Victoria in Future

Major Regions include the Greater Melbourne GCCSA (Greater Capital City Statistical Area) and the eight non-metropolitan SA4s (Statistical Area Level 4). Grand totals for these eight regions are presented as "Victorian Regions". All data for these regions can be downloaded in one file.

VIF2019 Major Regions 2056 (Excel, 4.1 MB)

Statistical Areas are those defined and used by the Australian Bureau of Statistics (ABS) in the Australian Statistical Geography Standard (ASGS). Data are provided for all regions from Statistical Area Level 2 (SA2) and above. Data for these regions are compiled into two files: one for population by age groups and one for total population, household and dwelling projections.

VIF2019 Population 5y ages ASGS 2036 (Excel, 2.4 MB)

VIF2019 Pop Hholds Dws ASGS 2036 (Excel, 918.7 KB)

Victoria’s 79 municipalities plus Unincorporated Victoria are listed as LGAs. Within each LGA data are presented at the VIFSA level. VIFSA is a customised set of regions designed to fit exactly within LGA boundaries, whereas some ASGS regions cross LGA boundaries. Data for these regions are compiled into two files: one for population by age groups and one for total population, household and dwelling projections.

VIF2019 Population 5y ages VIFSA LGA_2036 (Excel, 1.5 MB)

VIF2019 Pop Hholds Dws VIFSA LGA_2036 (Excel, 1.2 MB)

VIF2019 Population Service ages LGA 2036 (Excel, 229.0 KB)

VIF2019 Profiles (Excel, 1.2 MB)

Most of these area types are defined by the Australian Bureau of Statistics (ABS) in its Australian Statistical Geography Standard (ASGS) publications. The state of Victoria, the Greater Melbourne Capital City Statistical Area and the Statistical Areas from Level 2 (SA2s) to Level 4 (SA4s) are all standard ABS geographic areas. Local Government Areas (LGAs) are defined by the ABS and fit within the ASGS but are considered a Non-ABS Structure.

View an interactive map of ASGS areas.

For some previous editions of Victoria in Future, (e.g. VIF2012) projections were published for Statistical Local Areas (SLAs), part of the now obsolete Australian Standard Geographical Classification (ASGC). SLAs were geographic subsets of LGAs. There were 210 SLAs in Victoria. As these are no longer supported by ABS, they are no longer used for VIF publication.

The closest equivalent to the SLA in the ASGS is the Statistical Area Level 2 (SA2). There are 433 SA2s in Victoria.

It is recognised, however, that users would like to have data available for areas consistent with, but smaller than LGA to allow for local area information and decision making. To meet this need, VIF is published for a series of geographic areas known as Victoria in Future Small Areas (VIFSAs). VIFSAs are based on SA2s, but adjusted to be exact subsets of LGAs. In some cases SA2s are aggregated, some are split, and a small number of VIFSAs are identical to SA2s.

VIFSAs are:

• smaller than LGAs and fit entirely within them
• large enough to avoid the uncertainty attached to some projections at SA2 level
• as far as possible distinct from ABS classification, so as to avoid confusion and misinterpretations.

A list of the VIFSAs with corresponding maps (PDF, 1.2 MB) grouped by region is attached.

In addition, the VIFSA map files are downloadable in Mapinfo TAB (ZIP, 8.9 MB) and Esri SHP (ZIP, 8.1 MB) formats for use with Geographic Information Systems (GIS).

VIF2019 refers to the updated population and household projections Victoria in Future 2019 produced by the Department of Environment, Land, Water and Planning.

VIF2016 refers to the previous population and household projections Victoria in Future 2016 which have been superseded by VIF2019.

Greater Melbourne refers to the Melbourne Greater Capital City Statistical area as defined by the Australian Bureau of Statistics (ABS). Victorian Regions refers to the remaining 8 Statistical Areas Level 4 (SA4) either as individual regions or in total: Ballarat; Bendigo; Geelong; Hume; Latrobe – Gippsland; North West; Shepparton; Warrnambool and South West. Maps of these areas are available on the ABS web site. 

Metropolitan Melbourne in VIF2019 refers to the region covered by the 31 local government areas of Melbourne and also includes the growth areas of Wallan, Beveridge and surrounds, within Mitchell Shire (represented by the Wallan SA2).

Regional Victoria refers to the region covered by the 48 local government areas and unincorporated areas outside of Melbourne (except for that part of Mitchell Shire indicated above). Maps of the component LGA and SA2 areas are available on the ABS web site. 

VIF Small Areas (VIFSAs) are a set of small areas specifically designed by DELWP to provide more local information than the LGAs. These Victoria in Future Small Areas (VIFSA) fit within the LGAs and cover the whole of Victoria.

The year reference (e.g. 2016, 2036, 2056, etc) refers to the population at 30 June in that year.

For definitions of demographic terms, refer to the Glossary.

Victoria in Future (or VIF) is the official state government projection of population and households across Victoria, produced by the Land Use & Population Research team of the Department of Environment, Land, Water & Planning.

