How large is the housing gap?
25th Sep 2015
There's a "massive gap" of 100,000 units a year between what housing is needed and what is being built.
There's a few different ways of looking at it but 100,000 seems to be a pretty reasonable estimate of the gap (for 2014).
"Does the Prime Minister not think it is time to reconsider the question of the funding of the administration of housing, as well as, of course, the massive gap of 100,000 units a year between what is needed and what is being built?"
Jeremy Corbyn, Prime Minister's Questions, 16 September 2015
There are various different ways to measure the housing gap and Labour hasn't been able to confirm which figures Mr Corbyn was referring to.
Looking at a few of the ways of measuring it, 100,000 a year seems to be a pretty reasonable estimate.
If the government achieves its desired level of house building in this Parliament, the latest estimates (excluding any other additions to the housing stock) would suggest a gap of roughly 40,000-45,000 a year.
The easy way: comparing houses built to increases in households
England will have a projected 220,000 extra households per year between 2012 and 2022. The main house building figures put the number of homes built in England in 2014 at 118,000—so around 100,000 units less than that figure.
There are other house building figures though (incorporated in the net housing supply statistics), which aren't updated as quickly but are more complete. They show around 130,000 houses built in the 2013/14 financial year—making the gap about 90,000.
Households aren't the same as houses—they're groups of people rather than bricks and mortar. So we have to use a more complex set of estimates to take account of this, as well as more of the factors that feed into housing supply.
Two or more households can live in one house
A 'household' could be someone living alone, a family living in the same place, or a group sharing accommodation.
Two or more households can live in one dwelling. People living in the same place are counted as separate households if they don't share a living room or dining area.
So it's not as easy as to say that the 220,000 additional households a year would require 220,000 houses.
And housing supply is not just about how many new houses are built each year
The existing and future housing stock isn't just affected by new builds—though according to a report by Dr Alan Holmans for the University of Cambridge, they are a significant factor.
The number of housing units available can grow if large houses are converted into flats—instead of providing for one household then they may provide for three or four households. The number can also fall from demolitions and homes being made into office spaces, for example.
The figure of 130,000 homes built in England in 2013/14 doesn't factor in these additions or reductions of the existing housing stock. What does is the figure of net housing supply, which was thought to be around 137,000 in 2013/14.
The more complex way: estimating housing need from households
Dr Holmans' study took account of all of these things, and using his own household projections developed from the official projections for 2011-2021 (which have now been superseded by the 2012 ones), derived estimates of how many additional houses would be required each year. These aren't the only estimates that exist, but as far as we're aware they're the most recent.
His analysis estimated that we need to build about 170,000 additional private sector houses and 75,000 social sector houses each year—in total, an extra 240,000-245,000 houses each year, excluding any reductions in the existing housing stock. That includes allowing for second homes and vacancies in the stock. So that's a greater need than just the household estimates would suggest.
So what does this mean about the gap? Comparing the estimate of 240,000-245,000 houses with the existing net housing supply of around 137,000, the gap actually is still around 100,000.
That is based on the 2011 household data though so we don't know if things have changed with the 2012 household projections.
Earlier this week in an interview with the BBC Housing Minister Brandon Lewis said that success to him would mean building "something like one million" homes by the end of this parliament. If achieved that could leave a gap of around 40,000-45,000 per year (excluding additions or reductions in the stock) based on Dr Holmans' estimates.
But estimating future household numbers is complex
The official release which provides us with the estimates of additional household numbers each year explicitly points out that its figures are "projections" based on past trends, not forecasts (which would take account of things like changing economic circumstances).
Dr Holmans also argued (in relation to the 2011 projections) that the figures didn't account for different household patterns among recent immigrants, who he said tend to form fewer households than the rest of the population. He argued that means the official household projections may be too high.
We put this point to the statisticians who compile the projections and they told us they thought while they don't explicitly adjust for differences in average household size among new immigrants, any effect should be reflected in the overall average anyway. Each round of projections takes into account the most recent data on household formation.
Certain groups may be hidden from the figures—people who Dr Holman referred to as "concealed households". One example would be a young couple living with one set of parents in order to save for their own place. They wouldn't be visible as a separate household from the in-laws. But when they got their own place they'd become a new household.
Changes in rents and salaries will impact on how many couples do this in the future.
Levels of house building have been a long-term concern