Housing viability is the calculation that decides what gets built

The theory of development viability is well covered elsewhere. This is about the version you meet in practice.

Land value is what is left over

Developers calculate what a completed scheme will sell for, deduct construction costs, fees, finance and a required profit margin, and what remains is what they can pay for the land. This means land value is a residual rather than an input, which is counterintuitive and central to everything else. It also means anything that increases costs — including affordable housing requirements — reduces what a developer can pay for land.

Whether that reduces land prices or reduces affordable housing depends entirely on when the land was bought and at what price.

Viability assessments reopen policy

Where a scheme cannot meet policy requirements and remain profitable, many planning systems allow a viability assessment to justify reducing them. The assessment depends on assumed sales values, costs and profit margins, all of which are contestable.

This is why the same policy produces very different outcomes on similar sites, and why transparency of these assessments became a live political issue. The most sensitive input is usually the assumed profit margin, since a small movement in it shifts the conclusion further than any construction saving would, and it is the figure least open to independent checking because it describes what a business says it requires rather than what anything costs.

The land price problem

If a developer pays a price based on an assumption of minimal affordable housing, the scheme genuinely cannot afford more afterwards. The requirement has to be known and credible before land changes hands for it to be capitalised into the price. Policy that is routinely negotiated down is therefore self-defeating, because the market prices in the negotiation.

Land is frequently controlled through options rather than bought outright, so a promoter holds it for years while pursuing permission, and the price agreed at the beginning of that period becomes the constraint a planning authority runs into at the end of it.

Build costs are not the flexible part

Construction costs, particularly for taller buildings, are relatively fixed and have risen substantially. Tall buildings cost disproportionately more per square metre due to structure, cores, lifts and fire strategy, which is why height does not automatically mean affordability. The assumption that building higher makes housing cheaper does not survive contact with the cost curve.

The curve is not smooth either, since it steps at the heights where a different frame is needed, or a second staircase, or additional lift cores, or a full sprinkler and smoke strategy, which is why schemes so often stop one or two storeys below what the rules would have allowed.

Who captures the uplift

Granting permission to build creates a large increase in land value, and the question of who captures it is the core political argument in planning. Mechanisms include developer contributions, community infrastructure levies and public land ownership. Countries that have used public land assembly extensively — the Netherlands historically, and others — capture more of it, and the comparison is instructive.

Across a network, which mechanism is chosen matters less than when it applies, because a charge fixed per square metre and published in advance gets capitalised into the land price, while one negotiated scheme by scheme after permission comes out of the developer or out of the policy instead of out of the landowner.

The data here is patchy — most cities do not publish it consistently.

Build-out rate is the constraint no permission removes

A developer releases homes at a pace that does not undercut its own prices, so a large consented site delivers slowly no matter how much permission has been granted for it. That is why permission counts and completion counts diverge, and why a city can consent a great deal of housing and watch very little of it appear. Sites split between several builders, or carrying a mix of tenures that sell to different buyers simultaneously, absorb faster because the homes are not competing against each other.

In practice, finance sits on top of all of it, since a scheme pays interest for every month between buying the land and selling the last unit, which makes delay expensive and makes a change in rates alter what is viable without anything on the site changing at all.

The takeaway

The affordable housing number was decided when the land was bought, not when the plans were drawn.

The design decision is visible long after the people who made it have gone.