VIF2019 was released in 2019 and is the most up to date set of projections. The projections use the Estimated Resident Population (ERP) at 30 June 2018 as the base population (as published in Demographic Statistics (ABS 3101.0 September 2018). This is the most recent official population which is available for all geographic levels.

Victoria in Future projections are an estimate of the future size, distribution and composition of the population. The projections give an idea of what is likely to happen if current trends continue. They are developed using mathematical models and expert knowledge, relying on trend analysis and assumptions about future change.

VIF projections are an important guide for planning and building for our future. State and local governments and agencies use the projections to plan for land use and development, infrastructure, services and programs. Developers and businesses use them to analyse potential markets, for example future demands for dwellings, goods and services and labour supply.

Clause 11.02 of the State Planning Policy Framework explicitly requires planning for urban growth to have consideration to Victorian Government population projections.

VIF2019 is not an exact prediction or forecast of the future. Uncertainty about the future increases over longer projection horizons and with smaller disaggregations, geographic or sectoral. Different policy settings and changes in the economy could result in changes to the expected size, distribution and characteristics of the population.

All VIF2019 products and data can be accessed via the Victoria in Future 2019 pages on the website of the Department of Environment, Land, Water and Planning (DELWP).

The VIF2019 Brochure is also available in hard copy. Hard copies can be requested from Forward Policy and Research.

VIF2019 products

• VIF2019 brochure: Summary of VIF2019 highlights, graphs, maps and data

• One-page Profiles: Summaries of VIF2019 data and graphs for individual regions
• Data Files: Excel spreadsheets of detailed VIF2019 data (refer to table below)

VIF2019 Data Files

For the first time, in 2019, Victoria in Future projections will also be published for ASGS areas as small as Statistical Area Level 2 (SA2). In built-up areas an SA2 is usually roughly equivalent to a suburb or town. In rural areas SA2s can be much larger.

VIF2019 does not provide projections for the following geographical areas:

• Australia
• Other states/territories
• Suburbs and towns
• Postcode areas
• SA1s

The Australian Bureau of Statistics (ABS) produces population projections for Australia and other states/territories. These projections are available at the ABS website. 3222.0 - Population Projections, Australia, 2017 (base) to 2066.

VIF2019 was produced to take account of the results of the 2016 national Census and incorporates ABS population data as at 30 June 2018 (as published in Demographic Statistics (ABS 3101.0 September 2018). The VIF2019 projections will be tested against new published data when it becomes available, and a new Victoria in Future prepared as necessary.

VIF2019 projections are an indication of possible future populations if current demographic, economic and social trends continue. They are not predictions of the future, nor are they future targets.

The projections take into account current government policy measures that explicitly affect the sizes and distributions of future populations. They do not consider policies that may be introduced in the future.

The Australian Bureau of Statistics (ABS) produces population projections for Australia and other states/territories. These projections are available at the ABS website. 3222.0 - Population Projections, Australia, 2017 (base) to 2066.

Both the ABS and VIF use the cohort component model, which begins with a base population for each year of age and sex and advances it year by year, applying assumptions regarding future fertility, mortality and migration.

One key difference of methodology relates to migration within Australia. VIF2019 models the migration flows between Victoria and the other individual states and territories, and between the regions of Victoria, while the ABS models the flows between each of the Greater Capital Cities and Rest of State areas and the rest of Australia as a whole.

Further differences result from VIF2019 and the ABS choosing different assumptions for the key drivers of population change: fertility and mortality rates and level of overseas migration. VIF2019 assumptions are detailed in another question.

VIF2019 projections are more detailed geographically than the ABS. The ABS projections are produced for the Capital City and Rest of State Statistical Areas (15 areas) and aggregated into States and Territories and Australia. VIF2019 projections are prepared first for Victoria, the Melbourne Capital City Statistical Area and the 8 Statistical Areas Level 4 (SA4s) in regional Victoria. Projections for the 462 Statistical Areas Level 2 (SA2s) in Victoria are calculated, ensuring they add up to the larger relevant totals.

VIF2019 also aggregates the projections into Local Government Areas (79 LGAs in Victoria plus Unincorporated Victoria) and Victoria in Future Small Areas (191 VIFSAs) specifically designed by DELWP to provide more local information within each LGA.

For the smaller geographical areas, DELWP gathers, analyses and uses information about local land supply and patterns of residential development as a significant input to the VIF2019 projections to ensure that the VIF projections are realistic and take into account local influences on population change.

VIF2019 is independent of other projections of populations, prepared for example by local governments or private businesses

VIF2019 updates VIF2016 and broadly uses the same methodology but makes different assumptions, based on up to date data and any new trends detected since the publication of VIF2016. VIF2019 uses the results of the 2016 Census, whereas VIF2016 was based on the results of the 2011 Census. The base population for VIF2019 is the Estimated Resident Population as at 30 June 2018.

