Australia promised 1.2 million homes. When will each one finish?

The target is fixed. Pace, cost and build time are not.

The national housing council expects 980,000 homes in the five Accord years, and the 1.2 million mark more than a year late. It made that estimate before recent global disruptions to commodity prices and supply chains.

The federal programs look the same from inside: 1,432 homes built of 40,000 at 30 April 2026, most completions forecast for the final years, and delivery risk rated high for most of the period. An approval is not a home.

• 1.2 millionnew homes targeted over five years from mid-2024 under the National Housing Accord.National Housing Supply and Affordability Council [1]
• 980,000homes the housing council estimates will be supplied over the five Accord years, 220,000 short of the target.National Housing Supply and Affordability Council [1]
• 308,000homes completed with 7 of the 20 quarters gone, around one-quarter of the target.National Housing Supply and Affordability Council [2]
• 1,432 of 40,000social and affordable homes built under the two federal funding programs at 30 April 2026.Australian National Audit Office [3]
• 51%rise in house construction costs since before COVID-19, as at the June quarter 2026.National Housing Supply and Affordability Council [2]
• 32.9 monthsaverage time from approval to completion for an apartment in 2024-25.Master Builders Australia, using ABS data [4]
• Halfas many homes are completed per hour worked as in 1995.Productivity Commission [5]
• 248,733dwellings under construction in the June quarter 2026: the pipeline whose finish dates the system is counting on.Australian Bureau of Statistics [8]
• 3,325construction insolvencies in the 12 months to March 2026.National Housing Supply and Affordability Council [1]
• 206,000households on the social housing waitlist at 30 June 2025, up from 183,000 a year earlier.Australian Institute of Health and Welfare [6]
• 73.2%of public housing dwellings met minimum acceptable standards in 2025, a share that has fallen since 2016.National Housing Supply and Affordability Council [1]
• 53%of surveyed New South Wales strata buildings reported a serious defect in 2025, waterproofing the most common.Building Commission NSW [11]

Six questions worth testing first.

Public Australian evidence is thin. The clearest study used Victorian permit records to estimate building cost; its best model explained about two-thirds of the variation (r² of 0.66) [7]. CSIRO's RapidRate estimated energy ratings for 910,247 homes at an R² of 0.80 [9]. NSW linked service records to show that 32% of homelessness presentations can be attributed to 1% of the population [10].

Each question below is a decision made years before handover. For each: the gap in what is published, the test, the public data, and the pass mark.

1. When will this project finish?

   Federal completions are weighted to the final years, and the Auditor-General records delivery risk as high for most of the period [3].

   The gap

   Statewide average build times are published. Whether a project's storeys, prices and approval quarter improve on that average is not. The one Victorian study is on cost, not time [7].

   The test

   Model approval-to-completion time for multi-unit projects from dwelling type, storeys, state, approval quarter and input prices at commencement. Score it on unseen projects against the state's rolling average.

   Public data

   ABS Building Activity, ABS Producer Price Indexes, Victorian building permit records. No public dataset records final cost against contract cost, so overrun work needs a provider's own project records.

   Pass mark

   Lower error, in months, than the rolling average. If not, the published average was enough.

2. How many approved homes will start, and when?

   Starts depend on build cost, sale and rent values, and the cost of finance. The housing council modelled two scenarios because it could not tell a short cost spike from a longer shock [1].

   The gap

   Last year's conversion rate is the rule in use. We found no published test of whether construction prices, set against dwelling prices, predict starts better.

   The test

   Predict the share of approved higher-density homes that start within four quarters, from construction input prices and established dwelling prices. Score it on held-back quarters against last year's conversion rate.

   Public data

   ABS Building Approvals, ABS Lending Indicators, Reserve Bank statistical tables.

   Pass mark

   Forecast starts closer to actual starts than last year's rate, quarter by quarter.

3. Will the trades be there when the slab is down?

   Labour shortages continue, worst in Western Australia and Queensland, where housing competes with infrastructure and mining for the same workers [1].

   The gap

   Shortages are reported. We found no published result on whether a rise in engineering and non-residential work slows house and townhouse completions two to four quarters later.

   The test

   Add each state's engineering and non-residential work, lagged two to four quarters, to a build-time forecast. Compare it with a forecast that uses residential activity alone.

   Public data

   Jobs and Skills Australia shortage list, ABS Job Vacancies, NCVER apprentice and trainee collection.

   Pass mark

   Lower forecast error, state by state, on held-back quarters.

4. Which homes will need the most repair next year?

   The share of social housing meeting minimum standards has fallen since 2016 [1]. That points to maintenance money arriving after the fault.

   The gap

   Public repair figures are statewide totals, and age is the rule in use. Whether age, construction type and climate rank repair cost better than age alone can only be scored on a provider's own orders.

   The test

   Rank the tenth of dwellings likely to cost most in responsive repairs next year, from past orders, age, construction type and local climate. Compare it with a ranking by age alone.

