Using Community Profile to plan for future housing need
Build your housing baseline from population, household and dwelling data – and read unoccupied dwellings and building approvals with their limits in mind
This article is for Community Profile customers. Using Housing Monitor? See: Using Housing Monitor to plan for future housing need.
Community Profile holds the demographic foundation of a housing evidence base: population figures, household types, dwelling occupancy, structure and size, housing tenure and payments, with monthly building approvals alongside. This article explains how to use that data to establish your housing baseline, and how to read the figures that are most often misread – particularly in regional and remote areas, where second homes and transitory workforces are part of the housing story.
Key concepts
Unoccupied dwellings
'Unoccupied' in the Census means unoccupied on Census night. It does not mean vacant. A dwelling recorded as unoccupied could be vacant long term or short term, or its residents could simply have been away on that one night. The ABS codes people back to their area of usual residence, but not to their dwelling of usual residence, so a dwelling is recorded as unoccupied even where its usual residents were counted elsewhere and gave their home address on the form. The Census is a snapshot at a point in time, and the unoccupied count represents only that one day every 5 years.
Visitor only and other non-classifiable households
Two household categories are important, but often not shown by the ABS in standard output. They are particularly important in areas with second homes or transitory workforces. A 'visitor only household' is a dwelling that housed only people whose usual address is elsewhere on Census night. The usual address could be next door, elsewhere in Australia or overseas. It is treated as occupied (because it was occupied on Census night), although the data cannot say whether the dwelling is usually vacant. An 'other non-classifiable household' is a dwelling the Census field assessment judged to be occupied, but where no contact was made and no form was returned even after repeated visits – in some cases this may be a Census refusal. It is also treated as occupied, but nothing is known about the occupants, which makes it the category with the most potential for error. Check your place of interest on the Community Profile - it’s under the “Households summary” page and ID always shows both these categories - they are often missing from the raw ABS data. In areas with young and transient populations, often both visitor only and other non-classifiable categories have higher percentages than the state benchmark, and comparing your area's shares with the benchmark is a quick first test of how much weight the headline occupancy figures can bear. High rates of these categories also affects the percentages of other household and dwelling types, making them appear smaller in proportion.
Building approvals
The building approvals series in Community Profile is sourced from the ABS Building Approvals collection. It is updated monthly and records gross additions only – dwellings approved to be built, with no account taken of demolitions.
Net change in dwelling stock can only be derived from the five-yearly snapshots at Census time. An approval is also not a completed dwelling, so approvals indicate what is coming rather than what has been delivered. In some cases dwellings with building approval never get built.
Population figures
The Census count, usual residence figures and the annual Estimated Resident Population (ERP) are different measures, and the right one depends on the job. For housing planning, the important point is that a difference between them is not an error. See Which population figure should I use in Community Profile? for the full explanation.
Establishing your housing baseline
In many regional and remote areas, a declining population alongside a high count of unoccupied dwellings can make it look as though the area has enough housing. That reading often does not match what councils see locally: some town homes are used intermittently by the owners of surrounding rural properties, while others house fly-in fly-out (FIFO) or other transitory workers for part of the year. Dwellings used in these ways are counted in the housing stock, but they are not necessarily available to meet new demand.
To establish a credible baseline, read the occupancy figures as a set rather than relying on the headline rate. The dwelling occupancy data under 'How do we live' and the households summary show occupied dwellings, unoccupied dwellings and the household categories above, with a benchmark comparison. The Census is not a perfect vehicle for measuring occupancy, so where the visitor only or non-classifiable shares run well above your benchmark, treat the headline occupancy figures with caution and test them against local knowledge, such as rates records, where possible.
From today's baseline to future need
Your population forecast treats occupancy differently from the Census. The occupied and unoccupied rate in the forecast is not the Census figure: it is .id's best estimate – and only an estimate – of how many dwellings do not house usual residents, so it is usually lower than the Census rate, with a factor applied for residents temporarily away on Census night.
The population forecast does not itself measure housing need. Where your state government's projections are the required population basis for planning, they set the population total, but they do not tell you how many households or dwellings of what type your area will need. Estimating the number and type of homes needed combines forecast household change with the housing stock and the dwellings different household types occupy – Housing Monitor carries that data where your organisation subscribes, and Using Housing Monitor to plan for future housing need sets out a practical approach. .id can also scope a fuller housing needs analysis as consulting work.
Frequently asked questions
Q: A large share of our dwellings were 'unoccupied' at the Census. Does that mean we have enough housing?
A: No. Unoccupied means unoccupied on Census night, not vacant or available. The count includes dwellings that are genuinely vacant alongside homes whose residents were simply away on the night and second homes used intermittently, and the ABS does not code people back to their dwelling of usual residence. A high unoccupied count is a prompt to look closer – at the visitor only and non-classifiable shares, the benchmark comparison and local evidence – not a measure of spare supply.
Q: How do FIFO and other transitory workers show up in the data?
A: Workers who gave a usual address elsewhere are counted in their home area's population, not yours. The dwelling they were staying in appears either as a visitor only household, which is treated as occupied, or as unoccupied if it was empty on the night. Either way, homes used by a transitory workforce for part of the year are counted in the dwelling stock without being available to new households, so areas with large transitory workforces should expect elevated visitor only and unoccupied figures and read them in that light.
Q: Why is the unoccupied rate in our population forecast lower than the Census figure?
A: This is because the forecasts do not use the Census figures for population. Instead, the forecasts use the revised estimated resident population published by the ABS post-Census. The forecast unoccupied rate is because of ID’s reconciliation of available dwellings with the calculated number of households, and the number of people contained within these households. The number of households corresponds to occupied dwellings, and the unoccupied rate is a result of this reconciliation. Owing to the different population figures, vacancy rates will differ from the Census count.
Q: Do building approvals account for demolitions?
A: No. Building approvals are gross additions only. In many cases demolitions are negligible and can be ignored, such as a greenfield estate or a new apartment building, but in redevelopment areas – particularly public housing renewal – they can be a significant factor. The population forecast and the residential development forecast estimate net dwelling additions, so use those where net change matters; net change in the existing stock can otherwise only be derived from Census-to-Census comparison.
Q: Do building approvals tell us what has been built?
A: No. An approval records a dwelling approved to be built, not a completed one, so the approvals series indicates the pipeline rather than delivered supply. Measure actual change in the stock from Census-to-Census change in dwellings, or from the net dwelling additions in your forecast.
Q: Why don't the dwelling figures in Community Profile match ABS QuickStats?
A: In its standard outputs, the ABS breaks down dwelling structure only for occupied private dwellings and excludes visitor only and non-classifiable households, even though dwelling structure is collected for all dwellings. Community Profile presents the data on a different basis, so the two sets of figures will not match. See Why Census figures in .id tools can differ from the ABS website for the broader explanation.
Q: We don't have demographic specialists in-house. Can .id help us interpret the data?
A: Yes. Training and briefing sessions are part of your subscription and are encouraged annually, and you can contact support@id.com.au with interpretation questions at any time. Where you need analysis beyond the tools, such as a housing needs assessment, your .id client lead can scope it as consulting work.