A very technical session

Tuesday’s session on Matter 17, Housing Requirements was very different from earlier sessions – which had ranged widely across the core topics around land supply and good growth. This one was inherently more technical – and indeed many of the areas of discussion almost mirrored that held at the technical seminar last November. The Panel had set out four questions for discussion, only one of which was fundamental: whether the need for 66,000 homes based on the 2017 Strategic Housing Market Assessment was both justified and properly calculated with respect to market and affordable housing having regard to national policy and guidance.  The other three concerned the relevance of the new household projections issued by ONS last September; the impact of Brexit; and whether the Mayor had adequately considered the possibility of increasing housing figures to help deliver the numbers of affordable homes required.

The discussion was also very structured. The Chair of the Panel asked the GLA team for an explanation of what they had done and why; then put some questions for clarification; the discussion was then opened up to the invited attendees; and the GLA asked to comment and the Chair then moved on.  The atmosphere was quite relaxed, but people mainly spoken quite succinctly and rarely ranged outside the specific question. The result was that the session was complete by lunchtime, with the Panel clear that they had all the information they required.

So, what were the major issues of debate?  In a way this was framed by the Chair’s introduction which stated that the issue was really about housing need. This led to a distinction being drawn between need – seen as a normative standard unrelated to whether resources would even potentially be made available to address these needs – and requirements which take more account of implicit standards set by government (e.g. thorough the limits on local housing allowance for younger single people).   In this context some questioned whether, if delivery could not be achieved, an aspirational figure unbacked by resources was at all relevant.

In answer to the first question –  whether the need had been calculated properly – the GLA stated that they calculate need differently from elsewhere in the country – as they have a right to do –because London is clearly a special case, mainly because of the tightness of the housing market, and because of the relative importance of migration. One of the issues was technical – using 10 years’ migration data to project trends, rather than the five used at national level.  The other was more substantive – in that they included in projected household formation an addition reflecting the extent to which formation has been below projections in the past few years.

The discussion centred on whether it would be better to have consistency across the country, even though it is accepted that markets differ –  because this provides a clear basis for comparison (and indeed appeal if the land is not coming forward); or whether the differences between London and the rest of the country are large enough to merit a different approach to projecting housing need. Most discussants appeared to think the London methodology was more robust than the national approach, although not all thought this was enough to justify the complications arising. 

The second area where the GLA approach differs from national guidance is the transparent way in which they deal with the backlog.  Even spreading it out over 25 years the backlog makes a large difference to the annual requirement – and, almost by definition, pushes up the proportion of these requirement that must be met in the social/affordable sector, and changes the unit size-mix. 

This issue re-emerged in the context of the final question, as to whether total requirement numbers should be increased to ensure more affordable housing. Some argued that market housing crowded out affordable housing – and also that increasing total numbers could not be expected to reduce house prices. The second would be the generally accepted view of economists. The first is more problematic – the empirical evidence shows little or no crowding out in either direction. Moreover, the S106 and CIL regime inherently means that more development overall is the main mechanism for generating more affordable homes. 

On the two final questions there was general agreement: take no notice of the new ONS household projections or of Brexit. On which note the Panel moved on.

Overall, for someone attending the Panel sessions for the first time one thing above all stood out.  This is not a process which aims to determine what can and should be provided. In particular, it is about need but with the complication that no government is going to provide the resources to enable that need to be fully met.  This will then be compared to the estimate of capacity would looks far too close to the need estimate to be believed. So implicitly it is not the supply of land but simply delivery which will be to blame when the required dwellings are not built.    

A related question is whether the most important element of housing need –household projections –  can really be believed.  At best these numbers simply set out what might happen if the future is like the past – but inherently the future will be driven by factors which are changing rapidly.  At the least, there should be a better understanding of these factors as well as some capacity to take these changes into account within the lifetime of the Plan.