After the publication of the 2016 Census, the population of Victoria was found to be approximately 100,000 greater than previously reported by the ABS. This ‘rebasing’ has been included in the VIF2019 population. VIF2019’s base population for 2016 is 6.17 million, where VIF2016 estimated the 2016 population to be 6.07 million.  

VIF2019 projections extend to 2056 for the state and major regions, and to 2036 for Local Government Areas and smaller areas. VIF2016 extended to 2031 and 2056. Compared with VIF2016 at 2051, VIF2019 makes projections of larger population and greater growth. VIF2016 projected the state’s population to reach 10.1 million by 2051. VIF2019 projects the population to reach 10.6 million by 2051 and 11.2 million by 2056.

VIF2019 projections incorporate Census data in making assumptions, but the official population estimate used in the projections is the Estimated Resident Population (ERP). This is the official estimate used by the ABS to record the population and is used for purposes such as funding distribution and electoral boundary changes. It is different from the Census count as it uses a different method of calculation. The ERP is based on the population count from the latest Census, with the addition of quarterly components of population growth (births, deaths, population movements). It incorporates the estimated Census net undercount (to account for people who were missed from or counted more than once in the Census) and is adjusted to include usual residents who were temporarily overseas at the time of the Census but had returned by the end of the reference period, and to exclude overseas visitors who were temporarily in Australia on Census night.

The ABS produces ERPs quarterly for Australia, its states, territories and annually for sub-state/territory regions and for local government areas (ABS cat. no. 3301.0, 3218.0 and 3235.0).

Victoria in Future projects future ERP, based the ERP for Victoria at 30 June 2018 as published by the ABS in March 2019. Population projections are estimates of future populations and, unlike the ERP, are produced in advance of the reference period. Projections are developed by applying mathematical models and knowledge of likely population trends to the ERP and other existing data.

It is possible that during the life of the VIF2019 projections, the ABS will revise or update the ERP for one or more years in the past, or will publish a new estimate for one of the years for which VIF has made a projection.

DELWP recommends that users wanting population data for a period covered by published ERP data should use that source. Projected data should be used only if the ERP data is not available for the required reference period, or if the data required is for a population at a future time period.

For more information about Estimated Resident Population, refer to the ABS Australian Demographic Statistics, catalogue number 3101.0, available at the ABS website.

The projections are developed by analysing/quantifying past changes to estimate future population numbers and distribution. They are not simply a continuation of recent changes (either in number or in rate), rather they involve the application of specific assumptions about the causes/drivers of the change.

At the national, State and broad region level, fertility, mortality and migration behaviours are the drivers of change. At smaller levels of geography, dwelling availability (either through new land or redevelopment potential) and peoples housing preferences need also been taken into account.

The DELWP modelling process involves both approaches.

State and broad regions populations are projected by assuming that trends in fertility, mortality and migration patterns continue to apply to the current population. This provides the overall population by age and sex for each year of the projections.

For local areas within these regions, likely dwelling construction and housing capacities are developed from a range of DELWP, ABS and other sources including but not limited to building approvals, the Urban Development Program, DELWP’s Housing Development Data, land use zoning and growth strategies. Combined with likely person per dwelling determinants (household size and occupancy rates), this provides a measure of total population for each small area.

The large area totals (top-down) and small area totals (bottom-up) models are balanced through an implied migration model to ensure all local totals sum to the state and regional totals and that births and deaths in the existing population of each small area are plausible.

The VIF2019 population and household projections draw on three standard forms of modelling:

• The Cohort Component model projects the population by age and sex for each period using the basic demographic equation:

Population at end of period = Base population + Births – Deaths + Net migration

• The Housing Unit model projects the supply of dwellings available for occupation in a given area for each period using the basic equation:

Occupied dwellings at end of period = (Base stock of dwellings + new construction – demolitions) * (occupancy rate).

Using an average number of persons expected in each household, the housing unit model can also be used to determine the total projected population for each area.

• The Household Formation model allocates the population into households, based on living arrangement propensities (i.e. the distribution of people of each age and sex across household types).

The whole projection process can be summarised as follows:

Steps 1 to 3 are classified as ‘Top Down’ steps

Step 1 - The Cohort Component model is used to determine the population by age and sex for Victoria. Assumptions are made for the levels of overseas and interstate migration (as flows in and out) and for the fertility and mortality rates which are applied to the population by age to determine the number of births and deaths in the population each year.

Step 2 – The Cohort Component model is used to determine the population by age and sex for Greater Melbourne and the eight SA4 regions in the rest of the state. Assumptions are made for the levels of overseas, interstate and intrastate (within state) migration (as flows in and out) and for the fertility and mortality rates which are applied to each region’s population by age to determine the number of births and deaths in the population each year.

Step 3 - The Household Formation model determines an initial control total of households in each region. The Census provides Living Arrangements data which in turn provide household propensities (the probability that an individual of a given age will be living in any of a range of household types, such as living alone, as a parent in a family, living in a group household etc.).

