a/work· 31 October 2026 · 5 min read

Australia has eight empty homes for every homeless person. Who is the tax system paying?

Art: Omxus Press

Murray Barr, a former Marine, lived on the streets of Reno, Nevada, for about ten years. The two police officers who kept picking him up added up what his hospital bills and treatment had cost. "It cost us one million dollars," one said, "not to do something about Murray" (Gladwell, 2006).

Look how clean this pool is. Nobody's swimming in it.

Nobody had that million to give him a flat. We found it for ambulances. Most of us assume homelessness is a shortage: of homes, of money, of people willing to change. Australia took a count that says otherwise.

One night in August

On the night of 10 August 2021, Australia counted its people. Behind 1,043,776 front doors, nobody lived. The same night, 122,494 people were sleeping rough, in shelters, on couches, in boarding houses or in homes so crowded the Bureau counts them as homeless (Australian Bureau of Statistics, 2022, 2023). That is eight and a half empty dwellings for each of them.

Plenty of those are holiday homes, or between tenants, or half renovated. So ask a smaller question: how many must be usable for every homeless person to have one? One in eight and a half, or 11.7%. Every state clears that bar except the Northern Territory, where 10,404 empty dwellings face 13,104 homeless people, three in four of them in severely crowded homes. There the buildings really are short. Everywhere else, the shortage isn't buildings.

What the tax system pays for

In 2023–24, 1,276,669 Australians claimed $15.34 billion in net rental losses against their other income (Australian Taxation Office, 2026). We valued those deductions at each tax bracket's rate and got about $5.11 billion. The 16.6% of taxpayers earning over $120,000 received 55.5% of it. Grattan Institute found only 8% to 14% of investor lending goes to new dwellings; the rest buys homes that already exist (Daley et al., 2016).

At $20,000 a year for each supported place, one year of that would fund about 255,000 places, twice the number of people counted homeless. Removing the deduction wouldn't raise it all at once: Grattan put the immediate gain from quarantining losses at about $2 billion a year. But the money exists and it is being spent. It goes to the investor, not the tenant.

Meanwhile social housing fell from 4.8% of households in 2011 to 4.1% in 2022, with 188,300 households waiting (Australian Institute of Health and Welfare, 2023).

What a home first does

The old staircase says get sober, get treated, then earn a home. Housing First flips it: a home straight away, with support offered and never forced. It has been tested in randomised trials. In Canada, 2,148 people in five cities spent 73% of two years stably housed, against 32% in usual services (Goering et al., 2014). Pooled, people offered a home first were 2.46 times as likely to be housed up to two years later (Baxter et al., 2019).

It isn't magic. Mental health and drug use come out about the same either way (Aubry et al., 2020), and the evidence is strongest for chronically homeless adults. A home doesn't cure anyone. It gives a person somewhere to be sick, or sober, that isn't a footpath. And for the Murrays, the savings in hospitals and emergency services cover most or all of the cost (Goering et al., 2014). Australia would save about $29,450 per person a year if homeless people's use of those services fell to the average (Zaretzky & Flatau, 2013).

Finland counted, and kept counting

Every 15 November, every Finnish town counts its homeless. In 2008, the year Finland made Housing First national policy, they counted 3,597 people homeless for a year or more. In 2024: 1,010 (Asumisen rahoitus- ja kehittämiskeskus, 2010; Housing Finance and Development Centre of Finland, 2025). Long-term homelessness fell 6.7% a year to 2017 while everyone else counted in the same towns stayed flat.

Then Finland cut income support and ran short of small affordable homes, and in 2025 long-term homelessness rose 29% to 1,306 (Centre for State-Subsidised Housing Construction, 2026). The rise is the proof as much as the fall. Housing First needs homes to give and incomes to pay the rent. Take either away and the numbers move, because the numbers were never about the people.

What you can do

A country that treats housing as an asset gets rising prices, empty dwellings and homeless people as a matter of course. Homelessness isn't a failure of that policy. It's one of its outputs, so it can be changed by changing the policy.

