What’s really happening to the size of the private rented sector?

Importantly, because of the way data for these official stock and household estimates is collected, figures from these sources will lag behind current events. The most recent data run to 2024/25, meaning that if the sector began shrinking in 2025 or 2026, it won’t show in these official estimates until the 2027 or 2028 releases. But there are other sources of data we can look to, both for corroboration of the trends from these official government datasets, and for an indication of whether the story might have changed in the most recent period.

One of these sources is tenancy deposit data, which corroborates the stock and household trends from official government estimates to 2024/25. The number and value of protected deposits across England and Wales also reached an all-time high by March 2025: 4.7 million deposits of £5.5 billion, up 570,000 deposits (£1.2 billion) since 2020 (TDS, 2025). These figures aren’t a direct measure of the sector size —  legacy and non-compliant tenancies sit outside the deposit protection schemes, and a minority of landlords don’t collect deposits —  so the total reflects the stock of existing deposits, rather than the full stock of tenancies.

But as turnover continues and a growing share of tenancies flow into protection schemes over time, this data should become a more reliable proxy for the sector’s size, with any future fall in deposit numbers a plausible signal of shrinkage. For now, both the number and value of protected deposits continuing to climb is at least consistent with a sector that is at least holding steady, not contracting.

The latest HMRC property rental income statistics (2026a) show a small decline in the numbers of individual (unincorporated) landlords from 2.91 million in 2023/24 to 2.88 million by 2024/25 (a reduction of 30,000), and a slight reduction in the property income they declared from £59 billion to £58.99 billion. However, accounting for rental inflation over the period, this implies a decrease in the number of homes owned by individual unincorporated landlords from 4.1 million to 3.8 million.¹

Mortgage lending (MLAR) data also offer a useful, if somewhat narrower, lens on the issue, covering an estimated 40% of privately-rented homes financed via a mortgage —  mostly buy-to-let (BTL) mortgages —  the remaining 60% are owned outright or financed through other means, including by institutional landlords (Mortgage Works, 2024).

New BTL advances more than halved from £41 billon in 2022 to £18 billion in 2023 as the interest-rate shock hit, though have since recovered somewhat to £25 billion in the year to Q1 2026 (FCA, 2026). The aggregate value of outstanding BTL mortgages, meanwhile, has declined by 7%, or by £17.6 billion in value, since its peak in 2022, yet remains higher than at any point prior to 2020.

A lower stock of BTL mortgage debt since 2022, fewer individual unincorporated landlords, and lower rental property income for this group doesn’t necessarily mean a falling stock of private rented homes though, for several reasons.

First, the same aggregate balance of BTL mortgage debt doesn’t tell us about composition. The same balance of outstanding BTL mortgage debt could reflect a larger number of cheaper homes, fewer more expensive ones, or more homes purchased with less borrowing, so a falling total doesn’t map neatly onto a falling number of homes.

Second, and more importantly, a large and growing share of landlord activity is moving into limited-company structures, which aren’t captured in these figures. Increasing numbers of landlords are incorporating each year in response to tax changes in the late 2010s, and the incentives to incorporate are even stronger post interest-rate shock (Elliott and Baxter, 2025). New letting-company incorporations reached 67,000 in 2025, up from around 50,000 new incorporations in 2022 and 2023, and more than 6 times the 10,000 new incorporations in 2015 (Hamptons, 2026a).

This matters for interpreting both MLAR and HMRC property income statistics. MLAR statistics covers loans to individuals, so increased lending to incorporated landlords would not be captured in these figures; and the number of new lettings company incorporations in 2025 was more than double the net reduction in unincorporated landlords reporting rental income by 2024/25, demonstrating that the decline in 1 segment of the market was more than offset by expansion in another.

Third, BTL mortgage data also says nothing about the estimated 60% of the sector financed some other way (Mortgage Works, 2024). The English Private Landlord Survey 2024 shows that, among landlords adding to their portfolios, 56% financed their latest acquisition with a BTL mortgage, while 32% used no borrowing at all, meaning a substantial share of portfolio growth is happening outside of this mortgage lending data.

Taken together, the mortgage lending figures and the HMRC data may describe a genuine slowdown in one financing route for a segment of the sector, but are not, on their own, evidence that the overall stock of private rentals is shrinking.

Headline exit and landlord transaction statistics

Another way the argument that the PRS is shrinking is often substantiated is using analyses of housing transactions and rental listings. The problem with this is that these data describe the churn of ownership of homes or tenancies, not a net loss of homes, despite often being described as such. For years these data have been used to make the case that the PRS was contracting, while official data to 2025 shows the sector continued to grow slowly.

