The compilation methods of building price indices in Britain: a critical review

Abstract

The major message that I will be trying to convey is that we often misinterpret the available data because of inadequate attention to how they are produced … (Griliches, 1994, p. 2)

The importance of accurate measurement and pertinent modelling of the general level of construction prices cannot be overemphasized. Uses range from macroeconomic statistics such as real value of investment to micro‐level budgeting like construction project price forecasts. Numerous research studies posit that the measured productivity growth rates of the construction sector are distorted and that an inaccurate general construction price index is a main villain of the piece. The academic research published in this arena has primarily focused on models to forecast or predict changes in the general construction price level, whereas this research scrutinizes the compilation methods of the general construction price index in Britain and finds that the indices measure the price movement of more traditional building trades but almost completely ignore the mechanical and electrical services. The existing indices also do not gauge the tender price movements of new building work that is not procured through the conventional lump sum BQs route, such as many projects in private housing and PFI markets. These make out a case for allowing resources to the project of developing alternative methodologies: (1) a revised version of the present method; and (2) hedonic price index.

Keywords:

• Tender price index
• construction output deflator
• index compilation methodology
• construction statistics

Acknowledgement

We are grateful to Messrs Peter Fordham, Joe Martin, James Meikle, Ian Pegg and Lindsay Pullen for useful discussions and provision of information on the TPI method being adopted in UK. We also thank the four anonymous referees for their comments and suggestions. The opinions and views expressed in this paper are not necessarily those of the organisations the authors belong to and the authors are solely responsible for any errors that this article may contain.

Notes

1. The ‘repair and maintenance’ work actually includes, especially for housing, much improvement and alteration work. The actual share of ‘work to existing structures’ is unknown, but significantly higher than 47% as reported in Table 1 because while improvement and alteration of housing is reported as ‘repair and maintenance’, alteration and improvement of other buildings and infrastructure is reported as ‘new work’.

2. Cannon (1994) and Briscoe (2006) call for attention to the quality and usefulness of the general construction statistics in the UK. Regarding construction price and cost statistics, Fleming and Tysoe (1991) provide a comprehensive collection of British construction cost and price indices and Fleming (1966) is a valuable review of various indices of construction factor costs and market prices over the period between 1845 and 1964.

3. Since this paper was written, the Department of Trade and Industry (DTI) has been renamed the Department for Business, Enterprise and Regulatory Reform (DBERR), with some associated changes in the scope and functions of the Department. It is probable that in the near future responsibility for construction statistics will move to National Statistics. However, as of 2007 the tender price indices we discuss were published by DBERR in the same format, and using the same method, as formerly under DTI. We therefore have chosen to continue to refer to this as the DTI index, the name by which it is best known to users.

4. Fleming (1965) was an earlier attempt to produce an output price index of the construction industry from the bottom‐up approach. The method is to measure the movement of the factors that determine the movement of the output price namely, labour cost, material cost, productivity, and overhead and profit (OHP) of the construction firms. However, productivity and OHP are notoriously difficult to gauge. It has become more popular to measure them by using the tender price indices lately. For comments about material and labour cost indices, please see note 8.

5. Elemental bills have become more popular recently, in which each bill covers a major element or component of the building such as external walls, internal walls and partitions, and wall finishes.

6. A common criticism on the construction cost index is that the material prices are ‘list prices’ and discounts are ignored. However, the ONS report that they do attempt to collect the real transaction price for compiling their produce price indices and it is outside the scope of this paper to verify this. We are more concerned with the labour wage component of the construction cost index since it is generally believed that there is a variable time lag between the market conditions and the wages reflected in the national labour agreements. The average earning information collected by ONS in Annual Survey of Hours & Earnings and its predecessor, New Earnings Survey, would provide a more timely, but somewhat noisy, picture of the wages in construction industry.

7. In the PFI market, the special purpose vehicle (SPV), formed by a consortium in the private sector, enters a long period service provision contract with the public sector. The construction contract in the PFI project is made between the SPV and the building contractor.

8. A private firm, Carillion, continues to produce the PSA Schedule of Rates after the PSA dissolved. Please see their webpage for details: www.psa‐sor.com/.

9. The rates in the PSA Schedule of Rates complied by Carillion and other common building price books are compiled by using the historic labour constants (updated infrequently), labour wages, material and plant prices. The Schedule of Rates compiled by the then PSA is believed to have better captured the market rates because the Schedules are believed to have been tested against cost information of the large amount of projects procured by PSA. Davis Langdon also report that they have cross‐checked some of the rates in their price book against the rates in accepted BQs.

10. It is called ‘under construction’ as opposed to another index called one‐family house ‘sold’ price index that includes the land and other non‐construction related elements in the price which the house purchasers pay.

11. Other than this method, there are a few alternatives to construct hedonic price index. Triplett (2004) is a comprehensive study.

12. We thank Mr Jim Meikle for bringing our attention to these issues in this context.