We Built Too Many HMOs. What Happens to Them Now?

Something has shifted in our office over the last twelve months and I don't think we're imagining it. The conversations landlords are having with us aren't the conversations they used to have. For the better part of a decade the question was always where to put the next investment, what street had the best yields, whether such-and-such conversion would clear the planning hurdle. Now it's different. More of them are asking how to get out.

For about fifteen years HMO conversion has been one of the dominant strategies in the private rented sector. Not the only one. Rent-to-rent ran alongside it for a while, Airbnb pulled stock out of long-term lets in the city centre, and there are other variations besides, but HMOs were the headline play. The reason was the obvious one. A three-bed family terrace in Lenton or Beeston might let for around £900 a month. Convert that same house to six en-suite bedrooms and the gross rent climbs to roughly £3,000. You don't need an MBA to see why investors built the model they built. Capital flows where the yields are, and the family let, the long-term single-household tenancy that used to be the spine of the sector, got quietly priced out of the conversation by its own product class. Why house a family for £900 when your neighbour is taking £500 from one room?

What's now coming onto the market are the houses those conversions produced. Five, six, seven bedrooms apiece. Fitted kitchens, en-suite doors off a beige hallway, a licence number quoted in the listing description. We're seeing them sit. We're seeing prices reduced. We're seeing landlords come to us asking what their realistic options are, and we don't always have a good answer for them.

This is going to be a long piece because the question deserves it. It's going to get into the legislation that built the market, the demand that's gone soft on it, the regulatory wave that lands next May, and what I think happens to the stock over the next few years. The short version is that nobody actually knows yet. But the question is becoming impossible to ignore, and somebody needs to start asking it out loud.

How we got here

The HMO market we have today wasn't built by accident. It was built by two waves of regulation layered on top of each other, and it's worth setting them out clearly because the layers explain a lot about why the stock looks the way it does and why it's now stuck.

The first wave was licensing. The Housing Act 2004 brought in mandatory HMO licensing nationally, taking effect April 2006. As originally drafted it only applied to large HMOs, meaning three or more storeys with five or more occupants from two or more households. The same Act gave councils two further powers. They could extend licensing to smaller HMOs through additional licensing schemes. They could also apply selective licensing, which covers all privately rented properties and not just HMOs, in areas where the council judged housing demand or anti-social behaviour required it. Nottingham used all three. The city has had additional licensing for smaller HMOs for years, and from 2018 it brought in one of the largest selective licensing schemes in the country, covering most of the city. Then in October 2018 the mandatory regime got expanded again. The three-storey threshold was scrapped, so any HMO with five or more occupants from two or more households now needs a mandatory licence regardless of building height. That single change pulled a large slice of existing stock into the licensing net.

Each layer brought its own application fee, its own room-size standard, its own inspection cycle, its own paperwork. The cost stack added up.

The second wave was planning. In April 2010 the Town and Country Planning (Use Classes) (Amendment) Order created a new use class, C4, for small HMOs of three to six unrelated occupants sharing a household. Labour's intention had been that converting a family home (C3) to a small HMO (C4) would require planning permission. That intention lasted six months. In October 2010 the incoming Coalition government amended the General Permitted Development Order to make C3-to-C4 conversion permitted development, meaning no planning permission was needed unless a local authority specifically removed the right.

The mechanism for removing it is the Article 4 Direction. Nottingham City Council brought one in across the entire city boundary on 11 March 2012. From that point, anyone in the city wanting to convert a family home to a small HMO had to apply for planning permission. Broxtowe, Beeston's local authority just across the city boundary, held off for a decade. Their Article 4 didn't arrive until 26 March 2022, and even then only across parts of Beeston and Beeston Rylands rather than the whole borough.

That ten-year gap is, more or less, the gap that built modern east Beeston.

