Australia, 37 per cent and 29 per cent. In other regions, the differences were not as large. These differences are driven by projected differences in household types. It seems that some regions are more subject to variation in future household type than others.
The projected high demand for flats outside of the cities no doubt reflects the greater ageing of the populations outside of the cities but it also reflects the low base, that is, the relative dearth of flat accommodation outside of the cities in 2006.
| Region |
Ratio of Dwellings in 2021 to Dwellings in 2006 |
|||
|---|---|---|---|---|
| Total | Separate Houses |
Semi- Detached |
Flats | |
| 1. NSW CC | 1.22 | 1.22 | 1.22 | 1.21 |
| 2. NSW B | 1.23 | 1.22 | 1.28 | 1.28 |
| 3. VIC CC | 1.30 | 1.29 | 1.32 | 1.35 |
| 4. VIC B | 1.21 | 1.20 | 1.26 | 1.29 |
| 5. QLD CC | 1.36 | 1.36 | 1.37 | 1.37 |
| 6. QLD B | 1.44 | 1.43 | 1.46 | 1.44 |
| 7. SA CC | 1.19 | 1.18 | 1.22 | 1.22 |
| 8. SA B | 1.22 | 1.21 | 1.28 | 1.37 |
| 9. WA CC | 1.40 | 1.39 | 1.45 | 1.48 |
| 10. WA B | 1.38 | 1.36 | 1.47 | 1.52 |
| 11. TAS CC | 1.20 | 1.20 | 1.21 | 1.21 |
| 12. TAS B | 1.16 | 1.15 | 1.25 | 1.24 |
| 13. NT | 1.33 | 1.31 | 1.37 | 1.35 |
| 14. ACT | 1.29 | 1.29 | 1.29 | 1.29 |
| 15. SEQ | 1.42 | 1.41 | 1.43 | 1.42 |
Demand by tenure type
Table 10 shows the projected demand for dwellings in relation to their tenure under Scenario 3. The main conclusions from the table are that, in most regions, the demand for public rental dwellings increases more than the demand for all dwellings and the demand for private rental dwellings increases less than the demand for all dwellings.
The regions in which the demand for public housing is projected to rise most compared to the demand for all dwellings are Melbourne (37% for public; 30% for all), Adelaide (24% for public; 19% for all), Perth (53% for public, 40% for all), and Balance of South Australia (29% for public, 22% for all). This means that, in these localities, household types shift over the projection period towards those that are more likely to be public renters. Note that changing income levels within household types are not taken into account in making these estimates. If, for example, lone aged persons were to be wealthier in 2021 than they were in 2006, then the demand for public housing might not be as high.
Regions in which the additional demand for private rental dwellings is well below the additional demand for all dwellings by 2021 include all of the balances of state as well as South East Queensland, Hobart, Northern Territory, and the ACT.
In all regions, the projected demand for owner/purchaser dwellings is higher than or equal to the projected demand of all dwellings. Affordability is not taken into account in making these projections.
| Region |
Ratio of Dwellings in 2021 to Dwellings in 2006 |
|||
|---|---|---|---|---|
| Total | Owner/ Purchasers |
Public Renters |
Private Renters |
|
| 1. NSW CC | 1.22 | 1.23 | 1.21 | 1.18 |
| 2. NSW B | 1.23 | 1.25 | 1.24 | 1.17 |
| 3. VIC CC | 1.30 | 1.30 | 1.37 | 1.29 |
| 4. VIC B | 1.21 | 1.22 | 1.25 | 1.15 |
| 5. QLD CC | 1.36 | 1.38 | 1.39 | 1.31 |
| 6. QLD B | 1.44 | 1.47 | 1.46 | 1.35 |
| 7. SA CC | 1.19 | 1.19 | 1.24 | 1.15 |
| 8. SA B | 1.22 | 1.23 | 1.29 | 1.16 |
| 9. WA CC | 1.40 | 1.41 | 1.53 | 1.35 |
| 10. WA B | 1.38 | 1.42 | 1.42 | 1.25 |
| 11. TAS CC | 1.20 | 1.22 | 1.20 | 1.11 |
| 12. TAS B | 1.16 | 1.17 | 1.18 | 1.09 |
| 13. NT | 1.33 | 1.36 | 1.50 | 1.26 |
| 14. ACT | 1.29 | 1.32 | 1.33 | 1.20 |
| 15. SEQ | 1.42 | 1.44 | 1.45 | 1.35 |
If the alternative household formation and dissolution assumptions are made, the demand for public and private rental dwellings is lower (comparing Scenario 3 and Scenario 7). The regions for which this effect is largest are shown in Table 11. In these regions, Scenario 7 yields a lower total demand than Scenario 3 but all of the reduction relates to rental dwellings.
| Region | Tenure type |
Ratio of 2021 demand to 2006 demand |
|
|---|---|---|---|
| Scenario 3 | Scenario 7 | ||
| Brisbane | Total | 1.36 | 1.35 |
| Owner/purchaser | 1.38 | 1.39 | |
| Public rental | 1.39 | 1.31 | |
| Private rental | 1.31 | 1.26 | |
| Perth | Total | 1.40 | 1.40 |
| Owner/purchaser | 1.41 | 1.41 | |
| Public rental | 1.53 | 1.49 | |
| Private rental | 1.35 | 1.34 | |
| Balance of SA | Total | 1.22 | 1.20 |
| Owner/purchaser | 1.23 | 1.22 | |
| Public rental | 1.29 | 1.25 | |
| Private rental | 1.16 | 1.11 | |
| Balance of Tas. | Total | 1.16 | 1.14 |
| Owner/purchaser | 1.17 | 1.17 | |
| Public rental | 1.18 | 1.12 | |
| Private rental | 1.09 | 1.03 | |