New data on contractual controls on land: our views
You’d probably be upset if you discovered that the house you’d just bought backed onto land optioned for development. Even more so if there was no way you could have found it out before buying the house.
Well, as things stand, that’s a possibility. Contractual controls - where, for example, a farmer gives a builder the legal right to buy a piece of land in future - are often agreed privately. And that means no one, not even the Government, can get a complete picture of what can and can’t be done with land. We’ve written about this problem before.
So it was great news when the Government announced a plan to create a dataset of such agreements, and make it freely available. We think this is almost a great solution.
The rest of this post tackles two questions:
Why do we need this new data?
Why is the Government’s proposal only almost a solution?
Why do we need new data?
England and Wales have had a public register of who owns land since 1990, so what’s missing? As it turns out, a few vital bits of data.
Contractual controls allow people to control the future use of land, short of owning it. They’re used commonly in the land market, often by housing developers to secure land while they try to get planning permission.
But information about these agreements is not easily accessible, because many agreements are private. This means that smaller developers can’t easily find out which land is available, and local and central governments can’t plan land use.
This is where the proposed new dataset comes into play: it will gather and publish information on these contractual controls, for everyone to see. The Government recently consulted on its plans, and we submitted a response.
The solution … almost
Overall, we’re impressed. The consultation presents its data aspects clearly - we particularly like the table of proposed data. We think the proposals just need two small tweaks.
First, the new rules must cover unregistered land. The Department only plans to collect contractual control data on registered land - this means it would be blind to around 12% of land in England and Wales. It’s actually particularly important to include unregistered land, because:
It’s especially likely to be undeveloped. Unregistered land has not been transacted since 1990, so is quite likely to be undeveloped. Excluding it may particularly affect land under consideration for development - the opposite of what’s intended.
It’s often geographically concentrated. Excluding it may mean we continue to have a limited picture of land control in certain areas. (How do we know it’s concentrated? Because Anna, our director, previously used HM Land Registry data to map unregistered land.)
The second tweak: the dataset should include geospatial data. Under the current plans, data users will only have a description of the location to work from. Judging from current Land Registry titles, this may well be vague, e.g. “land north of Stone Road”. If communities want to know the precise land affected, or if researchers want to map them en masse, they’ll need to pay for title plans - but even those may not be precise enough. Including geospatial boundaries is critical to unlocking the value of the data, and supporting the policy goals.
Our full response has all the details. If you’re interested in discussing this, please get in touch - we’d love to hear from you.