What's important to be clear about is that none of this regulation made small HMOs illegal. Three unrelated people sharing a house in Nottingham is still a perfectly lawful C4 use today. The effect of the regulation was to push the model in a particular direction. Once you were carrying licensing fees on top of additional licensing fees on top of selective licensing fees, plus mandatory inspections, planning permission costs inside Article 4 areas, room-size compliance, fire-safety upgrades and proper professional management, you couldn't run a small casually-tenanted shared house and make the numbers work. The economics pushed everyone the same way. Bigger properties. More rooms per property. Higher rent per room. Professional management.

The informal version, three friends renting a 1930s terrace between them at a friendly rent with no licence and no en-suites, didn't disappear because the law banned it. It disappeared because nobody could be bothered to run it commercially under the new cost structure, and the casual landlords who used to run it without much thought weren't going to start now. (Lodgers are a different question. A homeowner with one or two paying guests inside their own household isn't an HMO and never has been. That arrangement is still common. What disappeared was the small unprofessional shared house in the middle.)

The professionalisation that followed was the precondition for the boom. HMOs became a financeable, scalable, predictable asset class. You could buy one, leverage against it, run it like a small business. The yields were strong, two big universities sat down the road, and inside an Article 4 boundary the existing C4 stock had a sort of de facto scarcity protection. The maths worked.

What happened next was predictable. Investors moved in. Family houses with three or four bedrooms got bought, gutted, and reconfigured into five, six, seven en-suite rooms. In Beeston, on the east side particularly, the rush before the 2022 Article 4 came in was something to watch. People were buying houses in a panic. The Article 4 implementation got delayed and the rules came in poorly drafted, which only stretched the window further. Investors were certain they were going to miss the golden goose if they didn't get in.

Some of them got in. Plenty of them are now sitting on the result.

Three things broke at once

Student demand for on-street HMOs has been weakening for a while, and several different things broke at roughly the same time. They're compounding each other in ways that I don't think were planned for.

The first thing is that the council changed direction. Back in 2022 Nottingham City Council was openly warning of a 5,500-bed student accommodation shortage and pointing to what it described as the largest PBSA development pipeline outside London. Council planning policy explicitly aimed to draw students out of on-street HMOs and into purpose-built blocks. That was the framework. Skip forward to the council's June 2025 student accommodation update and you can read the result in their own numbers. PBSA delivery is now exceeding the growth in student numbers. On-street student-occupied households are falling year-on-year, by roughly 2%, then 2%, then 4% across recent reporting cycles. The city has around 7,000 PBSA bedspaces in the pipeline with planning permission or likely to receive it. They've solved the bed shortage. The honest reading is that they've overshot it.

The second thing is that the international student tap got turned down. From 1 January 2024 the government restricted international students from bringing dependants except on PhDs and government-funded scholarships, and closed off the route from a student visa to a skilled worker visa before course completion. The numbers shifted immediately. Year-on-year international student visa applications fell 16%, from 428,100 to 359,600. Dependant applications collapsed by around 85%. International students don't only live in PBSA. Plenty lived in HMOs, often paying a year's rent up front in lieu of a UK guarantor. That route is closing too. From 1 May 2026 no landlord can demand more than one month's rent in advance, full stop, regardless of whether the tenant is offering it.

The third thing is that both Nottingham universities are visibly in trouble. The University of Nottingham posted a group deficit of £76.8 million for 2024/25. They're selling King's Meadow Campus. They're selling Castle Meadow Campus, which they bought for £37.5m in 2021, refurbished for over £40m, and which is now valued at somewhere between £14.8m and £18m. In November 2025 they announced the suspension of 46 courses for new students, including all modern languages and music, making them the first Russell Group university to drop modern languages entirely. NTU posted a smaller £2m deficit on income that fell 8.2%, having already paid £9 million to the 230 staff who took voluntary redundancy. Both universities cite declining student numbers, particularly international, as the cause.

You don't need a forecasting model to see what this means for HMO demand. The student pipeline that justified the conversion frenzy of the 2010s is shrinking. The international cohort that propped up rents at the top end is being squeezed. And the PBSA stock that students are increasingly being pushed towards is now, on the council's own numbers, in surplus, with another wave still arriving, including the 419-studio Cassidy Group scheme on Station Road in Beeston due for the 2026/27 academic year, on top of their other Nottingham developments delivering more than a thousand beds combined.