A widely cited figure in recent reporting is that 834,800 landlords (or properties, the terms are incorrectly used interchangeably), 18.6% of PRS stock, have ‘gone’ since 2016 (TwentyCi, 2026a). The method underpinning this figure tracks rental listings since 2016 and identifies which were later sold without returning to the market as rentals — a real measure of exits, but only on one side of the ledger. It says nothing about entries, and so cannot on its own tell us about the net change in sector size.

TwentyCi’s (2026b) transaction analysis shows a related trend from a different angle; the number of homes coming onto the sales market has risen from an average of around 390,000 a year in the early 2020s to over 440,000 by 2025, while the share of these that were formerly rented has climbed from 8-10% (Q1 2023-Q1 2024) to 12-16% (Q2 2024-Q1 2025).

Our own analysis of Energy Performance Certificate (EPC) data (MHCLG, 2026b) corroborates this direction of travel; the number of private rentals being evaluated for an EPC for sale, rather than relet or another reason, climbed from an average of 15,000 or 5% of EPC issuances for private rentals between 2016-2024, to 26,000 or 9% of EPC issuances in 2025 (though the share fell back to 4.4% in current data for 2026).

Likewise, Capital Gains Tax statistics (HMRC, 2026b) show an uptick in the number of residential properties being sold with capital gains liability (likely predominantly private rentals, although also including other residential property such as second homes), climbing from 165,000 in 2022/23 to 195,000 in 2024/25, before easing back to 173,000 in 2025/26. Again, however, these one-sided measures tell a story of rising churn, not a shrinking stock of homes, unless the outflow is genuinely outpacing the inflow.

A more robust method for tracking the changes in the size of the sector is to track sales from landlords to owner occupiers, and vice versa. Savills have published several pieces using this approach; one concludes that England’s private rented sector has contracted by 4.2% (200,000 homes) between October 2022 to November 2025 (Real Estate: UK, 2026), another concludes it contracted 6% (290,000 homes) between April 2021 and October 2024 (Savills, 2025).

A timeseries from Hamptons tells a similar story: the share of transacted homes that were sold by landlords exceeded those purchased by landlords every year between 2016 and 2025. Notably, though, that flipped in the most recent data: landlord purchases exceeded landlord sales in June 2026 for the first time since 2019, accounting for 10% of purchases against 9.2% of sale listings (Hamptons, 2024; Hamptons, 2026b).

Even this more robust method has structural gaps: they miss routes into the sector that don’t involve a sale, and even when a sale takes place it’s not always possible to be certain where the property is flowing to. This could be a significant gap. The English Private Landlord Survey (2024b) showed that just over half (52%) of landlords’ first rental properties were purpose-bought to be let out; but 38% converted a home they originally bought to live in themselves, and a further 7% inherited or were gifted a property that became their first rental.

EPC data corroborates this: EPCs issued for owner-occupied homes converted to private rentals averaged 18,000 annually between 2015 to 2017, but more than doubled to 38,000 annually between 2023 and 2025, from around 2% to 5% of EPC issuances for owner occupied homes (note again, this isn’t a stock measure: new EPCs are only required in certain circumstances, but could provide an instructive indicator of the direction of travel).

Inflows of inherited homes to the private-rented sector may also continue to grow as the share of inheritances that contain a property climbs; 80% of inheritances contained property in 2021/22, up from around 70% in the early-to-mid 2010s (Hamptons, 2024b).

A growing build-to-rent sector also contributes to the private rental stock without being captured in transactions data. The build-to-rent sector remains a relatively small share of the overall private rental market, but its growing in size with 157,000 built as of Q2 2026 (Real Estate: UK, 2026) and a further 154,000 in the pipeline for development, meaning increased inflow of new private rental properties.

In many of the years prior to 2025 analysis of these same data sources in industry reports concluded that the PRS was shrinking (such as Propertymark, 2022; Savills, 2023) while the official statistics showed it continuing to grow. The discrepancy between these 2 data series could be partially explained by the incomplete picture given by the data which industry reports focus on — because it misses conversions and new builds.

Reduced rental listings data

Declining volumes of new rental listings, identified by Zoopla and Rightmove, are also often read as evidence of shrinking supply. Rightmove’s Q2 2026 report found the number of homes to let 1% lower than a year earlier, while Zoopla’s June 2026 report found new rental listings still 20-30% below pre-COVID-19 levels.

However, volumes of new listings are a product of both rental market size and the rate of tenancy turnover. Average tenancy length in the PRS has increased from 3.5 years in 2013/14 to 4.7 years today (EHS, 2025b), meaning tenancy turnover is falling; consistent with an aging and older, more security-seeking mix of renters (Aldrige, 2026).