Steps 4 onwards are ‘Bottom Up’ steps

Step 4 – Expected dwelling construction from a number of project streams are brought together to determine total expected dwelling construction for each local area (SA2). The four project streams are: broadhectare development; major redevelopment sites; major urban renewal precincts; dispersed infill development.

Step 5 - The Housing Unit model uses the projected number of additional dwellings by small geographical area and applies to each an expected average household size depending on the type of dwelling constructed (separate houses, semi-detached and townhouses, apartments). The resulting additional population is added to the previous year’s population to determine the total population in each small area.

Step 6 - The Cohort Component model is used in reverse to infer the age/sex distribution of each small area population. Total population change is known (from Step 5). Assumptions about fertility and mortality rates, based on Census and ABS data, are used to determine the expected natural increase, and the resulting implied net migration  The age patterns of migration and therefore the age-based impacts of growth for each area is also derived from Census data (e.g. most migrants to inner city areas are young adults; suburban growth areas attract many families with children).

Step 7 - The Household Formation model determines the number of households of each type in each small area. As per the state and region level projections, Census data is used to determine the number and type of households in each area based on the ages of the population and the characteristics of the area.

Steps 8-10 – The models from steps 5 to 7 are iterated at least once more to ensure household and dwelling totals balance at each stage. Initial estimates may change during iterations.

Step 11 – Final small area totals are aggregated to region and state level and checked against control totals to ensure coherence of projections at all levels.

Summary of assumptions required for DELWP’s models

The Cohort Component model requires the following inputs:

• base year estimated resident population by age and sex
• age and sex specific mortality rates
• age-specific fertility rates
• sex ratio for births
• overseas, interstate, regional migration total flows (in and out)
• overseas, interstate and/or regional migration age and sex distribution

The inputs for the Housing Unit model are:

• base year’s stock of structural private dwellings
• new dwelling construction
• stock loss rates (through demolition or conversion to other uses)
• occupancy/vacancy rates (dwellings without usual resident households)

The Household Formation model, developed by McDonald and Kippen in 1998, has been adapted by DELWP to project household formation in Victoria. It applies a propensity (likelihood) to persons of each age and sex to live in each household type (including non-private households), and thus determines the number of households and average household size.

The living arrangement tables by age and sex from the 2016 Census provide the propensities used as input. The household types include:

• Couple with children
• Couple without children
• Lone parent family
• Other families of related persons
• Lone person household
• Group household
• Other household

Persons in non-private households are classified and projected separately.

VIF2019 provides a single set of projections to allow analysis and decision making on a consistent base for all users.

DELWP performs sensitivity analyses on the main assumptions, before choosing the most reasonable single set and using it to produce one set of projections. This information is strengthened by extensive research on local and regional trends, and by wide consultations with people with specialist and local knowledge.

Using VIF2019 as reference, it is possible to analyse the effects of different policy scenarios on population levels and distribution between and within the regions of Victoria.

Base Population

The VIF2019 projections use the Estimated Resident Population for 30 June 2018 as the base population, as published in the ABS Australian Demographic Statistics, cat 3101.0, September 2018, published March 2019.

Components of population change

Populations grow (or decline) in number from a combination of natural increase (number of births minus number of deaths) and net migration (number of people moving in from another area minus number of people moving out to another area). This constitutes the basic demographic equation:

Change in population = births – deaths + net overseas migration + net internal migration

For each of the four components of population change, one main assumption has been developed at the state level.

State-level assumptions applied in VIF2019

For sub-state projections, local fertility rates and migration patterns are specified. The assumption for mortality is the same across all areas.

Fertility:

In VIF2019, the number of births each year is obtained by applying age specific fertility rates (ASFR) to the number of women in the relevant ages. These ASFRs represent the likelihood of a woman of a given age having a child within the year. Influences on fertility are complex and interact to varying degrees over time. It is difficult to measure the impact of any one influence. Nonetheless, recent trends have been fairly consistent, with declining rates for the youngest ages and increasing rates for the 30 to 39-year-olds. VIF2019 assumes a continuation of these trends, within reasonable limits. Overall, the resulting Total Fertility Rate (TFR) remains fairly constant at around 1.6 children per woman for the projection period.

The combination of strong population growth and substantial numbers of women in the childbearing age groups projected for the future will result in relatively high (and increasing) numbers of births in Victoria. After declining sharply in the early 1970’s, the number of births averaged around 61,000 per annum between 1975 and 2000. The number of births each year has been over 78,000 since 2016 and is expected to reach 120,000 by 2056.

There are considerable variations in the fertility rates in different Victorian regions. Generally the more urban the region, the lower its fertility rate. Melbourne has the lowest rate, followed by the three largest regional centres: Geelong, Ballarat and Bendigo. VIF2019 assumes that these spatial differences in TFR will continue into the future.

