The secret clause stopping us building



Section 106 is the sort of policy you will be forgiven for not having spent much time thinking about. It is an obscure piece of planning law, tucked away in the Town and Country Planning Act, and easy to ignore.
That would be a mistake.
If the Government is serious about its promise to deliver 300,000 additional affordable homes by 2035, it cannot afford to ignore the mechanism through which most affordable housing in England is actually delivered. In July 2025, ministers made that pledge, but if they want any realistic chance of meeting it, they need to confront the fact that the current system is not fit for purpose.
In theory, Section 106 allows councils to attach obligations to planning permission, requiring developers to contribute to infrastructure, schools and affordable housing in order to ‘mitigate the impact’ of their developments. In practice, it is a way for the state to push the cost of delivering essential public goods on to the private sector.
That cost is not optional. If a developer is faced with a Section 106 agreement, compliance is effectively the price of securing planning permission. Sometimes developers can negotiate a financial contribution in lieu of on-site provision, but the principle remains the same: affordable housing is being funded through a hidden levy on housebuilding.
Consider a typical development of 200 homes. A council may require that 20% are affordable, and may also insist that these units are delivered before or alongside the rest of the scheme. That raises costs and slows projects down.
But as we at the Adam Smith Institute detail in a new report, Section 106 is not just inefficient, it is structurally flawed. Nowhere is this clearer than in London.
London is a city that needs roughly 43,000 affordable homes a year. Yet in 2023/24, only seven affordable housing starts in the capital were tied to Section 106 contributions. Seven; a collapse of 99.86% in a single year.
This was not simply an aberration. It exposed how fragile Section 106 is when housing market conditions deteriorate.
Section 106 is already painfully slow. A Home Builders Federation FOI campaign found that the average negotiation time for an agreement reached 515 days last year. Legal complexity, stretched planning departments and administrative delay all play a role. But these problems become far worse when the market weakens.
And London’s market has been weak. Over the past decade, house prices in the capital have grown by only around a third of the rate seen elsewhere in the country. New-build transactions have fallen dramatically, making up only 0.9% of overall housing transactions.
That matters because Section 106 depends on a healthy market to function; a market that builds. Councils cannot extract obligations from schemes that don’t exist. And when transactions slow, Registered Providers find it harder to make the numbers work, because the cross-subsidy model underpinning Section 106 becomes less viable.
In other words, the system only works when conditions are favourable. Section 106 turns affordable housing into a by-product of market confidence. In good times, it can limp along. In bad times, it seizes up. That is no basis on which to meet a national housing target.
Worse still, it imposes a major cost on development. By forcing developers to shoulder the burden of affordable housing provision upfront, Section 106 acts as an effective tax on housebuilding. Estimates by the Adam Smith Institute suggest that burden amounts to £1.3 billion. Unsurprisingly, taxing development tends to reduce it.
The Government says it wants more affordable homes. If so, it should stop relying on a mechanism that is slow, fragile, and fundamentally counterproductive. Section 106 should not be tweaked around the edges. It should be scrapped.
The good news is that there is an alternative. The policy road map set out in our research paper would abolish Section 106 and replace it with a system better suited to the scale of the housing challenge. If ministers want to meet their target, they should start there.