Converting turnover rates and stock sizes into an ‘implied new-listings rate’ makes the scale of this explicit: even as the number of private-rented households increased by 8% between 2013/14 to 2024/25 (400,000 more households), falling turnover would mean 18% fewer new home listings each year (240,000 fewer listings) even as the sector grew. Falling new listings, therefore, are what we’d expect to see from a stable-to-growing sector with longer tenancies, not a shrinking one.

Other recent data shows listings moving in the opposite direction: private rental supply (flow of properties onto the market for let) increased by 17% in the year to May-June 2026, and was at its highest for 7 years (TwentyCi, 2026a). A July 2026 Savills report (2026) offers a plausible mechanism: weak sales-market conditions have led some landlords who tested a sale to return stock to the rental market instead, and most Savills’ agents expect stock levels to increase over the next 3 months.

The demand side of the story, meanwhile, tells a consistent tale of a market normalizing after a genuine post-COVID-19 surge, not a supply collapse. Zoopla’s enquiries-per-listing measure has fallen from a 2022 peak of 15.5 to 5.6 by May 2026 (against a 2017-19 average of 3-3.7) (Zoopla, 2026). Tenancy Deposit Scheme (TDS) survey data shows the average number of properties viewed before securing a tenancy has fallen from a high of 3.75 in 2024 to 3.36 by March 2026 (TDS, 2026).

As noted above, build-to-rent (BTR) represents another structural shift in how the sector is supplied: 157,000 BTR homes have been completed by Q2 2026, with a further 153,000 in the pipeline (Real Estate: UK, 2026).

This is consistent with, and extends, our analysis from 2024 (Grayston et al., 2024), which found that while the evidence at the time didn’t support a large-scale exodus, some parts of the sector (specifically smaller-scale, BTL-mortgage dependent landlords) did appear to be contracting, even as the sector overall stagnated and consolidated rather than reduced in size. The pattern in this update looks the same: potentially some shrinking in the BTL-mortgage-financed segment, offset by incorporation, BTR growth, and non-mortgage entry, producing a sector that is larger and with more concentrated ownership than it was, not smaller.

Tenant side: a long-term increase in unmet demand and a genuine undersupply problem

Despite a lack of evidence that the private rented sector is shrinking, there is clear evidence that there is, and has been for a prolonged period, an insufficient number of homes overall available for residential use relative to housing need. The latest English Housing Survey (2026) data show around 1.6 million concealed households in England (households with an additional adult(s) who would like to own or rent their own home but can’t afford to do so) a figure that has held broadly steady as a share of households since 2012/13. A third of these concealed households (33% or 527,000) want to rent but can’t afford to; these are skewed towards those in higher-income households, and 88% are aged 16-34.

Source: English Housing Survey, 2026

For these households, renting privately becomes the default housing option, creating increased ‘demand’ in a narrow economic sense for privately rented homes, and this demand is often interpreted as evidence that the private rented sector specifically needs to grow. But that doesn’t follow. The same need could be met by growing the housing ‘pie’ overall: through new-build homes of different tenures and bringing homes back into residential use (for example, away from short-term letting), rather than by expanding the PRS itself.

If owner-occupation and social renting grew fast enough, absorbing unmet housing demand and outflow from private rentals, then demand for private rentals would be reduced even as the sector’s growth slowed or contracted — the resulting housing system would likely do a much better job of meeting the genuine housing needs of the population.

There’s strong evidence that renters themselves would welcome this exact outcome. The TDS (2026) tenant survey finds that three quarters of private renters (76%) would like to buy their own home, and 30% would prefer to live in social housing, rising to half of pension-age renters (49%), and higher among those receiving benefits (47%) or those with children in the household (39%).

A survey commissioned for the National Residential Landlords Association (Chamberlain Walker, 2023), found just 29% of private renters wanted to continue to be renters in 12 months’ time; 12% wanted to buy in the next year and thought they could, while 45% wanted to but didn’t think they are in the position to yet. For most private renters, they view the sector either as a staging post or as the housing option of last resort rather than a tenure of choice, which cuts against reading ‘more private rented homes’ as an unambiguously good outcome we should be seeking, separate from whether the sector is shrinking.

Unmet housing demand and affordability pressures are real and pressing policy issues, the appropriate solutions to which require consideration of what kind of housing system, and tenure mix, will best serve peoples’ housing needs (Baxter et al., 2022). The claim that the private rented sector is shrinking is a separate, empirically distinct claim and, on the evidence collated here, one that isn’t currently supported.