Mortality:

Life expectancy (the average age to which a person is likely to live) has been increasing steadily in Victoria over the past century. With advances in life-prolonging medicines and technologies, reductions in premature life-ending events (e.g. infant mortality, road accidents, preventable diseases) and overall better health in the community, the trend of increasing longevity is expected to continue, albeit at a reduced rate. Life expectancy for those born in 2018 is estimated to be 82 years for males and 85 years for females. Life expectancy for those born in 2056 is expected to have increased to 87 years for males and 88 years for females.

The Age-Specific Death Rates (ASDR) used in VIF2019 are continuing medium term trends in life expectancy improvements. They resemble more the ‘high life expectancy’ rates used by the ABS in its latest projections, than the ’medium life expectancy’ assumptions which were revised down considerably since the previous ABS projections and DELWP considered too pessimistic.   While there are regional variations in mortality rates, the data are too volatile to exhibit useable trends. Hence, regional differences in ASDRs have not been incorporated in the VIF2019 projections.

Despite the expected increases in life expectancy, the number of deaths in Victoria is projected to continue to rise, due to the larger population base.

Net overseas migration:

Net overseas migration (NOM) is the difference between people coming to live in Australia and residents leaving to settle overseas.

In the past, natural increase has generally contributed more to Australia's annual population growth than has overseas migration. However, since 2004-05, overseas migration has overtaken natural increase as the major contributor to population growth. In 2018, NOM accounted for over 60% of the annual population growth of Victoria (ABS cat. no. 3101.0).

It is important to note that NOM is not the same concept as permanent migration. To be counted in the resident population, thus potentially in the NOM, a person has to live in Australia for at least 12 of the past 16 months, regardless of nationality, citizenship or legal status. In recent years, the largest contribution to NOM has been from people on temporary visas. Temporary visa holders include students, tourists, working holiday makers and temporary business migrants.

In developing assumptions for NOM, Victoria in Future considers broad expectations for the main categories of arrivals to Australia (permanent, temporary, students, business etc) and the expected ratio of arrivals to departures in each category (eg temporary migrants are more likely to leave the population again within a short period of arrival). Also important is the share of each category of arrival and departure expected to be to/from Victoria compared to other states and territories.

Currently Victoria is receiving record levels of NOM, in the vicinity of 90,000 per annum, as the Australian total is large and Victoria’s share of national NOM is high. However, overseas migration can vary greatly from year to year. The variations can be affected by economic conditions in Australia and overseas, as well as policy changes in Australia. The overseas migration assumption used in these projections should be seen as a long term annual average, rather than a prediction that overseas migration will be at a particular level in a particular year.

VIF2019 expects that Victoria will continue to experience very strong NOM in the short term, while Victoria and NSW remain the most attractive states for overseas migrants. Over the longer term there is an assumption that other states will become more attractive, on average. VIF2019, therefore, reduces the expected NOM to Victoria over time such that it averages approximately 76,000 over the whole period to 2056. It should be noted that while this level is below the current extraordinary level, it is still very high in historical terms.

Net interstate migration:

Net interstate migration (NIM) between Victoria and the other States and territories had been a small component of population change since the 1990s. In recent years this has changed and Victoria has been experiencing positive net migration from all other states and territories. In 2017 approximately 86,000 people migrated into Victoria and 68,000 people moved out, leading to a net movement of 18,000 people. The net migration was somewhat smaller in 2018, with 14,000 people.

VIF2019 assumes that NIM will return to lower levels and contribute 8,000 per annum to the Victorian population for most of the projection period. As the age structures of exiting and entering interstate migrants are not significantly different, a small positive NIM is not expected to have a marked effect on the projected population beyond its simple additive contribution

DELWP takes into account local conditions and differences in developing the VIF2019 projections.

Locally specific fertility rates, overseas, interstate and local migration patterns are used in VIF2019. Common mortality rates are applied to all regions.

One key input to the VIF projections is the availability of land and housing, including data on local vacancy rates, recent development trends and capacities for additional dwellings. The Department’s Urban Development Program (UDP) provides updated analysis on the supply of and demand for residential and industrial land across the metropolitan Melbourne, Geelong and other regional centres over the short to medium term. In developing the VIF projections, DELWP takes into consideration the UDP information as well as information gathered through consultations with local governments, property and development experts and other stakeholders who have localised knowledge. For Metropolitan Melbourne, Housing Development Data which monitors the size and location of infill developments in residential areas is used to estimate the extent of future dwelling additions.

Whilst the vast majority of the population lives in households in private dwellings, non-private dwellings, which include for example hospitals, aged-care facilities, correctional institutions, boarding schools or army barracks, can represent a significant part of the population of a small geographical area. People living in these institutions tend to have specific demographic characteristics. For example, a boarding school represents a constant population of school-aged children.

In producing the projections for small areas, allowance is made for known sizeable institutions, whether they are likely to increase in size and the specific demographic characteristics of their inmates.

The numbers and characteristics of people living in non-private dwellings are explicitly accounted for in the household formation component of the VIF2019 projections, before the rest of the population is allocated to households.

