BLOGS

The UK residential development pipeline has rarely looked so contradictory. Thousands of planning permissions are being granted. Construction activity continues across hundreds of sites. And yet, beneath the surface, a growing number of large schemes — the kind that move the needle on housing delivery — are quietly being put on ice.

REalyse data tracking residential planning applications of 100 units or more across the UK shows that of the 2,546 large schemes with granted planning permission in the active dataset, 472 are currently flagged as On Hold or Shelved. That equates to roughly 148,000 consented homes sitting dormant — not refused, not in dispute, but simply not moving forward. The average consented scheme in this stalled cohort is 313 units in size, with an estimated average development value of around £53 million. At that scale, the cost of inaction is enormous, both for developers and for the nation's housing targets.

The three-way squeeze on viability

Viability appraisals sit at the heart of every large residential scheme. They model whether a project can generate an adequate return — typically a 15–20% profit on cost for a market-rate developer — after accounting for land, construction, finance, planning obligations, and sales or rental revenue assumptions. When any one of those inputs moves against the developer, the margin shrinks. Right now, several are moving at once.

1. Financing costs: eased but not forgiven

The Bank of England's base rate has come down from its 5.25% peak, but development finance remains materially more expensive than the ultra-low-rate environment of 2020–2022. Senior debt for large residential schemes is typically priced at 200–350 basis points above SONIA, putting all-in borrowing costs in the 6–8% range for many schemes as of mid-2026. On a 300-unit development with a £50 million-plus construction cost, that translates to millions of pounds in additional finance charges over a two-to-three-year build programme.

For schemes that received planning consent in 2022 or 2023 — when both land values and construction cost forecasts were based on very different rate assumptions — the gap between what was modelled and what the market now demands can make previously viable schemes look marginal or loss-making on reappraisal.

2. Construction costs: structurally elevated

Build cost inflation has moderated from the acute spikes of 2021–2022, but RICS and BCIS data consistently show that tender prices remain 20–30% above pre-pandemic levels in real terms. Labour shortages persist, particularly for skilled trades, and material costs — from structural steel to timber and mechanical and electrical components — have not meaningfully retreated. Biodiversity Net Gain requirements, mandatory since early 2024, and tightening Part L energy efficiency standards have added a further layer of compliance cost that was not priced into many original viability assessments.

For large urban schemes — the type most affected by infrastructure contributions and complex groundworks — build costs per square foot in London and major regional cities routinely sit above £300 per square foot, with some complex or high-specification schemes breaching £400. When those numbers are set against achievable sales values or rental income, the residual land value often falls short of what the vendor paid, creating a standoff that can last years.

3. Affordable housing obligations: less flexibility, more risk

Local planning authorities across England have been less willing to accept viability-based reductions in affordable housing proportions in the wake of scrutiny from the Housing Delivery Test and the revised NPPF. Many London boroughs maintain a 35–50% affordable housing requirement as a starting position, while Homes England grant funding — a critical subsidy for many registered provider-led schemes — has faced intense competition and does not stretch to cover all viable applications.

The result: developers find themselves caught between LPA requirements that assume a certain affordable housing quantum and grant subsidy levels that cannot bridge the resulting gap. REalyse data shows that 316 large refused schemes — totalling over 81,000 units — sit in a cancelled or on-hold state, a portion of which will have been refused or abandoned partly on viability grounds. Those represent a material opportunity cost to housing supply.

Build-to-rent: growing fast, but not immune

Build-to-rent has emerged as one of the most active corners of the large-scale residential market, partly because institutional investors can underwrite longer hold periods and accept lower initial yields than traditional for-sale developers. REalyse data shows a sharp acceleration in BTR planning approvals, reaching a peak of 66 granted schemes delivering nearly 21,000 units in 2024 — more than double the 2022 figure of 27 schemes and 8,200 units.

However, the 2025 data already shows a retreat: 65 granted schemes but only 13,725 units — a 35% drop in unit volume from the 2024 peak, suggesting that while scheme numbers held steady, average scheme sizes pulled back. With 2026 data still accumulating (23 schemes, 5,595 BTR units granted to July), the trajectory warrants close monitoring.

The challenge for BTR operators is that residential rental yields — while REalyse data shows strong performance in core urban markets — still struggle to absorb both elevated construction costs and financing rates simultaneously. A scheme targeting a 4.5–5.5% initial gross yield in a regional city looks very different when debt is costing 7% and build costs have risen 25% since the feasibility stage. Operators are responding by repricing rents — average asking rents across major BTR markets have risen materially — but there is a ceiling set by tenant affordability.

What the pipeline actually tells us

The headline figures mask a more nuanced picture. Of the large residential pipeline tracked by REalyse, 1,100 schemes are actively In Construction — a meaningful number — and 2,300 more are in progress through the planning system. But the 201 schemes that have been Withdrawn from planning, representing a further 33,000-plus units, indicate that developers are not simply waiting for better conditions; some are abandoning their positions entirely.

For the government's 1.5 million homes target to be met, unlocking the stalled consented pipeline is arguably more urgent than granting new permissions. REalyse data points to a latent supply of hundreds of thousands of consented or near-consented units that could, in theory, be brought forward — if viability can be restored.

Outlook: selective recovery, structural constraints

The conditions needed to unblock the pipeline are beginning to edge into place, but slowly. Further Bank of England rate cuts would reduce development finance costs and, critically, improve the buyer mortgage affordability that underpins sales revenue assumptions. A sustained period of construction cost stability — rather than active deflation — would at least allow appraisals to be underwritten with greater confidence.

Policy levers matter too. Accelerated Homes England grant allocations, greater flexibility in affordable housing claw-back mechanisms when viability is genuinely constrained, and faster Section 106 negotiations could all make a meaningful difference at the margin. Some local authorities are experimenting with design codes and pre-application engagement processes that reduce uncertainty costs — a real, if underappreciated, drain on viability.

For investors and developers using REalyse data to monitor the pipeline, the opportunity lies in identifying consented schemes where land has been written down, operators are motivated to transact, and the underlying residential demand — visible in local sales and rental market data — remains robust. Those are the conditions in which a stalled scheme becomes an acquisition target rather than a cautionary tale.