10/09/2007: The data cube associated with Information Paper: Experimental Estimates of Industry Multifactor Productivity, 2007 (5260.0.55.001) has been re-issued because of an error in Table 8 Capital Rental Price: $cost per unit of 2004-05 capital - (Incorporated and Unincorporated).
This information paper provides experimental estimates of industry level multifactor productivity (MFP). It is an extension of the ABS's research into productivity that began back in the 1980s. Estimates of MFP for the market sector were first published in 1985. Since that time, there has been growing interest in productivity measurement, particularly in the productivity performance of individual industries. The ABS commenced research into industry level MFP estimates in 2002. To assist this work the ABS established a Productivity Measurement Reference Group to advise on conceptual and practical issues involving productivity measurement. This paper has benefited from discussions with the Reference Group.
There are a number of conceptual and data issues surrounding productivity measurement. Statistical agencies and others involved in the field are still gaining experience in their compilation. As such, the industry level estimates contained in this paper should be considered experimental.
The paper discusses the general concepts and methodology for measuring MFP and then presents a brief summary of results. Much of the paper is centred around individual industry analyses of the estimates. Future directions are outlined at the end of the paper. The indexes that comprise the industry estimates are available on the ABS website at <www.abs.gov.au> in a spreadsheet format (cat. no. 5260.0.55.001).
The ABS plans to publish industry MFP estimates on an annual basis and is seeking user feedback regarding the estimates and the methodology used in their calculation. Please forward any comments to Paul Roberts (firstname.lastname@example.org).
The ABS has been producing multifactor productivity (MFP) estimates for approximately 20 years. Considerable development work took place during the 1980s leading to the publication of the first estimates of MFP in 1985. Since then, MFP estimates for the market sector as a whole have been produced each year and released in conjunction with the annual national accounts.
Over recent years, interest in productivity measurement has increased significantly with particular interest in the productivity performance of individual industries and the Australian economy relative to other countries. In response to this, the ABS commenced a project in 2002 aimed at developing industry level estimates of MFP. Estimates were released in a research paper in 2005 (Zheng 2005) and confirmed that while there were some limitations to the available data, meaningful estimates could be developed. Since then research has been undertaken to further develop and extend the estimates.
The focus of this paper is to present experimental estimates of MFP for the 12 industries that comprise the market sector. The paper presents an analysis and discussion of the results for each of the industries. The estimates of industry MFP presented in the paper use the same basic data that underlie the estimates of market sector productivity published in the Australian System of National Accounts, 2005-06 (cat. no. 5204.0). However, one difference of note between the industry level estimates and the aggregate market sector estimates is the treatment of taxes on products (sales tax, excise tax and GST). These are excluded from the individual industry estimates but accounted for in the published aggregate.
In developing these industry estimates a number of matters were identified as requiring further research. These research topics are identified in the industry chapters and summarised in Chapter 15 together with other general research topics. These topics provide a basis for future ABS work in this field.
The results presented in this paper should be interpreted with care, particularly with regard to short term movements. The methodology used in compiling the estimates implicitly assumes that the proportion of capital stock used in production (capital utilisation) does not change, therefore any real world change in the extent to which capital is utilised in production will be recorded as a change in productivity. The methodology further assumes that each hour of labour input is fully utilised in production. Further, improvements in output due to a firm's ability to produce more output as a result of their size, that is, economies of scale, will also appear as a measured productivity improvement.
The paper is structured as follows. The general concepts and methodology for measuring MFP are discussed in Chapter 1. Chapter 2 presents a brief summary of results comparing the industry estimates to the published market sector estimates, as well as comparing gross output based MFP estimates to value added based MFP estimates. The following chapters contain the individual industry analyses. The final chapter provides some concluding remarks, as well as identifying areas for future research. Appendix 1 provides further details on the methodology and data. Appendix 2 presents a sensitivity analysis on possible methodological approaches to estimating capital services.
The ABS plans to publish industry MFP estimates on an annual basis and is seeking user feedback regarding the estimates and their methodology.