The concept of ‘resident population’, as defined by the ABS and projected in VIF, is based on where a person is usually living, regardless of nationality, citizenship or legal status.  A person is regarded as a usual resident if they have been (or expected to be) residing in Australia for 12 months or more over a 16-month period, with the exception of foreign diplomatic personnel and their families. This 12-month period does not have to be continuous, to make allowance for brief stays overseas, which can be a common pattern for some long-stay migrant groups such as overseas students, business migrants and temporary employer-sponsored workers as well as Australian travellers.

In modelling overseas migration flows, DELWP makes separate analyses of, and assumptions for, the contribution of different categories of migrants. For example, the relationship between arrival and departure for temporary migrants versus permanent migrants (only a certain proportion of temporary migrants are assumed to become permanent residents following their short term stay).

The VIF2019 projections are based on the estimated resident population as at 30 June and not on seasonal, temporary fluctuations. People are counted at their place of usual residence.

Places with large ‘temporary’ populations, like holiday destinations or farming areas with very seasonal activities, often have low dwelling occupancy rates for most of the year. While the occupancy rates in some Victorian coastal towns are increasing as permanent settlement in these areas grows, it is not uncommon to have winter occupancy rates well under 50% in some locations.

These low occupancy rates (or high vacancy rates) are recognised in the Housing Unit model part of the projection process.

VIF2019 provides a table of projected occupancy rates for all projection areas from 2016 to 2036. Areas which include significant temporary populations can be identified as those with especially low occupancy rates.

Users of VIF2019 projections should be aware of these possible seasonal changes and may need to apply local knowledge to the projections to suit their particular purposes. For example, businesses and organisations providing services may need to be aware of both the usual resident and holiday peak populations in their business/service planning.

Over the projection period net migration is expected to account for 67% of the projected growth and natural increase for the remaining 33%. Of the 4.75 million extra population expected between 2018 and 2056, 1.56 million are expected to come through natural increase, 2.88 million through Net Overseas Migration (NOM) and 306,000 through Net Interstate Migration (NIM).

Victoria’s NOM is expected vary between a high of 84,000 and a low of 72,000 with an average of approximately 76,000 per annum. Victoria’s NIM is expected to decrease, on average, from its current high of over 14,000 per annum, to a gain of 8,000 per annum in the long term.

The number of births per annum in Victoria is projected to increase from 82,000 in 2019 to 120,000 in 2056. The number of deaths is also expected to increase, from 39,000 in 2017-18 to 80,000 in 2056. Thus, natural increase is projected to return to around 39,000 for most of the projection period.

The population of Melbourne will increase strongly through NOM, as over 90% of overseas migrants to Victoria are expected to settle there. But Melbourne is also expected to retain strong natural increase throughout the projection period.

In contrast, the main driver of growth in regional Victoria will be migration from Melbourne. It is projected that population ageing and the increasing numbers of deaths in regional Victoria will result in lower levels of natural increase in regional Victoria.

Within Metropolitan Melbourne, the areas expected to increase the most are those with the greatest capacity for dwelling growth, namely the outer Growth Areas and the inner city. While the middle suburbs are expected to regenerate and increase steadily in population, the designated Growth Areas including the Wallan SA2 (52 per cent) and the three inner LGAs of Melbourne, Port Phillip and Yarra (11 per cent) are expected to account for more than three fifths of population growth from 2018 to 2036.

In Victoria’s regions the largest numbers of projected extra dwellings, and thus largest concentrations of population growth, are in the major regional cities and the areas close to Melbourne. Between 2018 and 2036, the three largest LGAs (Greater Geelong, Ballarat and Greater Bendigo) are projected to account for around half of the population growth in Victoria’s regions.

Top five local government areas, largest and fastest growth 2011 to 2031:

Maps showing projected changes, in numbers and average annual rates, for LGAs in Victoria between 2018 and 2036 are published in the VIF2019 brochure.

A household is defined as one or more persons, at least one of whom is at least 15 years of age, usually resident in the same private dwelling.  Thus, the number of households equals/represents the number of dwellings needed to house the population. Many services are planned and consumed per household, rather than for individuals. The number of households changes when complete households move in or out of an area because of migration. New household formation also occurs when a person leaves the existing household to form a new household. This may occur, for example, when an adult child leaves the parental home to partner or live alone, or when a couple separates/divorces. Households also disappear when people previously living separately move in together, or when people move into non-private accommodation.

Thus, household number growth is function not only of the size of the population, but also the way in which individuals form themselves into households, which is related to the age structure of the population, partnering and de-partnering trends, the age at which children leave the parental home, and a range of socio-cultural factors.

Over the last twenty years, the average household size in Victoria has decreased, and household numbers have increased faster than the total population.

This is the results of several factors:

• an increase in the proportion of the population in the older age groups, which traditionally have the highest propensity to live in small households (alone or in a couple without children);
• an increase in the propensity for persons of most ages to live alone, and
• a decrease in the average number of children in families and an increase in the number of sole parent families.

