Density: a walkover for developers?

[+ update at the bottom of this page as the Examination ends in May.]

Today in the London Plan Examination in Public (EiP) was the discussion on density policy.

Here I’m capturing my personal reactions – largely from memory – before they fade. I’ll later merge them with other people’s more balanced and complete notes and with Gabi Abadi’s notes on the morning session. They will then appear on the blog at

The GLA proposes to dispense with the ‘density matrix’, and thus with any upper or lower limits on housing density, replacing it with a site-by-site setting of densities to be conducted by Boroughs.

I appeared in my own right (not as Just Space); Caroline Shah from Kingston appeared for Just Space, Pat Turnbull on her own account (not at LTF), Peter Eversden and Michael Bach for London Forum of Civic and Amenity Societies, Kate Gordon for FoE and Neil Sinden for CPRE. Ian Gordon from LSE London, who had written 2 of the research papers commissioned by the GLA and had a big influence on the policy was there. The amazing James Stephens appeared for the Home Builders Federation but had to leave before the end. Duncan Bowie appeared for the Highbury Group on Housing Delivery. London First and some developer interests. London Councils and just one borough (Bromley!). Nicky Gavron for the London Assembly.

There was a great deal of criticism of the GLA new approach – an approach so new that it was still being re-written, with a lot of further proposed changes at the end of their statement for the day.

In essence there seem to have been the following issues:

What was wrong with the old system? Nobody thought it was perfect. Just Space had been criticising it for years for over-reliance on PTAL and ignoring social infrastructure capacity and social impact, especially in inner areas. Others had other criticisms but wanted to see an evolution through improvements designed to make a transparent quantified basis for policy.  Ian Gordon and the GLA wanted to scrap it. Ian’s position is that it has had no detectable impact on the actual densities of what has been built, is therefore a failed policy which existed for symbolic purposes only so it would be hypocritical to have it in the plan. He considers that “…we were also arguing that, since the negative effects of height/over development or whatever were essentially local in their incidence (unlike most of the expected benefits),  it was appropriate for individual boroughs to set and police upper limits in whatever terms/ways were most appropriate  [as the GLA could not be trusted to do this or need to be involved ]- which we thought unlikely to be via some mechanically derived number for a broad class of area.…” I had counter-argued that the need to slow down or reverse the escalation of residential land prices was a London-wide and strategic need which would & should justify a London-wide set of limits binding on Boroughs.

Would the boroughs have the capacity (skills, staff levels) to develop these site-by-site density settings? Widespread doubts about this, though GLA defended themselves stoutly, referring to Practice Public and other upskilling initiatives.

What if boroughs set densities which, in the out-turn, meant they would not attain their total output targets? I don’t recall an answer being offered on this.

How democratic is all this? Caroline Shah and Pat Turnbull both spoke strongly against the shifting of these crucial decisions out of the sphere of public debate (on London Plan and Borough Plan policies) and into the closed world of developer/council pre-application discussions in which deals are virtually done before “applications” reach committees. Caroline illustrated this concern from Kingston and Pat Turnbull from Hackney.

My argument that the enforcement of clear, upper limits on density would help discourage over-bidding by developers when they are buying sites and thus help damp land price escalation – to the benefit of social housing providers as well as private developers and ultimately households. (Many GLA statements favour discouraging such over-bidding in other contexts.) GLA mounted what I thought was a rather lame defence that recent changes in national and GLA Viability guidance make it harder for developers to argue that the price they have paid for land should be taken into account in ‘viability’ calculations and used to justify social housing reductions. They  brought in their new head of Viability, Robert Wacher, who argued that this can be really effective in ensuring that schemes are evaluated in the light of truly policy-compliant valuations, citing the Islington Parkhurst Road case. This was a brave effort but I countered by saying that, under the new GLA policy, there would be no policy on density, and thus no concept of policy compliance. Thus the reforms they were relying on would not work.  (If there was time I should have added that Islington was a council with strong and embedded policy requirements. But imagine trying to do that in Redbridge or Bromley.)

Is all this compatible with national policy?  James Stephens in the morning had argued that the GLA’s design policies did not accord with the NPPF requirement that applicants need clear enough guidance on what will be permitted to make compliant applications. He didn’t actually say this again on the density issue but it was in his written statement and is just the kind of issue the Inspectors have to look out for.

So what will happen? Many voices (including the London Assembly) were calling for the existing matrix or something like it, to be retained, or retained until better quantified arrangements were developed by the GLA or by each borough. I thought a map would be better than a (necessarily crude) matrix. The logic of the debate should persuade the panel to recommend something like this.  But will they have the nerve?

“Once all this work has been completed for London several years from now, then we may eventually discover that the capacity for new housing within the developed area of London is much less than has been estimated by the Mayor through his SHLAA back in 2017. But by then it will be too late to stop the London Plan juggernaut continuing on its course of failure.

Of course, we appreciate why the Mayor would wish to dispense with the Density Matrix. This is because, if it had been retained, it would have had to be amended to indicate that much higher densities were to be tolerated in traditional suburban locations (characterised typically by low-density housing of about 20-30 dwellings per hectare). This would have attracted the ire of the residents of outer London. The London boroughs, however, know that this is what they will have to contend with.”

Later: 22 May 2019

  1. Today in the final session of the Examination, Robert Wacher (see above) said very firmly that, in their ‘viability’ testing they had estimated land costs for residential development per completed dwelling unit, e.g. £30,000 per dwelling, stressing that the value of sites depends on how much can be built there. This helps support our argument.
  2. Ian Gordon at the LSE has published his own blog post on this density issue, and I have been able to reply to his criticism on the comments section of his post. It is at [link updated 2022]

Later: 16 July 2020

Peter Bill article following the death of Tony Pidgley (boss of Berkeley Homes) on just hw he made decisions and, in particular, how adept he was at securing extra density after gaining an initial permission: what did I say?

I’m getting tired now.  May add more to this tomorrow. Meanwhile here are the last 2 paragraphs of James Stephens’ written statement (HBF) which many of us have enjoyed:

Author: Editors


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: