Housing Outlook Q3 2025

Welcome to our third Housing Outlook of 2025. [1] This quarter, we show that delivering 1.5 million new homes by the end of Parliament would be a real step-change in housebuilding in England, but not game-changing when it comes to rebalancing housing supply and demand.

Since 2022, housing stock relative to population has been on the slide, and we estimate that delivering 300,000 new homes per year over the next five years is necessary to simply arrest this trend and take us back to 2021 levels. But if net additions were maintained at this rate over a ten-year period, this would lead to a significant improvement in dwellings to population, boosting the number of homes per 1,000 adults from 535 in 2024, to 554 in 2034.

The picture when it comes to affordable (i.e. sub-market) homes is less encouraging. The ratio of affordable homes per 1,000 adults in England has been consistently falling since 1980, standing at 88 in 2024, an all-time low. The Government’s new £39 billion Social and Affordable Homes Programme (SAHP) is hugely welcome, but even in a best-case scenario it would only stop this decline, with the number of affordable homes per 1,000 adults increasing ever-so slightly over ten years, to 91 per 1,000 adults in 2034.

Finally, we show that higher levels of housing stock to population correlate with lower rents, which suggests building focused in high-cost areas could have an affordability impact over time. But perhaps more pertinently, the Government should focus building in areas that are, or should be, highly productive rather than simply less affordable to boost growth as well as reduce housing cost pressures.

Hitting the new homes target would be a step-change, but one that needs to be sustained

The Government has made housing one of its top priorities, pledging to preside over the delivery of 1.5 million new homes in England this Parliament – equivalent to an annual average of 300,000 added to the housing stock each year from 2025 to 2029.[2] This would mark an unprecedented increase: over the last decade (excluding 2020-21 when there was a substantial drop in building activity due to the Covid-19 pandemic), annual average net additions stood at 223,000, and since 1971, the figure has exceeded 250,000 just once (see Figure 1). To this end, the Government has already made a number of big changes: reforming planning rules with mandatory housebuilding targets for councils; creating routes for developers to access Green Belt land more easily; and establishing a National Housing Bank with £16 billion of capital, that aims to inject an additional £53 billion of private investment into housing projects.

Figure 1: 1.5 million new homes are needed over this Parliament simply to turn the decline in housing stock to population around

But when it comes to per person housing, what does 1.5 million additional new homes actually translate into? As Figure 1 shows, additional dwellings on this scale over the Parliament are required simply to reverse the recent downward trend in the ratio of housing stock to the adult population. 1.5 million additional new homes would mean the housing stock per 1,000 adults in England would increase to 543 by 2029, just surpassing its peak in 2021 (541 homes per 1,000 adults).

At least at the aggregate level, then, even if the Government does deliver on its promise of 1.5 million more homes over the Parliament, this does not look sufficient to reset the relationship between housing and the overall size of the population. What is key is maintaining this level of building over a longer time frame. Figure 1 also shows that if the same level of net additions was sustained over the following five years to 2034, this would result in a longer-term upwards trend in housing per person, with the ratio reaching a historic high of 554 homes per 1,000 adults by 2034.[3]

Boosting affordable housing levels relative to the population will be harder to achieve

Boosting England’s housing stock overall is key to 1.5 million target, but increasing the number of affordable homes is a further crucial aim of this Government. As Figure 2 shows, the ratio of affordable (i.e. sub-market) housing stock relative to England’s adult population has fallen drastically over time. At its peak in 1979, the figure stood at 154 dwellings per 1,000 adults; in 2024 (the latest outturn data), it was just 88 affordable homes per 1,000 adults. The near-halving of affordable dwellings to adult population is largely the result of failure to replace the social housing stock lost to Right to Buy sales: today, there are over 131,000 fewer affordable homes than there were when Right to Buy was introduced in 1980.[4]

Figure 2: Affordable housing stock has fallen drastically relative to population since the early 1980s

Given this, the Government’s manifesto pledge to deliver “the biggest increase in social and affordable housebuilding in a generation” is very welcome, as is the £39 billion allocated at this spring’s Spending Review to the new 2026-2036 Social and Affordable Homes Programme (SAHP). The SAHP aims to deliver 300,000 affordable homes over the next ten years, with 60 per cent (180,000) of these new homes for social rent, a meaningful change given that just 14 per cent (75,000) of additional affordable homes in the last decade were for social rent.

In Figure 2, we also show how the affordable housing stock per 1,000 adults would change over the next ten years if three assumptions held true: first, that the SAHP delivers 30,000 homes per year (to reach its promised 300,000 total); second, that the private sector boosts housebuilding to the level required to meet targets (1.2 million over five years) and builds the same share of affordable homes under section 106 agreements as it has on average over the last five years; and third, that losses through sales and demolitions continue to occur at the current rate.[5] Under these assumptions the steady downward trend of affordable homes per 1000 adults would be ever-so slightly reversed, increasing the ratio of affordable homes to adults to 91 per 1,000 by 2034.

This might sound like good news, but in reality, it should be a concerning conclusion for the Government, for two key reasons. First, housing need is acute: as of March 2024, 1.3 million households were on local authority waiting lists in England, the highest number since 2014.[6] At the sharpest end, the number of households in temporary accommodation has been rising steadily since 2012 and reached a record high of 131,000 households (containing 169,000 children) in March 2025.[7] Second, there are reasons to be cautious when it comes to future private developer contributions, not least because the Government has foregrounded the role that brownfield land could play in unlocking sites for further affordable homes, where viability concerns could drive down section 106 contributions on these sites.

