My idea to house homeless people in government owned buildings, sports halls, churches, religious buildings and community centres instead of paying landlords to house people in temporary accommodation.
Potentially tens or even hundreds of billions of pounds per year worldwide, but there isn’t a reliable global dataset that lets us calculate an exact figure.
For perspective, England alone spent £2.84 billion on temporary accommodation in 2024/25, with about 131,000 households in temporary accommodation at the end of March 2025. That works out very roughly at £21,700 per household per year of gross expenditure.
London boroughs alone have been spending around £4–5 million every day, with expensive nightly-paid privately supplied accommodation a major factor.
House of Commons Library +1
London Councils +1
Globally, the potential population is enormous. UN-Habitat reported at least 318 million homeless people in 2024, while its 2026 assessment says as many as 3.4 billion people lack secure, safe and adequate housing—although obviously most of those people are not currently being housed at public expense in hotels or temporary rentals.
UN-Habitat +1
A useful way to illustrate your proposal is therefore with scenarios rather than pretending we know the exact global bill:
People moved from expensive paid temporary accommodation
Average net saving per person/day
Annual saving
5 million
£15
£27bn
10 million
£20
£73bn
20 million
£20
£146bn
20 million
£30
£219bn
50 million
£20
£365bn
So a scheme operating on a genuinely worldwide scale could conceivably release something in the region of £50–£200+ billion annually under plausible large-scale assumptions. That is a scenario estimate, not a measured global saving.
Your underlying idea becomes considerably stronger if it isn’t simply “put homeless people in sports halls.” It could be a public/community emergency-accommodation network:
unused church buildings, village halls, community centres, sports facilities, former schools, government buildings and other suitable premises converted into small, professionally managed temporary residential centres. Governments would pay the real operating costs—staff, heating, washing facilities, food, security, cleaning and support—but avoid very high nightly commercial rents and hotel charges.
There is an important qualification. Large communal halls are generally not suitable as long-term housing, particularly for families, children, disabled people and people with significant support needs. Adequate housing means considerably more than having a roof.
The economically stronger model would consequently be short-term community accommodation → rapid assessment → permanent social/affordable housing, rather than replacing permanent homes with halls.
UN-Habitat
And there’s an interesting second-order effect: if, hypothetically, £100 billion a year were saved internationally and even half were redirected into permanent affordable housing at an average public capital requirement of £100,000 per dwelling, that would finance roughly 500,000 permanent homes every year. Over ten years, that’s around 5 million homes.
So the most compelling version of your proposal is not merely about saving money: stop transferring enormous emergency-housing budgets to expensive nightly accommodation providers, use community/public buildings for genuinely short emergencies, and recycle the savings into permanent housing.
If you’d like, I can next work this into a fully costed worldwide proposal, including estimates for the UK, USA, Europe and the rest of the world and calculate a 10-year saving.
A defensible worldwide model has to be presented as a scenario, because there is no harmonised global database of what governments currently pay landlords, hotels and temporary-accommodation providers. The OECD specifically warns that national homelessness-cost estimates use different definitions and methods and cannot simply be added together.
OECD +1
The scale of the underlying problem is nevertheless huge. UN-Habitat estimates that at least 318 million people worldwide are homeless, with 2.8 billion experiencing some form of housing inadequacy.
UN-Habitat +1
For a Community Emergency Housing Network, I would model only a fraction of those 318 million people—people currently receiving, or who could otherwise require, relatively expensive emergency accommodation. The halls would be used for short stays and would need partitions/private rooms where possible, beds, bathrooms, kitchens or catering, heating/cooling, cleaning, security, safeguarding and professional support.
Global scenario
People served at a time
Avoided commercial cost
Community-centre running cost
Net saving
Annual saving
10-year saving
Conservative
5m
£12/person/day
£6
£6
£11bn
£110bn
Central
10m
£20/person/day
£12
£44bn
£438bn
Ambitious
20m
£25/person/day
£10
£15
£110bn
£1.10tn
Those figures are illustrative assumptions, not claims about present worldwide expenditure. They show, however, how quickly the economics become significant.
England demonstrates why the idea merits investigation. In 2024/25, government expenditure on homelessness was about £3.7 billion, including £2.8 billion on temporary accommodation.
Spending on B&B and nightly-paid accommodation alone rose from £309 million to £620 million in a single year.
At 31 March 2025 there were 130,890 households in temporary accommodation.
London has been particularly expensive: borough homelessness expenditure reached around £5.5 million every day in 2024/25.
GOV.UK
GOV.UK
London Councils +1
So I would propose a system with four stages:
Emergency accommodation: suitable sports halls, church/religious buildings, community centres, village halls, unused public buildings and similar facilities become certified emergency centres. This would ordinarily mean days or weeks, not years.
Assessment: every person or family receives an individual housing assessment, benefits/income assistance and whatever health or social support they require.
Rapid rehousing: people move as rapidly as possible into self-contained permanent or transitional housing.
Savings fund: a legally defined proportion of money saved on commercial temporary accommodation goes into acquiring and constructing permanent social and affordable housing.
The fourth part is particularly important. Suppose the central scenario produced approximately £44 billion net savings each year. If just 60% were reinvested, that would create a worldwide housing investment fund of roughly £26.3 billion every year, or £263 billion over ten years.
At an illustrative average public capital contribution of £75,000 per permanent dwelling, £263 billion could support about 3.5 million permanent homes. At £100,000 per dwelling it could support about 2.6 million homes. These shouldn’t be interpreted as predicted construction numbers—land and building costs vary enormously between London, New York, India, Africa and elsewhere—but they illustrate the virtuous circle:
Expensive emergency accommodation → cheaper community emergency accommodation → savings → permanent housing → fewer people requiring emergency accommodation → further savings.
There is also evidence supporting the wider economic principle. The OECD finds strong evidence for shifting away from prolonged emergency accommodation toward Housing First and housing-led approaches. It reports that US Housing First programmes have generated around $1.44 in savings for every $1 invested, while a French programme found its approximately €14,000 annual per-person cost was completely offset by savings elsewhere, particularly reduced hospitalisation and accommodation costs.
OECD +1
So I would modify your original idea slightly: community buildings should be the inexpensive emergency bridge, not the final destination. A person shouldn’t spend two years living on a camp bed in a sports hall simply because it is cheaper for government. The financial objective should be to make the emergency stage as short and inexpensive as practicable and put the resulting money into proper permanent homes.
On that basis, I think a credible headline for the proposal would be:
A worldwide Community Housing Programme could potentially redirect hundreds of billions of pounds away from expensive emergency accommodation over ten years, while using those savings to create millions of permanent homes.
The £438 billion central 10-year scenario is a reasonable figure to use for discussion provided it is very clearly labelled a modelled scenario rather than an established global saving. The upper scenario passes £1 trillion over ten years, but I would not use that as the headline until country-by-country data had been assembled.
A particularly compelling next step would be to build this into a “Solving the World’s Problems – Global Community Housing Plan”, with a proper country-by-country financial model, implementation rules and a proposal suitable for sending to governments, the UN and charities
