Last week, Prof Glen Bramley published research on housing requirements across Great Britain. Here, he explores how household projections are made and the problems current methodologies present.
As I write this it is difficult to focus on anything other than the UK’s political meltdown over Brexit, and the apparent economic death-wish of sections of our political class. However, among our government’s neglected areas of domestic policy sits housing and the widespread view (if not total consensus) that we need to build more of all kinds, particularly in England.
We have a Conservative Government which has committed to a target of building 300,000 homes a year in England, and which claims to be building more than 200,000 already. However, it has to get these targets embedded within the planning system at local level, and that is where a ‘numbers game’ involving ‘Objectively Assessed Need’ (OAN) is important. It is deeply embarrassing and counter-productive for this Government that the Office for National Statistics recently produced updated ‘household projections’ for England, which said that the expected growth in household numbers would be only 165,000 per year, instead of the previously stated 216,000 per year.
The history of planning for housing in England since the late 2000s has featured a number of U-turns, including the scrapping of ‘top-down targets’ and then their progressive reintroduction. This culminated in the so-called ‘standard method’ of calculating ‘OAN’ issued in autumn 2017, which took the household projection number and then increased it by a factor depending on how much worse ‘affordability’ was in a locality than a national norm. In a highly embarrassing move, the Ministry of Housing, Communities and Local Government (MoHCLG), in a recent consultation, has been forced to propose that the latest household projections be effectively ignored or discounted for this purpose. This will not go down well in local planning authorities, who have struggled for years to reconcile attempts to assess housing needs mainly based on household projections, with local politicians who are only too aware of ‘NIMBY-ist’ sentiment among their electorates.
I agree with the government that we should ignore this set of household projections in setting planning targets, but would respectfully point out that I and others have been highly critical of over-reliance on household projections for a long time. That is because, especially at local level, there is a fundamental circularity – if you don’t build much housing, you won’t have very much household growth. I have expanded on this argument and produced a lot of evidence on this (Bramley & Watkins 1995, 2014, 2016). Other economists like Geoff Meen and Paul Cheshire have also broadly shared this view.
There are three factors behind the fall in the household growth projection:
- recent reversal of long-term trends of increased life expectancy;
- expected reduction in international migration post-Brexit;
- methodology change in basing trends on the propensity of different groups to form separate households in the short period 2001-11.
My view about these is: (1) yes, in short term, but surely health policy will be galvanised to reverse this adverse change; (2) yes, but this is extremely uncertain and the migration assumptions (ONS’s or mine) are well below the actual figures from every one of the last 15 years; (3) 2001-11 is too short a period, too characterised by shortage/affordability problems/financial crisis to be appropriate for a long term extrapolative approach, and we should not use such an approach anyway.
There is clear evidence from basic descriptive analysis of regional trends over 25 years and from econometric modelling (Bramley et al 2010, Meen, 2011, Bramley 2015, Bramley & Watkins 2014, 2016) that housing supply shortage and affordability pressures lead to reductions in household formation by younger adults. That means that there is a large backlog of unmet need and potential demand for separate housing in the adult population; in my just-published study of ‘Housing Requirements’ I estimated that to enable these ‘suppressed’ households to form over the next 15 years would require an annual addition of 69,000 to the basic household projection number. I also argued that a sensible housing supply policy would need to provide for a somewhat greater ‘vacancy reserve’ and numbers of demolitions – equivalent to about 40,000 extra units per year in England. This would enable the market to function effectively; accommodate the higher level of estate redevelopment expected; and to deal with the backwash from the Grenfell tower disaster.
Over the last decade, housing supply has collapsed, then very slowly recovered. Affordability remains terrible, particularly in London and the South. Sensitive indicators of housing need (e.g. rough sleeping, temporary accommodation) have got worse, while others remain significantly worse than 10-20 years ago. Yet each successive household projection has been lower than the last, with the latest showing the sharpest fall. The system has come to resemble a plane in a downward death spiral: the worse things get, the more the household growth indicator points downwards, and the steeper the dive. It needs a re-boot.
There are a range of practical things which could be done in the short term to improve the approach to planning for new housing requirements. These include changing the base figure; allowing for the suppressed backlog of households; and increasing the markup factors associated with affordability. However, I continue to argue that the basis for assessing housing requirements should change fundamentally, towards an approach based on outcomes. We should identify a range of relevant outcomes to do with different kinds of housing needs (including homelessness, concealed households, overcrowding, unaffordability, etc.); monitor how well or badly we are doing in each area; and make economics-based forecasts of where we are likely to be in 10-15 years time if we carry on as we are, or under various alternative supply scenarios.This is the key methodology used in my recent study, but in essence similar to that used in a major previous study for the then Department of CLG in 2010.
Bramley, G. & Watkins, C. (1995) Circular Projections: Household Growth, Housing Need and the Household Projections. London: Council for the Protection of Rural England.
Bramley, G., Pawson, H., Pleace, N., Watkins, D. & Pleace, N. (2010) Estimating Housing Need. London: Department for Communities and Local Government.
Bramley, G. (2013) ‘Housing market models and planning’ Town Planning Review 84(1), 11-34. DOI: 10.3828/tpr.2013.2
Bramley, G. & Watkins, D. (2014) ‘A sub-regional housing market model for England with endogenous migration and household formation: its role in assessing the adequacy of planned new housing’, British Society for Population Studies Conference, Winchester, 10 September 2014. Special session on Demographic projections and forecasts.
Bramley, G. (2015) ‘Housing need outcomes in England through changing times: demographic, market and policy drivers of change’, Housing Studies, 31, 3, p. 243-268 10.1080/02673037.2015.1080817
Bramley, G. & Watkins, D. (2016) ‘Housebuilding, demographic change and affordability as outcomes of local planning decisions: exploring interactions using a sub-regional model of housing markets in England’, Progress in Planning, 104, pp.1-35. 10.1016/j.progress.2014.10.002
Bramley, G. (2018) Housing supply requirements across Great Britain for low-income households and homeless people. Report of Research for the National Housing Federation and Crisis.
Fitzpatrick, S., Pawson, H., Bramley, G., Wilcox S., Watts, B., Wood, J. (2018) Homelessness Monitor England 2018. London: Crisis.