The Nightly Rate

How homelessness became an asset class, and why nobody holding the purse strings seems in any great hurry to end it

Picture a hotel room on the edge of a ring road. Two single beds pushed together, a kettle, a microwave balanced on top of the minibar, and a family of four living out of holdalls because the wardrobe was designed for a weekend, not a life. The children do their homework cross-legged on the duvet. The mother keeps every letter in a carrier bag by the door, because she has learned that the one telling you to move on tends to arrive with less notice than a dental appointment. Nobody in that room is on holiday, and yet the room is billed by the night, exactly as though they were.

Now pull back from the room to the ledger, because that’s where the real story lives. Councils in England spent a record £2.9bn on temporary accommodation last year, while 135,580 homeless households, including 177,530 children, were living in it. That spending has climbed 88% since 2020-21, when it stood at £1.56bn, and English councils have poured £13bn into temporary accommodation over that period. In London alone, boroughs were spending almost £5 million a day on it in 2024/25. Five million pounds a day. That isn’t a safety net; it’s a market, and markets have winners.

I spent years after the RAF working in private security management, close enough to outsourced contracts to know how they’re written and, more to the point, how they’re read by the people who bid for them. Nobody in that world twirls a moustache. What they do is read the specification, price the risk, and deliver exactly what’s measured. If the contract pays for beds occupied, you get beds occupied. If nobody pays for the problem going away, the problem doesn’t go away, and it would be naive to expect anyone to volunteer the fix for free. That’s the uncomfortable thread running through everything that follows. This isn’t a story about villains so much as a story about a system that pays handsomely for the crisis and almost nothing for its end.

A legal duty with no houses attached

The British version starts with something genuinely decent. Councils have a legal duty to house homeless families who qualify for help. The trouble is that a duty without housing stock is just an invoice waiting to happen. Analysis by Crisis found that between April and October 2024, fewer than three in every hundred private rented properties in England were affordable for people relying on Universal Credit or Housing Benefit. So the council does what it must, and rents whatever it can find, from whoever has it, at whatever price they name.

Then the maths turns quietly vicious. Councils have to pay landlords at current market rates, but their reimbursement from central government is still linked to 2011 rental levels. The gap lands on local budgets, and the councils’ own share of temporary accommodation costs has risen from 7.1% in 2009/10 to 50.6% in 2024/25. The consequences show up in odd places. Hastings Borough Council, for instance, spent £6.8 million, nearly 40% of its entire £18 million budget for 2024/25, on temporary accommodation alone. That’s money that would otherwise have gone on libraries, parks and potholes, which is to say the ordinary public realm that everyone else uses and quietly wonders why it’s crumbling.

The most expensive slice is the nightly-paid stuff. In 2024/25, councils spent £1.08 billion on nightly paid, privately managed self-contained accommodation. Nightly rates are how hotels price rooms for people passing through. We are using them for children who will do most of their growing up in the space between one move and the next.

Selling the family silver, then renting it back

If you want a single image that captures the whole thing, here it is. For decades, councils were obliged to sell homes to sitting tenants at a discount under Right to Buy, and were then left without the housing they now legally need. So they rent the same homes back. A 2019 London Assembly report found at least 2,333 former Right to Buy homes being rented by London councils, with Newham alone renting back 808 of them at a cost of £12.9 million a year.

It gets better, by which I mean worse. Some councils have started buying the homes back outright. Newham, which has England’s longest social housing waiting list, sold 130 properties for £10.6 million and then bought them back for £33.3 million, losing roughly £22 million on homes it had owned only a few years earlier. Across the country, a Big Issue investigation found that 53 councils had sold 20,836 homes under Right to Buy in five years, earning £2.25bn, while buying back 8,590 properties at a cost of £2.12bn. Put simply, they spent over £2bn on house purchases and still ended up with 12,246 fewer homes. And the grim punchline is that the councils say it’s still the sensible option. They argue that buying homes back, even at a loss, beats paying private landlords for temporary accommodation. When haemorrhaging money counts as the prudent choice, you’re not looking at a policy failure any more. You’re looking at a business model, just not one the public owns.

A different duty, the same logic

Here I want to be careful, because this is where most of the online commentary goes wrong. Asylum accommodation is a separate system with a separate legal duty, run by the Home Office rather than councils, and it houses a different group of people. Lumping it in with family homelessness is sloppy. But set the two side by side and the machinery looks remarkably familiar.

In 2019 the Home Office handed its asylum accommodation contracts to three suppliers, Clearsprings, Serco and Mears, who split the country between them. Since then, the projected cost of those contracts for 2019 to 2029 has more than tripled, from £4.5bn to £15.3bn, according to the Home Affairs Committee. The per-night figures tell you exactly where the money goes. A night in an asylum hotel averages £144.98, against £23.25 in dispersal accommodation such as shared houses.

Clearsprings is the one that makes the eyebrows climb. It estimated that the value of its ten-year contract for the south of England had grown tenfold since 2019 to £7 billion, and its profits tripled in just two years to £91 million in 2024. In fairness, the contracts do contain a profit cap, and the published margins aren’t outlandish on paper. The National Audit Office recorded operating margins of 2.8% for Serco, 4.6% for Mears and 6.7% for Clearsprings, though those figures include other business. But a modest margin on a tenfold contract is still a fortune, and the cap only works if someone collects. The Home Affairs Committee noted that Mears was expected to repay around £13.8 million and Clearsprings about £32 million, yet both said they hadn’t yet handed the money back. The committee’s own verdict was blunt. It called the Home Office’s handling unacceptable incompetence.

