A. Method
The ABS uses a propensity method to project the population by living arrangement and the number of households and families. The method identifies propensities (or likelihood) for people to belong to different living arrangement types from the Census of Population and Housing. These propensities are then applied to the projected population to give the projected population by living arrangement. The numbers of households and families are then derived.
These household and family projections were produced in four steps:
1. At State/Territory level, the assumed living arrangement propensities by five year age group were applied to the population projections to produce the projected population by living arrangement.
2. At the Capital City and Balance of State/Territory level, the assumed living arrangement propensities by five year age group were applied to the population projections and the resultant projected populations by living arrangement were constrained to their respective State/Territory figures.
3. At SLA level, the assumed living arrangement propensities by five year age group were applied to the population projections and the resultant projected populations by living arrangement were constrained to the respective Capital City and Balance of State/Territory figures.
4. Finally, the projected populations by living arrangement at the SLA level were used to derive the number of families, family households, non-family households and total households.
The population projections used are consistent, at the Capital City and Balance of State/Territory level with Series II in Population Projections, Australia 1999 to 2101 (Cat. No. 3222.0) published on 17 August 2000. Details on the SLA population projections are given in Part D of this document.
B. Living Arrangement Assumptions
Only one set of assumptions was used for these projections. It was assumed that the living arrangement propensities observed at the 1996 Census of Population and Housing would remain unchanged throughout the projection period. This assumption is one of the three alternative assumptions used in Household and Family Projections, Australia 1996 to 2021 (Cat. No. 3236.0) published on 28 October 1999.
A small number of SLAs required special attention. This was where the initial population was very small (less than 300) and a very high rate of growth was projected. In these cases it was necessary to substitute the observed propensities with more meaningful propensities, for example, with the propensities at the Balance of State level or of an area with similar characteristics.
C. Nature of Projections
The nature of the projection method and inherent fluctuations in population dynamics mean that care should be taken when using and interpreting the projection results. The projections are not exact forecasts but simply illustrate future changes which would occur if the stated assumptions were to apply over the projection period. While the population projections take account of land planning and other decisions by governments known at the time the projections were derived, the ABS does not always have access to the policies or decisions of Commonwealth, State and Local Governments and businesses that assist in accurately forecasting small area populations. The projections do not attempt to allow for non-demographic factors (eg. major government policy decisions, economic factors, catastrophes, wars) which may affect future demographic behaviour.
The unpredictability of migration trends, especially in the short-term, can have a significant effect on projection results. Special care should be taken with small SLAs (populations under 1,000 people).
All SLAs in these projections are based on the Australian Standard Geographic Classification boundaries as they existed at the 1996 Census (ASGC 1996). The actual boundaries for a given SLA, or for other geographic regions such as Local Government Areas (LGAs) derived from this SLA, may change over time making the projections no longer comparable with other data.
D. Population Projections
D.1 Method
The cohort-component method was used for these projections. In this method, the base population is projected forward annually by calculating the effect of births, deaths and migration within each age-sex group according to the specified fertility, mortality and migration assumptions.
These projections were produced in two stages. Firstly, the resident population of each capital city and balance of State/Territory was projected by single year of age and sex. Secondly, the population of each SLA within each capital city and balance of State/Territory was projected by single year of age and sex, and constrained to sum to the respective capital city and balance of State/Territory projected population for each year.
D.2 Assumptions for Capital City and Balance of State/Territory Projections
The base population for the capital city and balance of State/Territory projections was the preliminary estimated resident population in each area, at 30 June 1999. The assumptions for fertility, mortality, overseas migration and interstate migration are the same at the capital city and balance of State/Territory level as Series II outlined in Population Projections, Australia 1999 to 2101 (Cat. No. 3222.0) published on 17 August 2000. Series II incorporates assumptions for Australia of 'low' fertility (a total fertility rate of 1.60 births per women from 2008), 'medium' overseas migration (an annual net overseas migration gain of 90,000) and 'medium' interstate migration.
D.3 Assumptions for SLA Projections
The base population for SLAs in NSW, Victoria, South Australia, Western Australia, Tasmania and ACT was the preliminary estimated resident population in each area by single year of age and sex, at 30 June 1999. For SLAs in Queensland and Northern Territory, the revised estimated resident population in each area at 30 June 1999 was used as the base.
The fertility assumptions were based on average age-specific fertility rates observed in each SLA between 1991 and 1997, projected to change in line with the State/Territory assumptions outlined in Population Projections 1997 to 2051 (Cat. No. 3222.0) published on 14 July 1998. The mortality assumptions were based on the State/Territory age-and-sex-specific mortality rates. Assumptions for 1998-2006 were calculated using the rates of mortality decline observed in each State/Territory between 1987-91 and 1992-96. Assumptions beyond 2006 were calculated using the rates of mortality decline observed in Australia during 1967-71 to 1992-96. The mortality assumption was then adjusted to reflect the recent mortality experienced in each SLA.
In constraining the SLA population projections to the projections published in Population Projections, Australia 1999 to 2101, the fertility and mortality assumptions used in the latter projections will in effect filter down to the SLA projections.
The assumed migration levels were based on historical trends of net migration in each SLA, the assumed levels of State/Territory overseas and interstate migration, and any recent State/Territory government dwelling and population projections or local land planning information. The assumed SLA net migration levels were constrained to the capital city and balance of State/Territory assumptions in Series II. The age-sex distributions for the assumed migration levels were based on overseas and inter-SLA migration rates used in the calculation of published ABS SLA age-sex population estimates, which were originally derived from 1996 Census of Population and Housing migration data.
E. Liability
It is important to recognise that the projection results given in this report simply reflect the assumptions made about future fertility, mortality, migration and living arrangement trends. While the assumptions are formulated on the basis of an objective assessment of demographic trends, there can be no certainty that they will be realised.
While ABS takes responsibility for the method employed, the assumptions used are the final responsibility of the client, and the projections are not official ABS population statistics.
No liability will be accepted by the Australian Bureau of Statistics for any damages arising from decisions or actions based upon these household and family projections.
F. Referencing these statistics
These projections should be referred to as projections prepared by the ABS according to assumptions agreed to by the Commonwealth Department of Health and Aged Care. The short reference is - Commonwealth Department of Health and Aged Care projections.
G. Additional Information
For further information on population and household projections refer to the ABS publications Population Projections, Australia 1999 to 2101 (Cat. No. 3222.0) and Household and Family Projections, Australia 1996 to 2021 (Cat. No. 3236.0) available from ABS offices in all capital cities. For more information, or quotes on the preparation of projections with other assumptions, please contact:
Matthew Berger
Demography Section
Australian Bureau of Statistics
Tel. 02 6252 6639
Email: matthew.berger@abs.gov.au
Print Page
Print All