Changes in household formation propensities over time are extremely difficult to measure and problematic to forecast. Consequently, such changing patterns have generally not been integrated in the VIF2019 projections model. The exception is the propensity for older people to live alone. The life expectancy of males is increasing at a faster rate than that of females, and therefore the gap between male and female life expectancy is decreasing. As this is assumed to continue, a lower proportion of older people are expected to be living alone at the oldest ages in the future.

Nonetheless, the projected changes in the population age structure result in a further decrease of the average household size. VIF2019 projects that average household size will decrease from 2.54 in 2016 to 2.40 by 2056.

The Australian population is said to be ageing because the proportion of people in the older age groups is increasing. This change in the age structure reflects two separate phenomena.

Firstly, life expectancy at all ages, especially the older ages, has risen, due to improvements in medical science and the quality of health care, and generally healthier diets and lifestyles. This increases the average age of the population.

Secondly, over the coming years, more of the ‘baby boomer’ cohort will reach the age of 65 and become ‘old’. This cohort is large in Australia. In Victoria there are currently 1.3 million people born between the years 1946 and 1965. This cohort was boosted by the large number of births as a result of high fertility rates and large number of young adult migrants in the 1950’s and 1960s.

However, over the projection period, the increase in births and continued migration of young adults will increase the number of young people in the population, therefore slowing the rate of ageing.

The full set of VIF2019 products is available on the Victoria in Future website.

For further information or assistance about VIF2019, please contact Forward Policy and Research

The Estimated Resident Population (ERP) is the official measure of population, used as the base (starting) population for Victoria in Future projections.

The ERP is based on the concept of usual residence. It is derived from the count from the latest Census, with the addition of quarterly components of population growth (births, deaths, population movements). It incorporates the estimated Census net undercount (to account for people who were missed from or counted more than once in the Census) and is adjusted to include usual residents who were temporarily overseas at the time of the Census and to exclude overseas visitors who were temporarily in Australia on Census night.

The ABS produces ERPs for Australia, its states, territories and sub-state/territory regions and for local government areas (ABS cat. no. 3301.0, 3218.0 and 3235.0 respectively). Australian and state data is published quarterly, with a yearly age/sex distribution. Sub-state data is published annually, with a 5-year-age group/sex distribution.

The Census of Population and Housing is conducted every five years under the Census and Statistics Act 1905 and aims to measure accurately the number of people and dwellings in Australia on Census night, and a range of their key characteristics. The most recent Census of Population and Housing was conducted on 9 August 2011. The Census contains two measures of population,

• Population by place of enumeration, based on where people were on Census night,
• Population by place of usual residence, based on people’s declared usual address.

In contrast with the Estimated Resident Population, the Census population includes short-term visitors and excludes people temporarily absent from Australia.

Victoria in Future projections are an estimate of the future size, distribution and composition of the population. They are developed using mathematical models and expert knowledge, relying on trend analysis and assumptions about future change. VIF2019 is not an exact predication or forecast of the future.

Uncertainty about the future increases over longer projection horizons and with smaller geographic disaggregations. Different policy settings and changes in the economy could result in changes to the expected size, distribution and characteristics of the population.

The births and deaths data used in VIF2019 are consistent with those used in the ABS resident population estimates, based on usual residence and year of occurrence.

Natural increase is the difference between the number of births and deaths occurring during the period. If the number of deaths exceeds the number of births, it is called natural decrease.

NOM is the difference between the number of incoming travellers who stay in Australia for 12 months or more and are added to the population (NOM arrivals) and the number of outgoing travellers who leave Australia for 12 months or more and are subtracted from the population (NOM departures). NOM is based on an international traveller's duration of stay, i.e. being in or out of Australia for 12 months or more, regardless of nationality, citizenship or legal status. Note that this 12-month period does not have to be continuous and is measured over a 16-month reference period.

NIM to Victoria is the balance between the number of people moving to Victoria from other States and Territories and those moving from Victoria. Data on interstate migration cannot be directly estimated. Instead, post-censal estimates of interstate migration are modelled using administrative by-product data (for example Medicare changes of address).

Net intrastate migration is the balance of movements of population between Victorian regions.

For the smallest geographical areas, migration cannot be measured directly. The combined effects of overseas, interstate and intrastate migrations are estimated as the changes in population not explained by births or deaths.

Implied net migration = Population at end of period – population at start of period – births + deaths

Age-specific fertility rates (ASFR) are the number of live births (registered) during the calendar year, according to the age of the mother, per 1,000 women (female estimated resident population by age at the mid-point of the year 30 June). For calculating these rates, births to mothers under 15 years are included in the 15-19 years age group, and births to mothers aged 50 years and over are included in the 45-49 years age group. Pro rata adjustment is made for births for which the age of the mother is not given.

The total fertility rate (TFR) is the sum of age-specific fertility rates (live births at each age of mother per 1,000 women of that age) divided by 1,000. It represents the number of children a female would bear during her lifetime if she experienced current age-specific fertility rates at each age of her reproductive life.