I’d gently point out that “incompetence” is doing some heavy lifting there. Incompetence implies accident. When the same outcome keeps arriving on schedule, year after year, across different governments, it starts to look less like a mistake and more like a design nobody wants to redraw.

The respectable face

Hotels and nightly rates are the crude end of this. The sophisticated end wears a nicer suit, and it’s called the Social Impact Bond. The idea is elegant. Private investors put up money for a social programme, and if it hits agreed targets, the government pays them back with a return. Britain pioneered the model and ran several on homelessness. The Fair Chance Fund, for example, combined £10 million from the housing ministry with £5 million from the Cabinet Office to fund seven social impact bonds aimed at young homeless people. It ran entirely on payment by results, with social investors funding the providers. The model travelled well, too. In Australia they’re called Social Benefit Bonds, in the US Pay For Success schemes.

I’ll be honest, this is the part of the story where I’m least interested in easy outrage. Many of the investors aren’t sharks at all. A scoping review of homelessness bonds found that the most common investors were charities and non-profits, followed by social merchant banks and impact investment firms. Some of these projects helped real people. The problem isn’t that everyone involved is cynical. The problem is what the structure teaches us to believe. Once a young person’s route out of homelessness is expressed as an outcome metric that triggers a coupon payment, human hardship has been converted into something with a yield. And the accounting isn’t always easy to check. The same review concluded that homelessness SIB projects lack transparency, which is a polite academic way of saying we’re not entirely sure what we bought.

The view from my cave

I write this from a cave in rural Spain, which gives me a slightly unusual vantage point, and Spain is where the whole pipeline becomes visible from end to end. After the 2008 crash, Spanish banks were rescued with public money and left holding mountains of repossessed homes. Those banks and the state’s “bad bank”, SAREB, then sold large chunks of their portfolios to international equity funds, helped along by laws giving very favourable tax treatment to SOCIMIs, Spain’s version of the REIT. In other words, the public rescued the banks, and the banks sold the rescued homes to the funds.

The results are now part of the landscape. Blackstone holds around 19,600 rental homes through 27 subsidiaries, 13,000 of them in the Madrid region, where the Civio foundation describes it as the largest private landlord in the city. Prime Minister Pedro Sánchez has publicly blamed speculators and so-called vulture funds for distorting the market and turning housing into a financial asset. When Catalonia tried rent caps, the funds didn’t hand the homes back to the people who needed them. They sold them into the retail market for a higher margin instead, which is the market working exactly as advertised and the housing crisis carrying on regardless.

None of this is a secret. Back in 2019, the UN’s housing rapporteur, the Canadian lawyer Leilani Farha, wrote formally to Blackstone and to several governments. She and the UN working group on business and human rights raised serious concerns that the firm’s conduct was inconsistent with international human rights law, and wrote to the Czech Republic, Denmark, Ireland, Spain, Sweden and the United States, noting that each had enabled the financialisation of housing through preferential tax laws and weak tenant protections. Their core point was that housing, unlike gold, isn’t a commodity but a human right. Blackstone replied that it had improved the supply of well-managed rental housing through private capital. Both statements can be true at once, and that’s rather the problem. A home can be well managed and still priced beyond the reach of the family who used to live in it.

Where the logic ends up

If you want to see what happens when this approach runs its full course, look west. In the United States, the argument has moved beyond who profits from homeless people to what can be done to them. In the 2024 case Grants Pass v. Johnson, the Supreme Court ruled 6-3 that an Oregon city could effectively ban homeless people from camping on all public property, meaning cities may punish unhoused residents for sleeping outside even when they have nowhere else to go. Breaching the ban can ultimately bring up to 90 days in jail and further fines. The Court reasoned that homelessness was a complex issue better left to policymakers than federal judges.

Hold that next to everything above and a pattern appears. Where the state has sold, outsourced or never built the housing, it can pay a premium to rent the solution back; it can issue bonds that let investors profit from partial fixes, or, failing all that, it can simply fine the people it hasn’t housed. Every option is available except the boring one, which is owning enough homes to make the crisis stop.

The question nobody wants on the agenda

Here’s what keeps nagging at me. A problem that costs money gets solved, eventually, because somebody gets tired of paying. A problem that makes money gets managed, because nobody profiting from it has any incentive to see it end, and the people paying for it are spread so thinly across council tax bills and Home Office budgets that none of us feels the full weight of it. Homelessness in Britain, and increasingly across the developed world, has quietly crossed from the first category into the second. It has become recurring revenue.

And assets are not solved. Assets are managed for yield. They’re refinanced, repackaged and sold on, and their value depends on demand holding steady. I don’t think anyone sat in a boardroom and decided that 177,000 children should grow up in rooms billed by the night. I think something worse happened. We built a system in which nobody needed to decide it, and then we lost the ability to decide otherwise.

Somewhere tonight, in a room by a ring road, a kettle is boiling next to a microwave on a minibar. Someone is being paid for that room, and paid well. I just can’t find the line in any contract that pays anyone to make it empty.

Until Next Time