Dear Jonathan

NATIONAL STATISTICS STATUS OF CONSTRUCTION PRICES AND COST INDICES, CONSTRUCTION OUTPUT AND NEW ORDERS STATISTICS

Thank you to you and your teams for engaging effectively with us as you have developed your construction statistics following our assessment.

On behalf of the Board of the Statistics Authority, I am pleased to confirm the re-designation of Construction Output Price Indices, GB Construction Output Statistics and Construction New Orders as National Statistics. We will reconsider the re-designation of sub-national estimates of construction output, once further development work has been undertaken.

Since de-designation in December 2014 and our assessment, we are pleased that ONS has improved many aspects of this suite of statistics. ONS has improved their accessibility and transparency by publishing detail on the strengths and limitations of these data and publishing all industry data in one location on its website. Guided by requests from stakeholders and in collaboration with industry experts, ONS has enhanced its analysis and supporting commentary of developments in the construction industry and within the wider context of short-term economic indicators. ONS’s work with its stakeholders through the Construction Statistics Steering Group (CSSG) and with industry experts via the Consultative Committee on Construction Statistics (CCCIS) has been positive, yielding significant improvements in the quality assurance of data and the development of methodologies, concepts and interpretation of construction statistics respectively. Further details on actions taken by ONS to address the requirements of the assessment report are provided in the annex to this letter.

The April 2014 assessment report, and ongoing reviews of development work, raised several areas of concern relating to data quality which formed the basis of our considerations. These included concerns over the quality of construction output price indices, bias in early estimates of the output measures, the divergence between new orders and output series and movements in sub-national construction output statistics.

We considered at length ONS’s response to the assessment report and these concerns. In particular, I welcome the work to improve construction price data, through the introduction of a new construction output prices index, incorporating the input costs of labour, materials and plant, supplemented by a proxy for margin; this has improved accuracy, transparency and accessibility of these price estimates for users.

Working closely with users and methods experts, ONS has improved the early estimates of construction output by introducing an improved method to impute data for businesses that have not yet returned their survey responses, along with a further adjustment to address additional potential bias in early estimates. This has allowed those engaged in industry and wider economic planning to obtain an earlier, more stable, view of the industry’s performance. Further support has been provided to industry analysts through the clarification of the scope and coverage of the new orders and output series. Conventionally, analysts have focused their work on the assumption that these two series are directly comparable, often using new orders data as a leading indicator of the performance of the construction industry and the wider economy. Publishing detail on the make-up and coverage of these data however, has assisted users in understanding the relationship between new orders and output data, and re-defined the use of these data in analysis.

Sub-national and sub-sectoral estimates of construction output

ONS has worked closely with CSSG and CCCIS on improving sub-national and sub-sectoral breakdowns of construction output. Whilst these users have acknowledged the improvements ONS has introduced to the modelling of these estimates, some users remain uncomfortable with the path of sub-national estimates of construction output. Consequently, these detailed breakdowns were excluded from consideration for re-designation. We welcome your ongoing work to improve these detailed estimates of construction output and look forward to considering the National Statistics status in due course. To make it clear to users, ONS should therefore prominently flag within published statistics that these sub-national and sub-sectoral estimates of construction output do not carry National Statistics status.

Given the importance of construction statistics to those engaged in industry and wider economic planning, we support ONS’s plans to continue to work closely with its stakeholders and industry experts to determine what further insights can be generated into the performance of the construction industry more widely. These might include, for example, capturing aspects of “off-site” activity as part of a wider definition of the construction industry, measuring the impact of improvements in the quality of building materials and productivity developments arising from better building and construction design.

I am copying this letter to Frances Pottier at (BEIS) as the chair of CCCIS and to Grant Fitzner, ONS’s Chief Economist and Craig McLaren and Ceri Lewis of ONS’s construction statistics team.

Yours sincerely,

Ed Humpherson
Director General for Regulation