In VIF2019, future changes in ASFRs are modelled and the ASFR applied to the expected future female population by age to project the number of births each year. The result of the ASFR modelling is that the TFR declines from around 1.65 recently to just over 1.6 by 2036.

Age (and sex)-specific death rates (ASDR) are the number of deaths (registered) during the calendar year, at a specified age, per 1,000 of the estimated resident population of the same age (and sex) at the mid-point of the year (30 June). Pro rata adjustment is made in respect of deaths for which the age of the deceased is not given.

Life expectancy refers to the average number of additional years a person of a given age and sex might expect to live if the age-specific death rates of the given period continued throughout his/her lifetime. While life expectancy can be expressed as years of life remaining at any age, in comparing projections it is usual to refer to life expectancy at birth.

In VIF2019, projections of ASDR are applied to the expected population to calculate the number of deaths each year. This modelling results in a life expectancy at birth of 87 years for men and 88 years for women by 2056.

In general terms, a dwelling is a structure which is intended to have people live in it, and which is habitable. Some examples of dwellings are houses, motels, flats, caravans, prisons, tents, humpies and houseboats.

Both private dwellings and non-private dwellings are counted in the Census. People in non-private dwellings (hotels, hospitals, prisons, etc.) are enumerated and counted in the population, but do not form households.

Occupied and unoccupied dwellings are counted. In VIF2019, it is assumed that each occupied dwelling contains one household (it may be more than one family).

The proportion of private dwellings which are not occupied is called the Vacancy rate. In some areas, such as holiday destinations, the vacancy rate can be quite high, exceeding 50%.

The Occupancy rate is the proportion of dwellings which are occupied. By definition, Occupancy rate + Vacancy rate = 1. Victoria in Future modelling and outputs refer to the occupancy rate, as it is the occupied dwellings which accommodate the population.

A household is defined as one or more persons, at least one of whom is at least 15 years of age, usually resident in the same private dwelling. Therefore, the total number of households is equal to the total number of occupied dwellings.

Households are classified according to their composition, into the following main types:

• Couple with children
• Couple without children
• Lone parent family (with children)
• Other families of related persons household
• Lone person household
• Group household
• Others household

All these households (except the lone person households) can also include other people, related or not to the household reference person.

The term ‘child’ describes the relationship to the reference person in the family, not the age of the person. Children can be of any age.

Persons in non-private households are classified separately.

The distribution of people by age according to their position in the household is used to analyse and project household formation. This analysis has shown that the impact of the population age structure is most significant, while the effect of changes in living arrangement propensities is volatile and difficult to predict. Accordingly, in VIF2019 most of the living arrangement propensities are held constant throughout the projection period.

The average household size is calculated as the total number of people in occupied private dwellings divided by the number of occupied private dwellings. Over the past decade, the increased proportion of small households (lone person, couple without children and one parent families) has led to a decrease in the average household size.

VIF2019 uses the ASGS developed by the ABS. (see ABS cat. no. 1270.0).

VIF2019 publishes projections for the Greater Melbourne Capital City Statistical Area (2GMEL) and the eight Statistical Areas level 4 (SA4) in the remainder of Victoria, namely:

Ballarat

Bendigo

Geelong

Hume

Latrobe-Gippsland

North-West

Shepparton  and

Warrnambool and South West.

To provide more local information, VIF2019 publishes projections for local government areas (LGA)  and for a set of specifically designed small areas within each LGA. A list of these Victoria In Future Small Areas (VIFSAs) and maps are available through this link (add the link here).

For the first time in VIF2019, projections will also be published for ASGS areas as small as Statistical Area Level 2 (SA2). In built-up areas an SA2 is usually roughly equivalent to a suburb or town. In rural areas SA2s can be much larger.

The growth areas include the six LGAs of Cardinia, Casey, Hume, Melton, Whittlesea and Wyndham. When the data allows, Wallan, Beveridge and the surrounding areas (within the Shire of Mitchell) are also included.

The Inner City LGAs are Melbourne, Port Phillip, and Yarra.

The middle suburb LGAs are: Banyule, Bayside, Boroondara, Brimbank, Darebin, Frankston, Glen Eira, Greater Dandenong, Hobsons Bay, Kingston, Knox, Manningham, Maribyrnong, Maroondah, Merri-bek, Monash, Moonee Valley, Mornington Peninsula, Nillumbik, Stonnington, Whitehorse, and Yarra Ranges.

The Cohort Component model projects the population by age and sex for each period using the basic demographic equation:

Population at end of period = Base population + Births – Deaths + Net migration

The Housing Unit method projects the number of dwellings available for occupation in a given area for each period using the basic equation:

Occupied dwellings at end of period =

(Base stock of dwellings + new construction – demolitions)* (1- vacancy rate)

Using an average number of persons expected in each household, the housing unit model can also be used to determine the total projected population for each area.

The Household Formation model allocates the population into households, based on living arrangement propensities (i.e. the distribution of people of each age and sex across household types).