New homes are most needed in the least affordable and most productive areas, but there can be a trade-off here

Delivering on the aggregate housing targets (total and sub-market) is, then incredibly important, but building homes in the right places is also key to ensure that supply broadly matches housing demand, both now and in the future. As Figure 3 shows, there is a relationship between housing stock per 1,000 adults and private rental prices across English local authorities, with higher levels of stock relative to population associated with lower average rents. This suggests that focused building in high-pressure local areas could have a material effect on housing costs there: we estimate that if the ratio of dwellings to adult population increased from 550 to 570 in a typical local authority, average monthly rent would fall by £34.[8]

Figure 3: Higher levels of housing stock per population are associated with lower average monthly rents

So, just where should house building efforts be focussed to achieve the greatest effect? As the chart shows, local authorities in London and cities such as Oxford and Bristol are some of the most expensive housing markets in the country, suggesting they should be prime targets. Indeed, the Government’s housing targets tend to skew towards the least affordable areas in England. But there is a potential productivity and affordability trade-off for the Government here. Why? Because the least affordable areas are not always the most productive, and building in the most productive areas would help generate economic growth. While London, for instance, is both highly productive and a city renowned for eye-watering housing costs, there are local authorities like South Hams in Devon which combine low productivity with high housing unaffordability.[9] Equally, there are currently relatively more affordable areas such as Birmingham and Manchester which the country needs to be more productive, where housing cost pressures will increase as productivity grows and housing stock to population does not keep pace.

Pricey areas have more affordable housing to population, but also higher housing need

Finally, Figure 4 plots affordable housing stock relative to population against rents in English local authorities, revealing that areas with higher average monthly private rents are also in general blessed with higher affordable housing stock per 1,000 adults than elsewhere. However, this is largely driven by London, which has the highest rents in the country. The ratio of average affordable housing stock per 1,000 adults in the capital stands at 113, compared to an overall average of 84.

Figure 4: There is more affordable housing relative to population in high-rent areas in England than in low-rent areas

But the good news story ends there. Analysis also shows that many of the local authorities with highest rents also have large numbers of households in temporary accommodation. Housing need is particularly acute in London despite higher levels of affordable housing relative to population, indicating there remains substantial unmet need in these areas. As of March 2025, there were on average 20 households per 1,000 households in temporary accommodation in London, compared to the England average of 5 per 1,000 households. Across the country, on average local authorities have 49 households per 1,000 households on social housing waiting lists.

And as always, these aggregate measures obscure acute crises in some areas, particularly in London. At the extreme, in Newham, 3 in 10 households are on local authority waiting lists, and over 1 in 20 households are in temporary accommodation. Moreover, there are also warning signs from some core cities outside of London: for example, in Birmingham and Manchester, there are more than double the number of households per population in temporary accommodation than the national average (12 per 1,000 households), and one-in-ten households in Bristol are on social housing waiting lists. This implies there is scope for affordable housebuilding to be targeted to these areas with the highest supply constraints, somewhat at odds with the Government’s current plans which analysis suggests could see nearly half of all new social housing over the next ten years being built in rural areas.

Conclusion

Delivering 1.5 million new homes by the end of Parliament would be a genuine step-change in housebuilding in England – but not game-changing when it comes to housing per person levels. In fact, building at this rate across the Parliament would take housing per person levels back to just 2021 levels. Sustaining this level of housebuilding over 10 years and more is what is required to significantly shift the dial when it comes to housing relative to the population. For affordable housing, delivering these targets would reverse the consistent decline in affordable homes per 1000 adults in the UK, but sustained building at this rate over 10 years would only take levels of affordable housing per person back to 2017. However, any boost to the affordable housing stock will have a meaningful impact for the many adults and children currently housed in temporary, unsuitable or expensive accommodation.

[1] Contributions and comments from colleagues at the Resolution Foundation are gratefully acknowledged, particularly: Lindsay Judge and Hannah Aldridge. All views and errors remain those of the authors.

[2] Although the Government’s housing target is often interpreted as requiring 1.5 million new builds, in fact it is net additional dwellings by which the Government will measure progress (details set out in the Government’s Plan for Change document). Alongside new builds, the net additional dwelling measure takes account of dwellings from conversions and change of use as well as demolitions. Over the last decade (excluding 2020-21 when there was a substantial drop in building activity due to the Covid-19 pandemic), there was an annual average of 223,000 annual net additional dwellings, 13 per cent higher than the new builds total.

[3] The observation that a higher level of new homes is required over a sustained period of time to shift the dial on housing costs is far from new. See, for example, K Barker, Review of Housing Supply, Final Report – Recommendations, March 2004.

[4] The Government is reforming the Right to Buy scheme but previous Resolution Foundation analysis found their proposals were unlikely to reduce the outflow of council homes. Notably, the Government’s latest proposal has removed the commitment to replacing homes on a like-for-like basis, and dropped the replacement target altogether.

[5] Taken together, this sums to around 50,300 affordable homes added to the stock each year, demonstrating the significance of losses through sales and demolitions. Losses may be under-estimated as they do not account for other types of losses such as due to conversions or change of tenure. These estimates are inherently uncertain given the range of factors that may affect our assumptions.

[6] Source: MHCLG, Live Table 600: number of households on local authority housing registers (waiting lists), by district, England, from 1987, June 2025. Waiting list data was unavailable for seventeen local authorities.

[7] Source: MHCLG, Statutory homelessness in England: January to March 2025. Temporary accommodation data was unavailable for eleven local authorities.

[8] This follows the trendline seen in Figure 3, which shows a negative correlation between dwelling stock per 1,000 adults and average monthly rents – as dwelling stock increases, rent decreases. While this holds on average, there is a different relationship in London, where higher dwelling stock is associated with higher rents. This estimate does not take into account other factors affecting rent such as income.
[9] Source: RF analysis of ONS, Private Index of Private Rents; ONS, Subregional productivity: labour productivity indices by local authority district