The repurposing problem

The thing nobody quite says out loud, but everyone in the industry knows, is that a seven-bed HMO converted from a three-bed family house is a deeply specific product. It isn't a family home anymore. It isn't really anything else either.

Take a typical case. A 1930s semi in Lenton, originally three bedrooms and a box room, sold to an investor in 2014 for around £160k. By 2017 it's six en-suite bedrooms, a small kitchen-diner, parking for one car, half the garden gone to a side extension. To put that back to a family layout you're ripping out five bathrooms, knocking down stud walls, restoring an original layout, refitting heating and lighting, and once you're done you've got a three-bedroom house in a street where most of the neighbours are still HMOs and the family-buyer demographic has been priced out for the better part of two decades. The market doesn't pay for what something used to be. It pays for what it is.

Could the stock become professional rental? Maybe. Some operators are pitching themselves as co-living providers and trying to reach a young-professional demographic. The pool is narrower than the marketing suggests though. Two friends will share a flat. Far fewer people want to live with five strangers once they're past thirty, and we see that in the lettings book. Could it move to social rental? Theoretically, but the configuration of multiple en-suites, no proper communal living, fragmented kitchen layouts and no real outdoor space isn't what social tenants need.

There's a human side to this that agents sometimes get accused of ignoring, and it's worth saying out loud. HMOs at their worst, particularly with older tenants in their thirties and forties who've ended up there because nothing else is affordable, aren't great places to live. They're transient. They're not conducive to mental health. People rarely stay long. Anyone who's managed HMOs for any length of time has seen tenants at the wrong end of life ending up in a six-bed share because the alternative was nothing. If the next phase of this market involves a glut of cheaper, harder-to-let HMOs drifting into housing the people they're not really designed for, that's a public policy problem nobody currently has a plan for.

The Renters' Rights Act and what it actually does to HMOs

On 1 May 2026 the Renters' Rights Act 2025 comes into force. There's been a lot written about it elsewhere, so I'll keep this part to what specifically lands on HMO landlords.

Section 21 evictions are abolished. Fixed-term assured tenancies become assured periodic tenancies, with no end date and termination only on specific grounds. Rent in advance is capped at one month. The traditional pre-Christmas pre-let cycle for the following September gets disrupted, because the new Ground 4A possession route, which is what allows landlords to recover possession at the end of an academic year, only applies to tenancies granted no more than six months before the start date.

Code-registered PBSA, meaning members of the ANUK or Unipol national codes, is exempt from the assured periodic regime entirely. Private HMOs are not.

In practice that means from May next year, a student in an HMO can give two months' notice and leave at any point, including mid-academic-year, or for the summer. The eleven-month income guarantee that has underpinned a lot of HMO yields just disappears. PBSA keeps its fixed-term flexibility provided the operator is code-registered. HMO landlords have to price in void risk, which means rents either rise or yields fall.

Read at first glance this looks like it's bad for HMO tenants. It's actually rather the opposite. A cost-conscious student who doesn't want to be locked into an eleven-month PBSA contract can move into an HMO and end the term early to save on summer rent, which wasn't available before. The question for the HMO landlord is whether the loss of guaranteed income is offset by the demand bump from offering a more flexible product. The honest answer is it depends on the property, the area, and what PBSA does with its pricing.

What's clearer is that as the price gap between PBSA and HMOs narrows, because HMO landlords have to price in voids and institutional product is becoming more competitive, the structural case for paying a premium for PBSA gets stronger. Some private landlords will exit the student market entirely. Sector lawyers writing on the Act are converging on the same point. A two-tier student rental market is forming, and the bottom tier is the one with the converted family houses in it.

So what does the landlord actually do

In our office we're starting to see the shape of the realistic options. There isn't a clean one, and most of what we end up advising depends on the property, the location, and the landlord's broader portfolio.

