If the Government wants to solve the problem of land banking, it should start by revealing where it’s happening.
This week, the Communities Secretary, Sajid Javid, is set to unveil a new Housing White Paper. Its likely provisions have been heavily trailed in the press, including a promise to crack down on land banking.
Land banking is the practice of hoarding land in the hope that it will accrue in value over time, rather than developing it immediately.
Javid has previously attacked developers for land banking, demanding that they “release their stranglehold” over land. Housing developers have hit back saying that they simply need to keep a supply of land for construction. But as homeless charity Shelter have exposed, the top ten housing developers are sitting on enough plots of land for almost a million homes.
I decided to take Shelter’s analysis one stage further. A ‘plot’ is enough land for one house: Shelter and DCLG say that the average density for new-build homes is 12.5 per acre. But how do the developers’ land banks compare to the acreages of freehold land registered in their names with the Land Registry? To do this, I used data from the Land Registry released under FOI (passed to me by a contact – this is the same FOI results I used to assess the 50 companies who own a million acres of England & Wales). With this I was able to check the acreage of land registered in the name of each of the top ten housing developers. Here are the results:
|Developer||Land owned (acres)||Current land bank (plots)||Strategic land bank (plots)||Notes on land owned|
|Persimmon||9,839||93,519||17,500 acres/ ~218,750 plots||Sum of subsidiaries: umbrella org = 8,523 acres, plus Persimmon Midlands, NE, NW, South Coast, SW, Wessex, W Mids, Wales, Yorks|
|Taylor Wimpey||14,684||77,805||103,806||Sum of Taylor Wimpey UK Ltd (main landholder) & other subsidiaries: Taylor Wimpey Holdings Ltd and Taylor Wimpey Developments Ltd.|
|Barratt Group||386||71,351||70,000||Sum of Barratt South Wales Ltd, Barratt Manchester Ltd and Barratt East Midlands Ltd|
|Berkeley Group||1,064||42,125||5,000||Sum of Berkeley Strategic Land Limited, Berkley Seventy-Six Limited, Berkeley Homes (Southern) Limited, Berkeley Homes (Oxford & Chiltern) Limited, Berkeley Homes (Fleet) Limited and Berkeley Homes (Eastern) Limited|
|Galliford Try||415||14,200||?||Sum of Galliford Homes Limited and Galliford Southern Limited|
|Crest Nicholson||1,839||15,377||17,820||Sum of Crest Nicholson Operations Ltd and Crest Nicholson (Eastern) Limited|
|TOTALS||41,578 acres||404,040 plots||481,910 plots||Grand total: 885,950 plots|
As you can see from the table, the results look a bit odd. Why does Persimmon have so many declared plots in their land bank, but comparatively so little land registered in their name? Why does Taylor Wimpey, for example, appear to own more land than Persimmon, but declare fewer plots in their land bank?
Maybe it’s a problem with the data that I have. But to be clear: the data I’m basing this on is a huge spreadsheet released by the Land Registry giving the land areas owned by every single company and corporate body in England and Wales. I searched it for the names of the top ten developers, combining the totals for obvious subsidiaries, as recorded in the Notes column in the table above.
So whilst it’s possible that some of the big developers are registering land in entities with entirely different names to their own, I suspect something else is going on.
I think what this shows is that many developers’ declared land banks consist not of land they own outright, but land on which they have signed Option Agreements. This means they have struck deals with other landowners that they will buy out their freeholds if they get planning permission for development.
However, there is no central public register of Option Agreements on land. (You can’t even search for these for a fee.) And developers rarely reveal the location of their Optioned land.
This is problematic for several reasons. It means no-one knows where housing developers hold interests in land and are seeking to develop. It means we can’t properly assess whether they are improperly land banking or not – we don’t know quite how much they have a stake in, where it is, the value of the land, or whether it has yet received planning permission. And it’s problematic for local authorities trying to draw up Local and Neighbourhood Plans: they are trying to work out where to best place housing allocations, but may be hit with intense lobbying if council plans don’t match developers’ landed interests.
To fix land banking, the Government must enforce greater transparency in the land market. First, open up the Land Registry: that way, we’ll all get to see where the big housing developers own land. Second, create a mandatory register of land where developers hold Option Agreements. Put all that into the public domain for communities and councils to see clearly. That will lead to greater scrutiny of developers’ lobbying, more informed discussions about where housing is best built, and better Local Plans. And it will lead to a more complete answer to the question of who benefits from land being built on, and who benefits from sitting on land while it rises in value.
5 thoughts on “How to solve Land Banking? Reveal where it’s happening”
Excellent discussion – thanks for moving the debate on land forward with your recent Guardian article.
Of course, if we implemented an annual land value tax (LVT) on all land – urban and rural – we would not need to know where the land banks are.
The effect of an annual LVT payment would be to bring these land banks into use or the owners could “avoid” paying the tax personally by selling the land. Any potential purchaser would only want to take on the liability if they had an immediate use for the site and in addition they would deduct the LVT liability from their purchase price – so reducing the selling price of land and the cost of any new homes or other developments on the site.
LikeLiked by 2 people