
Counting Houses Is Difficult & The Changing Market Is Making It More So
This Limits Our Ability To Accurately Assess New And Existing Policies
Measuring new build completions should be easy as there’s a physical object you can see and feel but it turns out to be quite difficult. Limitations in the available housebuilding data mean we don’t accurately know how many new homes are being built, where they are being built, and who is building them. That is a big problem.
The most commonly used housebuilding data for England is MHCLG’s House Building series of tables and this data is regularly used to highlight a number of housing problems including a lack of new build supply and the decline in local authority house building. Unfortunately, there are several issues with this data and it is frequently misused. The biggest issue, as Figure 1 shows, is an undercount in the number of completions when compared to the housebuilding completions recorded in MHCLG’s more comprehensive Housing Supply data. For example, the total undercount across England for the 2018/19 financial year was 44,580 dwellings based on Housing Supply data released today.

The last couple of years have seen an improvement in the MHCLG House Building data but the undercount remains and so recent MHCLG statistical releases recommend that this data is only used as a leading indicator while the least frequent and less timely completions data released in MHCLG’s Housing Supply release is the most comprehensive measure.
The exact reasons for the undercount are not confirmed but it appears to be linked to the falling market share of the largest provider of new home warranties. NHBC have historically provided a substantial share of the underlying data used to create the New Build statistics. Their data, alongside other sources of building control inspection data, was grossed up to provide a measure for the whole market. Figure 2 below shows that the MHCLG New Build series had, until 2017, tracked the NHBC data when grossed up by a market share of 85%.
However, the more comprehensive Housing Supply data shows that NHBC’s market share was closer to 75% from 2010 to 2015 and has fallen further since. Today’s data suggests an NHBC market share of 60% during the 2018/19 financial year. MHCLG have recognised the problem and the House Building data has diverged from the underlying NHBC data since 2017. However, it still undercounts and therefore should only be used as a leading indicator.

The fall in NHBC’s market share may also reflect the broader range of groups that are now delivering new homes. Volume housebuilders may still build the majority of new homes but there has been an increase in activity by SME housebuilders, high-density luxury developers, build-to-rent investors, housing associations, and others. For some of the organisations in these groups, an NHBC warranty may be too expensive or not attractive compared to the alternatives.
MHCLG’s Housing Supply data may provide us with a more comprehensive measure of new build completions but the lag means we’re only just today finding out what happened in the 2018/19 financial year and there’s no breakdown on who is building these new homes or what type of homes they are.
Unfortunately, our limited ability to accurately measure housebuilding and our constrained understanding of who is doing the building make it very difficult to accurately assess the success or failure of existing policies and identify new ones that could increase new supply.