Is the Private Rented Sector Dying? A New Study Examines the Evidence

5 Oct

The many changes in the PRS in recent years have led some to claim that landlords are leaving in droves and that the sector is dying. With this in mind a well known charity has put together a briefing examining some of the data on the subject and which tries to come to a definitive conclusion. We will review their findings here.

Who conducted the study

The Joseph Rowntree Foundation (or JRF) describes itself as an independent social change organisation. To quote from their own website, they say that they are working to support and speed up the transition to a more equitable and just future. They say they believe that the main obstacles to change are forms of capitalism which concentrate the ownership and control of scarce assets in fewer hands and the defenders of such inequality.

Reasons for changes in the size of the PRS

Most of us will be aware of the reasons, but it is worth noting the factors which the briefing says it relies on here. It refers to financial reasons: Restrictions on claiming mortgage interest on buy to lets (Section 24), enhanced Stamp Duty for landlords and higher mortgage interest rates. It also refers to legislative reasons: The Renters’ Rights Act, more widespread housing licensing, the likelihood of greater enforcement, higher penalties plus coming higher energy efficiency standards (MEES).

A closer look at some of the findings

Although it is something of a self-inflicted spoiler, the briefing starts out by saying that its findings are that the private rented sector is consolidating and not contracting.

To establish this, the authors examined a number of different reports, surveys and statistics and we will look at those more closely next.

Rental stock levels

The briefing says that the Ministry of Housing, Communities and Local Government’s (MHCLG) dwelling stock estimates show that the PRS had reached 5m homes by 2025. It increased by 200,000 homes annually on average from 2005-2015 but only 45,000 homes annually on average since 2021. It points out that while this is indeed a slowdown it is still a significant increase.

The briefing adds that the English Housing Survey (EHS) confirms a similar story: The PRS reached 4.7m households in 2024-25 having increased by 55,000 annually since 2021-22.

However, the authors point out that data from such surveys lags the situation on the ground. Even if the PRS contracted in 2025-26 this would not show in the figures until 2027 or even 2028.

Tenancy deposit data

The briefing reports that figures from the Tenancy Deposit Scheme (TDS) show that by 2025 they held 4.7m tenant deposits valued at £5.5bn. This was a rise in 570,000 deposits since 2020.

However, the report says that while this information supports the continued size of the PRS, it is not a direct measure of it as such. It says reasons include that not all deposits sit within deposit schemes and some landlords do not even take them. It suggests that this data should become a more reliable indicator of the size of the PRS in future as more deposits flow into it.

HMRC rental income statistics

The briefing says that property rental income statistics from HMRC show a small decline in individual, non-incorporated landlords from 2.91m to 2.88m in 2024-25 – a fall of 30,000 landlords. They add that rental income declared suggested a reduction from 4.1m to 3.8m in homes such landlords owned after accounting for rental inflation.

The briefing later casts doubt on the usefulness of this statistic, due to the move towards landlords owning their properties through limited companies.

Mortgage lending data

The authors looked at mortgage lending (MLAR) data. The briefing says that new buy to let advances roughly halved from £41bn to £18bn over 2022-23 (although recovered somewhat to £25bn in 2026). The aggregate value of outstanding balances on buy to let mortgages has declined by 7% from its peak in 2022, although it still higher than before 2020.

The briefing casts doubt on the usefulness of this data in judging the size of the PRS. It says that only 40% of PRS properties are actually financed through buy to let mortgages. It adds that this data says nothing about the actual number of properties within the PRS, just the lending against them.

Again it refers to the increasing tendency for rental properties to be owned by limited companies – with new company registrations growing from 10,000 in 2016 to 67,000 in 2025 – something this data does not take account of. The briefing suggests that the fall in unincorporated private landlords using mortgages has been more than offset by the increase in limited company landlords.

Landlord sentiment surveys

In recent years several landlord surveys have appeared to show that landlords intend, or say they intend, to reduce their buy to let holdings. The authors took a closer look at some of these.

Data from the National Residential Landlords Association (NRLA) shows that the so-called ‘net sell intention’ of landlords rose from +5 percentage points in 2017-18 to +39 in 2025. The English Private Landlord Survey of 2024 suggests landlords intending to sell up rose from 16% to 32% from 2018 to 2024. Landlords’ reasons were both legislative and financial. The authors comment that larger landlords and those with BTL mortgages seemed more likely to sell.

The briefing surmises that there could be a so-called ‘cohort effect’ at play here, however. It suggests older, long term landlords could be cashing in and retiring. Company landlords on the other hand were more likely to be considering buying rather than selling.