You can sell to another HMO landlord, if you can find one. That buyer pool is shrinking and the prices being offered reflect it. Properties with planning permission, a current licence and a tenant book are sitting longer than they would have done two years ago. Reductions of 10 to 15% on the marginal stock aren't unusual.

You can sell back into the family market, theoretically, if the conversion was light-touch and the area still has genuine family demand. Most conversions weren't light-touch. Most areas don't.

You can convert to self-contained flats, if the planning will allow it. Reverting C4 to C3 family use is generally permitted. Converting to multiple flats requires planning permission and won't always get it, especially in streets where the council is now actively trying to rebalance toward family use. The conversion cost is also significant because you're effectively redoing the work.

You can hold and ride it out, which is defensible if you've got cash flow and the rental market doesn't deteriorate further. The Act changes the maths but a well-located, well-maintained HMO with a stable tenant base will keep working. The marginal stock, meaning the over-converted, badly maintained properties in over-saturated streets, is what worries me.

And you can walk away. The least-discussed option, but it's becoming a real one. Sell at whatever the market gives you, take the loss against the original conversion cost, move on. We're quietly seeing this conversation happen more often than I'd have expected this time last year.

Why this matters beyond landlords

Some readers will get this far and think this is a landlord problem. They built the bubble, let them eat the bubble. Fair enough. But the consequences don't stop at landlord balance sheets.

There are streets in this city, particularly in east Beeston, Lenton, parts of Radford and Dunkirk, that are 60-80% HMOs. If those streets start drifting into distressed stock, half-empty in summer, less well maintained over time, the spillover hits everyone living near them. Families who bought into mixed neighbourhoods that aren't really mixed any more. Owner-occupiers whose home values track their street comparables. Councils trying to rebalance through HMO-to-family policy that depends on family buyers actually existing in those streets in the first place.

There's also the wider economic picture. A city that produced over 60,000 student bedspaces' worth of housing infrastructure assumed a particular trajectory for higher education. If the universities continue contracting, with more course closures, more campus sales, an international intake that doesn't recover, that's not just a housing problem. It's a regional economic one. The two universities together contribute somewhere in the region of £3.8 billion to the local economy and support around 25,000 jobs depending on whose figures you take. Lose even a modest fraction of that and it ripples through the rental market, the retail base, the hospitality sector, and yes, the value of the houses sat on those streets.

There's a quieter question underneath all of this that I think deserves more honesty than it usually gets. Did we as a city, as agents, as landlords, as a council, build the right kind of housing for the people who actually need housing? The answer is plainly no. We built the housing the market wanted to build, in the framework the legislation allowed. Those aren't the same thing.

What I think happens next

I don't want to pretend I know exactly how this plays out. Anyone who tells you they do is probably selling something. But the broad shape looks roughly like this.

The lower end of the HMO market is going to take a real correction. The marginal properties, meaning the over-converted ones in over-saturated streets that haven't been kept up, sell at meaningful discounts to where they were valued in 2022, or they get repossessed. Distressed sales become more visible. We're already starting to see early signs.

The better-located, higher-spec HMO stock will probably survive, but at lower yields than its owners are used to. Three- and four-bed houses in mixed-demand areas, run well, tenanted carefully, will continue to work. They just won't make landlords rich the way they did. The asset class matures from an active investment play into something closer to traditional buy-to-let.

Alongside that, a slow patchwork conversion back to family or hybrid use begins. It will be expensive. It will be encouraged in some wards and obstructed in others. It will leave a permanent mark on streets that effectively had a generation of family ownership replaced with a generation of investor ownership, and not all those streets will recover their previous character. The council will keep promoting PBSA over on-street student housing. The Renters' Rights Act will keep nudging private landlords out of student lettings. The slow grind of cost and policy will finish what the demand picture has already started.

We built a lot of HMOs. The market is asking us, in a way it hasn't asked before, what we want them to be next. We don't have a clean answer yet, and we should probably be sceptical of anyone who tells you they do.