Measuring housing need

Crystal ball or rear-view mirror?

30th January 2015, words by Neal Hudson

 

Current planning guidance places great importance on household projections as a starting point in identifying the number of new homes needed in a local market. With the 2012-based household projections expected sometime in February it is worthwhile looking at some of the issues that their use presents.

At a national level, the household projections play an obvious and important role in identifying the total amount of new housing supply that should be delivered. The current 2011-interim projections indicate around 220,000 new households forming every year in England. Work by Alan Holmans (2013) extends the projections out to 2031 but also recognises our increased consumption of housing and so the need to build additional dwellings that cater for second homes and vacant dwellings. This process raises the annual increase in households to 243,000 which is unsurprisingly similar to the previous 2008-based projections. This level of housing demand equates to roughly a one percent increase in dwelling stock each year.

Household projections are heavily based on population projections. As such they are highly dependent on the assumptions contained within the population projections. Perhaps the biggest unknowns within the population projections are the future trends in migration. Therefore any change in the attractiveness of this country as a place to live, work and study could substantially impact on the underling population projections (see chart opposite) and hence the household projections. Current planning guidance identifies households projections as only the starting point for estimating housing need and it is worth repeating the footnote that is found on the CLG’s 2011 household projections spreadsheet:

“They are not an assessment of housing need or do not take account of future policies, they are an indication of the likely increase in households given the continuation of recent demographic trends”

Given the continuation of recent demographic trends, it is also worth noting that they make minimal compensation for the massive historic under-supply of new homes. As the chart below shows, the net change in dwelling stock rarely fell below 1% of stock prior to 1980 but in recent years it has been consistently below 1%. Housing stock has actually increased at or above the rate of new housebuilding thanks to the more intensive use of existing dwellings. Houses converted into flats, empty properties brought back into use and beds-in-sheds have all helped to house a growing population in the face of low new build delivery. Looking ahead we should recognise the need for homes not just to meet demographic trends but also to meet past under-supply, future economic growth and the increased consumption of housing.

Housebuilding, Change in Dwelling Stock and Population Growth since 1853

Between 2001 and 2011 the actual number of new households formed in England was 158,000 per annum, well below expected levels (figure 2). One of the underlying differences between the actual and projected rate of household formation was a slower than expected decline in household size. The average household size had reduced through the 20th century, falling from 4.3 people in 1911 to 2.4 people in 1991. The rate of decline slowed post 1991 and the projections expected a continued decline at this slower rate. However, the release of the 2011 Census showed that the average household size had barely changed between 2001 and 2011 (2.37 & 2.36 respectively).

Analysis of the underlying demographic trends by the University of Cambridge for the RTPI (2014) suggests that there are a number of reasons for this stagnation in household size relative to previous projections. The main cause is the challenge faced by younger potential households in accessing the housing market. Because of the difficulties in buying and affordability pressures in the private rented sector, many more young people are living with their parents for longer or moving back in after university. They are also more likely to live in shared private rented accommodation for longer than previous generations. While these pressures may have been exacerbated by the credit crunch post 2007, many of the trends and affordability pressures had begun to affect the market much earlier in the decade.

One of the biggest challenges for the team putting together the 2012-based projections will be ascertaining how much of the stagnation in household size is due to reaching an optimum level of household size and how much is due to the housing crisis and a lack of affordably priced housing.

Figure 2, Figure 3

Circular References

There are a number of issues with both population and household projections at a national level and these are magnified at a local level. Perhaps the greatest local level issue is the circularity of the projections and the fact that households can only form if there is a home (household space) for them to move into.

For example, if a local authority is building sufficient homes for people to move into from outside the area, then the population is likely to grow. When the next round of household projections are published, they will pick up on the recent population growth and this will result in higher household projections for the local authority in question. Meanwhile, a local authority with an unaffordable housing market and minimal housing delivery may see limited net inwards migration and hence lower population growth. Future household projections will then reflect this lower growth and the household projections’ “evidence” for more housing need will be limited (no doubt much to the delight of a development resistant local authority).

It is for this reason that an over-reliance on the backwards looking household projections can be misleading. There may not be a crystal ball telling us exactly how much, of what type and where new housing is needed but there are more intelligent ways to objectively assess housing need. Not least is the recognised but under-used requirement to assess need based on market evidence such as land prices and housing affordability.

It is also important to recognise that they are projections of recent trends, not forecasts. Household projections will not necessarily reflect any additional household growth generated by future economic growth, employment, transport improvements or other drivers of migration. As discussed earlier, they certainly don’t account for the historic under-delivery of housing, particularly in some of the most unaffordable markets.

It remains to be seen how the upcoming 2012-based projections deal with the effects of the recession and the backlog of delivery. They should offer a better picture of housing need than the 2011-interim projections but given the period they are based on (2007-12), they will certainly not provide the full story and should still be treated with caution due to their reliance on past trends. Therefore, when assessing local housing need it is essential to investigate the full range of evidence available and understand all the interlocking factors that contribute to our need for housing now and in the future.

Given the crisis facing the housing market, perhaps an issue as important as where and how much housing needs to be built should be dealt with at a national level. This could help ensure that it adds up to the right national figure, is built in the right locations, is the right type and mix of housing, is linked into national infrastructure plans but also makes appropriate allowances for protected land, flooding and other limiting factors. The challenge will be in balancing this national approach with the priorities of local communities and their elected representatives.

Affordability Population Projections and Change in Dwelling Stock

There is minimal correlation between housing affordability, where population growth is projected to occur and historic housing supply.

A focus on only the published population and household projections will not ensure that housing is delivered where the housing crisis is most acute.

Annual Change in Dwelling Stock, 2007-2012

There is minimal correlation between housing affordability, where population growth is projected to occur and historic housing supply.

A focus on only the published population and household projections will not ensure that housing is delivered where the housing crisis is most acute.

 
 

Receive the latest research

Key Contacts