How Reliable Are Our Housing Statistics? Some Thoughts on a New Report

As most of us will know, a key issue in housing enforcement, and indeed housing generally, is the availability of reliable statistics upon which to make decisions. This is an issue that impacts not only enforcement, but wider housing policy and legislation in the sector.

With this in mind we will look at some findings from a recent report – Private Rented Sector Statistics from Across the UK: 2025. The report is interesting not just because of the statistics it presents but because of the way it shines a light on some of the shortcomings in data quality.

Introduction – about the report

This report has been published by the ONS and prepared by the Government Statistical Service (GSS) Housing Coherence team. Briefly, the purpose of this initiative is to ‘join up’ housing statistics from across the UK and find ways to make them more accurate. This is the first such report in this series and it focusses solely on the private rented sector. It suggests that such reports may become regular. Reports on differing PRS policies across the UK and empty homes statistics are in the pipeline for 2026.

Overview of the statistics used

The report points out that, since housing is a devolved matter across the UK, there are now few national statistics on housing. Each country – England, Scotland, Wales and Northern Ireland – has their own statistics and ways of processing them. This is in addition to their own housing standards, rules and regulations.

A number of different sources of statistics have been used in the report. For example, for England, the English Housing Survey (EHS) is drawn on frequently – but is by no means the only source.

To complicate things, there are often several sources of statistics on the same subject area in each case. The reports point out that while some of these different sources are comparable others are not.

The report draws our attention to the fact that, in some cases, the best available statistics are time-expired in practice. For example, the latest housing condition reports from Wales and Northern Ireland date back to between 2016 and 2018. (In these cases the statistics have not been used utilised in their report.)

Now let us summarise and discuss some of the findings.

Size and characteristics of the PRS

This section of the report draws on the annual (by financial year) Family Resources Survey (FRS) for UK-wide statistics. It draws on the official census for country-level statistics. The census is a 10-yearly event last undertaken in 2021 (2022 in Scotland).

The report estimates that the UK PRS has 5.5m households and accounts for around 19% of all households (2023 and 2024 figures). The social sector accounted for 17% of households and owner-occupied for 65%. It says the proportion of households in the PRS increased steadily between 2004 and 2017 but has since levelled out.

According to the FRS 32% of households were aged 25-34, making this the largest single group, and 24% were aged 35-44 (2024 figures). In comparison the census uses different age groups and suggests that the 30-39 age group is the largest tenant group.

The report suggests that the FRS does not have useful statistics on household composition. However, census data shows the ‘single family household: couple family’ was the largest household type in the PRS in England at 38% of households. Elsewhere one person households were the most common type.

Census data shows that the most common property type in the PRS was flats/maisonettes/apartments in both England and Scotland. In Wales it was terraced homes. Northern Ireland’s PRS was predominantly houses of various different types.

Houses in multiple occupation

The report collects some interesting information on HMOs. A key takeaway is that, due to the plethora of different licensing regimes, statistics on HMO numbers are very patchy indeed.

Both Scotland and Wales have reliable statistics on HMOs, however, with 15,274 and 3,050 licensed properties respectively. There are no reliable statistics for Northern Ireland.

Most interestingly, the report confirms that there is no single source of statistics on HMO numbers in England. It suggests a commercially run, charged-for source (Property Data) is the best available source. Their register relies on collating statistics from 310 separate local authorities. Property Data estimate there are around 239,000 HMOs in England whilst pointing out that their database has only 93% coverage.

Housing quality in the PRS

The report looks at statistics in two areas which are of particular interest to those of us in housing enforcement – housing standards and energy efficiency.

Housing standards

In England the report collates EHS data. According to this, 10% of PRS homes assessed using the Housing Health & Safety Rating System (HHSRS) had a serious Category 1 hazard (2023 figures). This compared with 4% of social and 8% of owner occupier homes.

The EHS says that since 2013 the PRS has had a consistently higher percentage of homes with Category 1 hazards than any other tenure. (However, the hazard rate has declined since then in all sectors of the housing market.)

In Scotland, the report collates statistics from the Scottish House Conditions Survey or SHCS. It points out that this is not directly comparable with the EHS. The SHCS assesses homes against the tolerable standard. In 2018 4% of PRS homes failed to meet the standard but this had risen to 27% in 2023. (The Scottish Government attributes this to increased smoke, heat and carbon monoxide alarm standards.)

Again, the report suggests that the principal sources of this information for Wales and Northern Ireland are time-expired and so does not incorporate them.

Energy efficiency

The Standard Assessment Procedure (SAP) across the UK assesses the energy performance of a building. It scores it between 0-100+ and assigns a banding from A (most efficient) to G (least efficient) to produce an Energy Performance Certificate (EPC).

In England and Wales PRS properties must meet at least Band E. Scotland is in the process of reforming its EPC system. There is no requirement in Northern Ireland.

The report points out that there are different ways of assessing energy efficiency and not all are comparable.

For example, official statistics use the EHS and SHCS statistics. These are based on a relatively very small sample of properties which are assessed for the surveys and EPC rated. (Again Wales and Northern Ireland statistics are considered time expired.)

Another method uses data taken from the official registers of EPCs. This is based on when a property was last issued with an EPC (if it has ever had one), which could be up to 10 years ago.

As it happens, on this occasion both these measures seem to produce not dissimilar end results. For example, the latest EHS/SHCS figures say 97% of PRS homes in England and 95% in Scotland were Band E or higher. Figures from EPC registers suggest that 98% of homes in England were Band E or higher, 97% in Wales and 90% in Northern Ireland. (Scotland does not routinely publish their aggregated data). But the fact remains these two measures are world’s apart in the way they are prepared. So perhaps questions should be asked about what is actually the best way of producing statistics on energy efficiency.

Rents

Lastly, the report collates information on rent levels. While the report quotes information on rent rises over the last decade (actual figures are available in the report) it is what it reveals about the way rental statistics are compiled that is perhaps of most interest. Several methods are in use: For example, in England and Wales rent data is collected from achieved rents for new and existing tenancies. In Scotland and Northern Ireland it is based on advertised rents. An important implication of this is that statistics on rent levels in those countries rely heavily on estimates.

Some final thoughts

This new report collates some interesting statistics which, although drawn from external published sources, are not always easy to find and compare. That it itself makes the report useful.

However, a key takeaway from presenting all this information in one place is this: Statistics on the PRS in the UK are generally pretty mediocre. They are fragmented, often incompatible and in some cases long out of date. The situation hasn’t been helped by devolving powers to the individual countries of the UK.

In some cases the statistics could be considered really very poor. The situation regarding energy efficiency and the information on HMOs in England stand out as particular low points.

In times when housing legislation is developing apace, for example selective licensing, rising Minimum Energy Efficiency Standards (MEES) for PRS property and of course Renter’s Rights , this ought to be a concern for those in the industry.

Of course, this report wasn’t just about presenting statistics. It was about highlighting the issues behind compiling housing statistics and the limitations of them. In that regard, this report has done a good job. We can only hope that it ultimately leads to some improvements in the quality of housing statistics.

Further information

 The full report, which provides more detail on the statistics used in it, can be read here:  Private rented sector statistics from across the UK: 2025