The 2026 Property Investor Survey by Handelsbanken, a bank aimed at high net worth individuals, found that 84% of professional property investors planned to expand their portfolios. This was up from 54% in 2025.

The study suggests that even when landlords express an intention to sell up this does not seem to translate into a reduction in the PRS according to the statistics.

Housing transaction data

The briefing says an issue with some data that suggests the PRS is reducing is that it concentrates on sales alone and not churn in the market.

TwentyCi, a provider of property data, suggest 18.6% of PRS stock has been lost since 2016. Also, the number of previously rented homes being offered for sale has increased – from 8-10% in 2023-24 to 12-16% in 2024-25. The authors point out that this only refers to exits from the market however.

JRF’s own analysis of Energy Performance Certificate (EPC) data, when homes were surveyed for an EPC prior to going up for sale, shows only 5% were rentals in 2016-24 but this rose to 9% in 2025.

HMRC’s Capital Gains Tax data shows that more properties are being sold leading to a CGT liability. Some of these may be rentals, but not all. Again the authors comment that this could be due to a churn in ownership rather than any reduction in the PRS.

The authors considered surveys from agents Savills and Hamptons which looked at how housing stock moves between landlords and private homeowners and vice versa. Both appear to show small reductions in the size of the PRS annually in recent years (except for 2026 according to Hamptons). They say that this is more useful data on the changing size of the PRS. However, it does have a slight flaw in that it doesn’t account for properties which entered or left the PRS without involving a sale or purchase. For example inherited properties or build to rent (BTR) homes.

Again, in this regard, the authors conclude that whenever sales-related data suggested that the PRS was reducing in size other official statistics suggested it was continuing to grow.

Rental listing volumes

The briefing says that both Rightmove and Zoopla rental listings are sometimes used to support a claim the PRS is declining. Rightmove suggest the number of rental homes reduced 1% over the year to Q2 2026. Zoopla found that in June 2026 rental listings were 20-30% below Covid levels. The authors suggest longer tenancies (up from an average of 3.5 years in 2013-14 to 4.7 years today according to EHS data) could be the reason.

It quotes other data from TwentyCi which shows rental listings have increased 17% over the last year. The authors put forward a Savills report which suggests this could be due to landlords putting properties back on the market after attempting to sell them.

The briefing suggests the way in which Covid distorted the property market may still be impacting this type of data even today.

As an aside, it is worth noting that the briefing doesn’t appear to take account of the fact that not all rented homes appear in rental listings.

Other points made in the briefing

The briefing advances the idea that, on the landlord side, the PRS is seeing ‘consolidation and compositional change rather than contraction’. It brings forward data from the Deposit Protection Service (DPS) saying that, while the number of landlords owning one or two properties fell, the number owning more either held steady or rose. It suggests that this likely fits in with the growth in limited company landlords and other corporate landlords such as build to rent.

It also advances a theory that the PRS is holding steady and even growing because of issues on the tenant side. That is, there are substantial numbers of people who rent because they are unable to buy but want to – 1.5m people according to some data. Although this is to some extent JRF banging the drum for their cause it is a very valid point: Continuing demand for PRS homes has much to do with the wider housing shortage.

The briefing comes to the conclusion that the PRS is not shrinking as some might suggest. Indeed it is growing, if much more slowly than in the past. And it is changing. To quote from the briefing itself: ‘As we have shown in this briefing, none of the available data support with any real weight the idea that the PRS is shrinking, and the most recent official data available in fact shows the sector continuing to grow, albeit at a slower rate than previously. However, significant changes have taken place in the sector, on both landlord and tenant sides.’

Some final thoughts

The project undertaken by the Joseph Rowntree Foundation is an interesting one. It addresses an issue many have been wondering about for some time. It is unique (probably) in that it is a ‘study of studies’ rather than one that offers yet more data.

It also identifies a few issues for anyone trying to understand the size of the PRS. Most of the data on the PRS isn’t well suited to monitoring the changing shape and size of it. Some data is inconclusive. Some data is flawed in one way or another.

It shows us that there is no single source of data on the performance of the PRS in this country. In this regard, maybe the forthcoming landlord database will help with providing more accurate data about it. (Although if it does it will take several years for this to come through.)

One thing that is worth mentioning, which the briefing does not, is that some of the organisations who publish data on the PRS are promoting their own agenda in the process. This is likely to influence the data they publish as well as any study of it.

And, ultimately, whether the PRS is growing, shrinking or staying about the same depends on which figures you use – or believe.