Australian cities have started counting the way Finland does. Advance to Zero keeps lists of every homeless person in a town by name, so that people stop being a statistic and start being a to-do list (Australian Alliance to End Homelessness, n.d.). Find out whether your council takes part, and if not, ask at the next meeting why not. Then give your federal MP two numbers: 1,043,776 and 122,494. Ask what the $5 billion is buying.

The obstacle isn't knowledge, money or buildings. It is the decision to treat a house as a place to live. Housing first. Everything else follows.

A building with a pool and nobody home is a decision. So is a person on the street.

Sources

  1. Gladwell, M. (2006, February 13). Million-dollar Murray. The New Yorker.
  2. Australian Alliance to End Homelessness. (n.d.). Advance to Zero. https://aaeh.org.au/atoz
  3. Asumisen rahoitus- ja kehittämiskeskus [ARA]. (2010). Asunnottomat 2009 [Homeless people 2009]. https://www.varke.fi/fi/document/asunnottomat-2009
  4. Housing Finance and Development Centre of Finland [ARA]. (2019). Homelessness in Finland 2018. https://www.varke.fi/en/document/homeless-people-finland-2018
  5. Housing Finance and Development Centre of Finland [ARA]. (2025). Homeless people 2024. https://www.varke.fi/en/document/homeless-people-2024
  6. Centre for State-Subsidised Housing Construction [Varke]. (2026). Homeless people 2025. https://www.varke.fi/en/statistics/homelessness/homeless-people-2025
  7. Goering, P., Veldhuizen, S., Watson, A., Adair, C., Kopp, B., Latimer, E., Nelson, G., MacNaughton, E., Streiner, D., & Aubry, T. (2014). National At Home/Chez Soi final report. Mental Health Commission of Canada. https://mentalhealthcommission.ca/resource/national-at-home-chez-soi-final-report/
  8. Baxter, A. J., Tweed, E. J., Katikireddi, S. V., & Thomson, H. (2019). Effects of Housing First approaches on health and well-being of adults who are homeless or at risk of homelessness: Systematic review and meta-analysis of randomised controlled trials. Journal of Epidemiology and Community Health, 73(5), 379–387. doi.org/10.1136/jech-2018-210981
  9. Aubry, T., Bloch, G., Brcic, V., Saad, A., Magwood, O., Abdalla, T., Alkhateeb, Q., Xie, E., Mathew, C., Hannigan, T., Costello, C., Thavorn, K., Stergiopoulos, V., Tugwell, P., & Pottie, K. (2020). Effectiveness of permanent supportive housing and income assistance interventions for homeless individuals in high-income countries: A systematic review. The Lancet Public Health, 5(6), e342–e360. doi.org/10.1016/S2468-2667(20)30055-4
  10. Australian Bureau of Statistics. (2022). 2021 Census all persons QuickStats: Australia. https://www.abs.gov.au/census/find-census-data/quickstats/2021/AUS
  11. Australian Bureau of Statistics. (2023). Estimating homelessness: Census, 2021. https://www.abs.gov.au/statistics/people/housing/estimating-homelessness-census/latest-release
  12. Australian Institute of Health and Welfare. (2023). Housing assistance in Australia 2023: Households and waiting lists. https://www.aihw.gov.au/reports/housing-assistance/housing-assistance-in-australia/contents/households-and-waiting-lists
  13. Zaretzky, K., & Flatau, P. (2013). The cost of homelessness and the net benefit of homelessness programs: A national study (AHURI Final Report No. 218). Australian Housing and Urban Research Institute. https://www.ahuri.edu.au/research/final-reports/218
  14. Australian Taxation Office. (2026). Taxation statistics 2023–24: Individuals, table 3. https://data.gov.au/data/dataset/taxation-statistics-2023-24
  15. Daley, J., Wood, D., & Parsonage, H. (2016). Hot property: Negative gearing and capital gains tax reform. Grattan Institute. https://grattan.edu.au/wp-content/uploads/2016/04/872-Hot-Property.pdf

Keep reading