   Public data

   Report on Government Services, housing, AIHW National Social Housing Survey, Bureau of Meteorology climate data. Public data is aggregate. Dwelling-level work needs the provider's own repair and condition history.

   Pass mark

   A larger share of next year's repair cost in the model's top tenth than in the age-only top tenth.

5. Which buildings will carry serious defects?

   More than half of the NSW strata buildings surveyed in 2025 reported a serious defect, most often waterproofing [11]. An insurer or funder needs to know where to look first.

   The gap

   The survey is published only in aggregate [11]. Whether defects concentrate in buildings that overran the typical build time for their size is not known at building level.

   The test

   Model the likelihood of a waterproofing defect report in apartment buildings completed since 2018, from each build's overrun for its size, holding height and registration year constant.

   Public data

   Building Commission NSW strata defects research, NSW Planning Portal open data, Victorian building permit records, ASIC insolvency statistics. Survey data is published only in aggregate and inspection records are not open; building-level work needs the regulator's or the insurer's records.

   Pass mark

   On held-back buildings, defect reports concentrate where the model points.

6. Where will the waitlist grow next?

   The public housing waitlist rose by nearly 21,000 households in a year, the largest rise in over a decade [1]. Homes take years to deliver, so they must be sited where need will be.

   The gap

   Queensland publishes its social housing register, and population growth is the rule in use. We found no published local model that tests rent-assistance counts and advertised rents against it.

   The test

   Predict the 12-month change in new social housing applications in each Queensland local government area, from rent-assistance counts and advertised rents. Compare it with population growth alone.

   Public data

   Queensland Social Housing Register, DSS payment data, ABS Regional Population.

   Pass mark

   Forecast applications closer to actual, area by area, on a held-back year.

We start with the published number. We return a range.

Our own method takes the number already in use and returns a range for this project, this paddock, this farm. It is scored on records the model has never seen. If it does not beat the number in use, we say so.

For housing, the method covers the finish date, starts from approvals, trades on the day the slab goes down, next year's repairs, defect risk and waitlist growth. Each is a range, set against the statewide or last-year figure, on the day money is committed.

1. Lock the ruleThe number in use today: a statewide average, a map cell, last year's rate, a no-change forecast.
2. Build the modelOn Australian records, for the decision that has to be made now.
3. Give a rangeA range you can check. Not a single figure.
4. Score itOn records it has never seen, against the rule. Misses are named.
5. Keep scoringAs each quarter, season or five-minute interval arrives.

What stays private

The models, the features, the scoring, and the way a miss shapes the next range. None of it is published. Client data stays with the client. This site shows the questions and the tests, not the working.

Every figure on this page comes from one of these publications. Official statistics are revised, so check the source before relying on a number.

1. National Housing Supply and Affordability Council. State of the Housing System 2026. 30 April 2026. https://nhsac.gov.au/sites/nhsac.gov.au/files/2026-04/ar-state-housing-system-2026.pdf
2. National Housing Supply and Affordability Council. Quarterly Report, August 2026. 21 August 2026. https://nhsac.gov.au/reports-and-submissions/quarterly-report-august-2026
3. Australian National Audit Office. Auditor-General Report No. 3 of 2026-27: Design and Delivery of the Housing Australia Future Fund and the National Housing Accord Facility. 21 July 2026. https://www.anao.gov.au/work/performance-audit/design-and-implementation-of-the-national-housing-accord-and-the-housing-australian-future-fund-0
4. Master Builders Australia, using ABS data. Turning point in home building times. 20 October 2025. https://masterbuilders.com.au/?p=12404
5. Productivity Commission. Housing construction productivity: Can we fix it? (media release). 16 February 2025. https://www.pc.gov.au/media-speeches/media-releases/2025/reversing-30-years-of-poor-productivity-growth-can-help-fix-housing-affordability
6. Australian Institute of Health and Welfare. Housing assistance in Australia 2026: Households and waitlists. 23 June 2026. https://www.aihw.gov.au/reports/housing-assistance/housing-assistance-in-australia-2026/contents/households-and-waitlists
7. Gurmu and Miri, Construction Innovation 25(2). Machine learning regression for estimating the cost range of building projects. 2023. https://www.emerald.com/insight/content/doi/10.1108/CI-08-2022-0197/full/html
8. Australian Bureau of Statistics. Building Activity, Australia, June 2026. 7 October 2026. https://www.abs.gov.au/statistics/industry/building-and-construction/building-activity-australia/latest-release
9. CSIRO. Australian Housing Data portal: RapidRate energy ratings. read 10 November 2026. https://ahd.csiro.au/
10. Homes NSW. Pathways to Homelessness: early identification of people at risk of homelessness. read 10 November 2026. https://www.nsw.gov.au/departments-and-agencies/homes-nsw/social-housing-resources/pathways-to-homelessness/key-findings-early-identification-of-people-at-risk-of-homelessness
11. Building Commission NSW. 2025 Strata Defects Survey. 1 November 2025. https://www.nsw.gov.au/departments-and-agencies/building-commission/building-and-construction-resources/2025-strata-defects-research-report