Household projections have always had their limitations as a tool for planning the numbers of homes that need to be built in local authorities across the country. But the signs are that these limitations are about to become greater, at the very time that Government has hardwired them into the formulation of local plan targets. 

Until recently, household projections came with the disclaimer that ‘they are not an assessment of housing need [...], they are an indication of the likely increase in households given the continuation of recent demographic trends’.

This means that places where planning restrictions in the past have resulted in few new homes being built, and therefore fewer new households forming, get lower household projections. This then feeds into lower targets for future housebuilding, a circularity that has been widely explored before (see Measuring housing need).

Another concern is that if household formation has been constrained by economic circumstances at a particular point in an economic cycle, that too might temporarily change household formation rates which then influence the projections. But the use of the long-term trend in declining household size since 1971 meant these short-term effects had a limited impact.

That is no longer the case.

With the shift of responsibility for producing the projections from the Ministry of Housing, Communities & Local Government (MHCLG) to the Office for National Statistics (ONS) comes a change in methodology. Gone are the long-term trends. Instead projections will be based on trends going back only to 2001.

The 10 years from 2001 to 2011 surprised demographic forecasters.  Against all expectations, the long-term decline in household size stopped (see chart below).

^{Source: MHCLG}

Had household size bottomed out?  Perhaps the average household size couldn’t get any smaller.

But local trends reveal a much more complex picture than the national average and one that is strongly linked to the housing market. The 2000s were a period of unprecedented house price growth, particularly in the South of England where affordability underwent a marked deterioration and more people were excluded from the housing market (or, put another way, excluded from forming households).

It was in these least affordable markets, where the decline in household size didn’t only stop, but actually reversed between the 2001 and 2011 censuses.  Where housing remained more affordable and therefore accessible, across much of the Midlands and North of England, household sizes continued to decline (see What can household size tell us about housing need?).

More recent data from the Labour Force Survey suggests that even at a national level household size is now increasing, while the 2014-based projection continues to anticipate falls. By any measure we still are not building enough new homes, particularly in the least affordable parts of the country, stalling new household formation and boosting the size of existing households. That prevents the continuation of long-term trends; households simply cannot form if there are not homes for them to form into.

^{Source: ONS}

Using the shorter-term trends will result in reduced housing need figures in the least affordable housing markets in the country  –  the very places where the ‘housing crisis’ is most acute. This is according to the methodological note released on the ONS website in advance of the new household projections.

MHCLG has noticed the potential for its new standard methodology to be disrupted by the new population projections (see previous blogs in this series, below). There is a note on the Planning Practice Guidance web page promising a consultation on the impact this has on the standard methodology outputs. The change in methodology for the household projections will be an additional complication.

All of this will generate huge uncertainty for Local Planning Authorities, landowners and developers about how many new homes need to be built where. It was expected that cementing the link between local housing need and the household projections would bring greater clarity, but reforms to the demographic projections look set to derail that. We’ll find out more when the new household projections are published on 20 September.

Read the other blogs in this series: 

Why population projections are so important in house planning

Can we rely on population projections 

Further information

Contact Savills Research or read more on housing shortage