How population growth has changed housing demand in Australia since 1990

Australia's population rose from 17.07 million in June 1990 to 27.80 million in December 2025. In raw terms, the completed decades added progressively more people: 1.96 million in the 1990s, 3.00 million in the 2000s and 3.62 million in the 2010s. A further 2.15 million people were added in the five and a half years from June 2020 to December 2025.

Raw increases alone are not a fair comparison, since each decade started from a larger population than the last. Measured instead as a compound annual growth rate, the 2010s were the fastest of the four periods at 1.53% a year, followed by the 2000s and the partial 2020s period, both at 1.48%. The 1990s grew the slowest, at 1.10% a year.

Population growth since June 2020 has therefore matched the annual pace recorded in the 2000s, rather than setting a new record.

Overseas migration accounted for 71% of Australia's population growth over the five years to 2024–25, compared with 51.1% in the early 2000s. In the latest completed financial year, 2024–25, its share was 72.5%.

The shift has not been a smooth trend. Overseas migration fell to a net outflow of 85,000 people in 2020–21 during Australia's international border closure, before reaching a record year-ending level of 556,000 in the September quarter of 2023 as arrivals increased after the border reopened.

The most recently completed financial year, 2024–25, recorded overseas migration of 302,200 in the quarterly population series. The ABS's separate annual migration release publishes 306,000 for the same year after later revisions.

The clearest available benchmark projects a 37,000-dwelling shortfall over the 2024–25 to 2028–29 Housing Accord period, with net new supply of 862,000 dwellings against new underlying demand of 900,000 households.

The National Housing Supply and Affordability Council describes this as a forward-looking flow model. It does not represent the total housing shortfall accumulated before the Accord period.

No single, consistent estimate exists for cumulative housing undersupply since 1990. A genuine supply-demand balance needs two comparable flow measures: net new supply after losses, and new underlying demand.

The two series, gross completions and Census household counts, do not meet that standard. Completions data do not track losses to the dwelling stock, and no consistent annual household-formation series exists back to 1990.

New South Wales recorded a net interstate outflow in every year from 1989–90 to 2024–25, while Queensland recorded a net inflow in every year over the same 36-year period.

Net interstate migration measures the net movement of residents between Australia's states and territories and is separate from overseas migration. Queensland and Western Australia, both states with large resources sectors, recorded different interstate migration patterns during the resources boom.

Across the financial years ending 2003 to 2013, an analytical period used in this article rather than an official ABS classification, average annual net interstate migration was:

• Queensland: a net inflow of about +20,100 people a year.
• Western Australia: a net inflow of about +4,100 people a year.
• New South Wales: a net outflow of about −21,500 people a year.

Western Australia's smaller interstate inflow does not mean the resources boom generated less labour demand there. The Productivity Commission found that temporary overseas migration and fly-in fly-out arrangements were important sources of labour for resources regions, alongside permanent relocation.

Average annual population growth rose from 229,600 people a year over 2000–01 to 2004–05 to 392,360 a year over 2020–21 to 2024–25. Most of the increase came from overseas migration rather than natural increase.

In the earlier period, natural increase supplied 52.5% of growth and net overseas migration supplied 51.1%. In the more recent five years, a period that includes the COVID border closure and reopening catch-up, it supplied 71% of growth.

This makes recent population growth both faster and more migration-dependent than in the early 2000s. This has made annual growth more sensitive to changes in migration policy, international travel and temporary visa movements, including student arrivals and departures.

Australia's population has grown every decade since 1990, from 17.07 million to 27.80 million. Three major demand drivers changed over that period:

• Migration now drives most of the growth. Net overseas migration has gone from contributing about half of population growth in the early 2000s to close to three-quarters of it today, making growth faster but also more sensitive to migration policy and global conditions.
• Households are forming faster than population growth alone explains, partly because the average household has grown smaller, adding to demand for separate dwellings even in years when population growth was slow.
• Internal migration patterns have changed by state. Queensland and Western Australia have responded differently to both the mining boom and COVID, while New South Wales and Victoria have recorded net interstate migration losses through most of the past decade.

No single, methodologically consistent estimate exists for cumulative housing undersupply since 1990. The clearest current official benchmark is NHSAC's forward-looking Housing Accord outlook, which compares annual net new supply with new underlying demand.

Across the 2024–25 to 2028–29 Accord period, NHSAC projects a shortfall of 37,000 dwellings. Annual net supply is projected to exceed annual demand from 2027–28, but the later annual surpluses are not large enough to remove the published shortfall across the full five-year period.

Beyond 2040, the ABS and Centre for Population project continued population growth, while the ABS also projects continued household growth. The pace of growth is expected to ease under the central scenarios, but the projections remain dependent on assumptions about migration, fertility, mortality and household living arrangements.