Skip to content

Conceptual differences between an aggregate of construction output measures and the GFCF dwellings measure This product is designated as National Statistics

Released: 14 June 2013 Download PDF


This note is in response to users seeking advice on the appropriateness of making comparisons between certain gross fixed capital formation (GFCF) and construction output data. The note outlines the similarities and key conceptual differences between GFCF new dwellings data and an aggregate of three ONS construction output series: ‘new public housing’, ‘new private housing’ and ‘repairs and maintenance housing total'.

Measurement of GFCF new dwellings and construction output

It may be expected that GFCF dwellings would be broadly comparable with the construction aggregate due to similarities in coverage.

  • GFCF dwellings measures investment expenditure in property assets for business or residential use; while

  • Construction measures the building of structures for business or residential use.

GFCF dwellings is produced as a part of GFCF, a component of the expenditure approach to measuring gross domestic product (GDP). According to the conceptual and legal framework for National Accounts, the European System of Accounts 1995 (ESA95), GFCF dwellings consists of “resident producers’ acquisitions less disposals of fixed assets during a given period”.

All three series used to construct the aggregate construction measure can be found in the construction output release.  The series used are part of the ‘all work’ measure of construction, a component of the output approach to measuring GDP.

The components of GFCF new dwellings and the Construction aggregate are identified in Table 1.

Table 1: Components of GFCF dwellings and the construction aggregate

GFCF dwellings components Construction aggregate components
New dwellings New builds (published construction series)
    Private housing     Private new housing
    Public housing     Public new housing
    Housing fees
    Self builds
Dwellings improvement Repairs & maintenance (published construction series)
    Contracted improvements     All repairs and maintenance housing total
    Hidden improvements
    DIY improvements

Download table

Similarities between these series

Figure 1: Components of GFCF dwellings

Figure 1: Components of GFCF dwellings
Source: Office for National Statistics

Download chart

Both of these measures have components related to new residential buildings and the maintenance/improvement of existing property. Roughly 80% of expenditure on new residential property (GFCF new dwellings) is equal to the construction output series. Public housing new dwellings (1) and private housing new dwellings (2) are equal to ‘public new housing’ and ‘private new housing’ within the construction series.

The dwellings improvement component of GFCF dwellings is also derived in part from output in the construction industry.  This is due to contract improvements (3) containing part of ‘all housing repairs and maintenance’. However ‘contract improvements’ only includes the part of repairs and maintenance that add to the value of the property. In addition, ‘contract improvements’ contain contract charges and VAT.

Conceptual differences between the aggregate and GFCF new dwellings

The first way the two series differ is in geographical coverage. The GFCF dwellings measure covers expenditure on dwellings within the United Kingdom whereas construction output is limited to Great Britain.

GFCF dwellings data contain additional components that are not included in construction output. This is because some building activity happens outside the construction industry. For example, the main area of difference for new dwellings is self-building, which contributes 10% to expenditure on total private housing within GFCF dwellings. This is equivalent to roughly 4% of total GFCF dwellings.

Similarly, dwellings improvement contains activity outside of the construction industry. ‘Hidden improvements’ is work carried out by people outside of the construction industry or work not recorded by the construction industry, whereas, ‘DIY improvements’ is unpaid improvement carried out by the household. Both ‘hidden’ and ‘DIY’ improvements are measured within GFCF dwellings but not construction output and contribute 27% and 5% respectively to the ‘Dwellings Improvement’ component of GFCF; these components together contribute 15% to total GFCF dwellings.

GFCF is an expenditure measure, and therefore is subject to other costs besides those spent on building/improving a property. The ‘new dwellings’ component includes housing fees that fall outside the definition of the construction output series. In ‘dwellings improvement’ the ‘contract improvements’ sub-component also includes contract charges and VAT. In both cases, these additional charges are attributed to the industry that charged the fees in the output approach to measuring GDP and therefore not included in the construction series.


In conclusion, although it may be expected that GFCF dwellings and the aggregate of the three construction series should be closely related, they differ due to:

  • geographical coverage - GFCF dwellings is a UK measure compared with the construction series which covers GB

  • GFCF dwellings is an expenditure measure whereas the aggregate is taken from construction output series; therefore

  • GFCF dwellings contain taxes, fees, and charges whereas the output measure does not; and

  • GFCF dwellings includes expenditure spent on houses/improvements that are outside the construction industry.

Notes for Conclusion

  1. Further information on the expenditure and output approaches to GDP can be found in a dedicated ONS article.

Background notes

  1. Details of the policy governing the release of new data are available by visiting or from the Media Relations Office email:

Content from the Office for National Statistics.
© Crown Copyright applies unless otherwise stated.