5220.0.55.002 - Information paper: Gross State Product using the Production approach GSP(P), 2007  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 14/09/2007  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All

PREFACE

Since 1987 the Australian Bureau of Statistics (ABS) has published annual estimates of Gross State Product (GSP) as part of the Australian National Accounts: State Accounts (cat. no. 5220.0). Over recent years there has been an increased effort applied to improving and expanding the quality of the State Accounts. In 2003, the ABS established a specialist team with a focus on the State Accounts.


Compiling GSP using the Production approach (GSP(P)) was the major project on the research program. Another aspect was the establishment of a State Accounts User Group (SAUG) in 2004. Since 2004 consultation and discussion has taken place with SAUG on the development of the GSP(P) volume estimates.


This information paper provides the results of the GSP(P) project. Results are presented for each state followed by a discussion on Gross Value Added (GVA) by industry (including Ownership of dwellings and Taxes less subsidies on products) for each state.


The GSP(P) estimates are compared to the current official volume estimates of GSP growth which are based on the Income/Expenditure approach (GSP(I/E)). The GSP(P) estimates contained in this information paper are considered indicative.


The compilation of the GSP(P) estimates means that there are now alternative measures of economic activity available for each state. There are three possible volume measures of GSP that the ABS could publish as the headline measure of economic growth:

  • the new GSP(P) based volume estimates
  • the current official GSP(I/E) based volume estimates
  • an average of these volume estimates, described as GSP(A).

In considering the merits of the various options the ABS, in consultation with the SAUG, concluded that the average measure is preferred. The ABS considers this measure maximises the use of information about state economic activity and that it will be more stable over time (i.e. subject to smaller revisions) than the two alternatives. This approach is also consistent with the approach used nationally for the latest year estimates and for the quarterly national accounts.



INTRODUCTION


INTRODUCTION

The ABS has published estimates of GSP as part of the Australian National Accounts: State Accounts (cat. no. 5220.0) on a regular basis since 1987. Since this time there has been ongoing work to improve the estimates of GSP and expand the amount of information contained within the State Accounts. The main improvements to the State Accounts include:

  • 1988-89 - introduction of Market Prices
  • 1991-92 - introduction of State Final Demand
  • 1993-94 - constant price estimates introduced using prices at 1989-90 and Industry structure changed to the Australian and New Zealand Industrial Classification, 1993 (ANZSIC93) (cat. no. 1292.0.15.001) basis
  • 1997-98 - implementation of accounts based on the System of National Accounts, 1993 (SNA93) from 1989-90 onwards and change from constant price estimates to chain volume measure estimates.

Over recent years there has been an increased effort applied to improving and expanding the quality of the State Accounts. In 2003, the ABS allocated additional resources to establish a specialist team with a focus on the State Accounts. Investigating the possibility of compiling GSP(P) was the major project on the research program. Another aspect was the establishment of a SAUG in 2004. Since 2004 consultation and discussion has taken place with the SAUG on the development of the GSP(P) volume estimates.


This information paper provides the results of the GSP(P) project. Results for each state are presented, followed by a discussion on GVA by industry (including Ownership of dwellings and Taxes less subsidies on products) for each state. Comparisons of GSP(P) are made to the current official volume estimate of GSP growth which are based on GSP(I/E). The GSP(I/E) volume estimates are derived by deflating current price GSP compiled using the income approach (GSP(I)), with a deflator compiled using the expenditure approach (GSP(E)).



HEADLINE MEASURES OF GSP

The introduction of GSP(P) estimates means that there are now alternative measures of economic activity available for each state. There are three possible volume measures of GSP that the ABS could publish as headline measure of economic growth:

  • the new GSP(P) based volume estimates
  • the current official GSP(I/E) based volume estimates
  • a simple average of these volume estimates, described here as GSP(A).

In considering the merits of the various options the ABS, in consultation with SAUG, concluded that the average measure is preferred. The ABS considers this measure maximises the use of information about state economic activity and that it will be more stable over time (i.e. subject to smaller revisions) than the two alternatives. This approach is also consistent with the approach used nationally for the latest year estimates and for the quarterly national accounts.


It is planned to publish GSP(A) as the headline measure in the Australian National Accounts: State Accounts, 2006-07 for the period from 1989-90 onwards. The volume estimates published on the GSP(I/E) and GSP(P) bases would, therefore, contain statistical discrepancies to reconcile them with GSP(A).


The development of the volume estimates of GSP(P) will also have an impact on the presentation of the current prices estimate of GSP. The volume measure of GSP(P) will be reflated using the GSP(E) deflator to produce a current price GSP(P). This will then be used with the existing current price GSP to calculate a simple average of GSP, estimated using the income approach, in current prices. This is will result in a statistical discrepancy for the individual measures of current price GSP.



EXPERIMENTAL STATUS OF ESTIMATES

The GSP(I/E) estimates have been regarded as experimental estimates since their inception due to issues relating to the measurement of Interstate trade and Changes in inventories. The development of GSP(P) has allowed an assessment of the current price GSP(I) results. Both the volume GSP(P) and volume GSP(I/E) generate similar outcomes. Consequently the ABS considers the volume estimates to be sufficiently robust, enabling the experimental status be dropped. The robustness of the estimates is also reinforced by the use of GSP(A) as the headline measure. Thus, from the 2006-07 publication of Australian National Accounts: State Accounts, the headline GSP volume estimates will no longer be labelled as experimental.


Nevertheless, users should be aware that the State Accounts estimates are likely to be of lower quality than the equivalent national level estimates. One reason is the inherent problems associated with the allocation of multi-State activities, especially in industries such as long distance transport, communication and finance. Another reason is the data sources are generally sample surveys designed to optimise quality at the national level, not the state level. This is likely to impact more on the quality of data for the smaller states and territories.



STATUS OF THE ESTIMATES IN THIS PAPER

The GSP(I/E) estimates shown in this paper are those published in the 2005-06 issue of Australian National Accounts: State Accounts, which are the most recent official estimates. It is expected that these estimates will be revised in the 2006-07 issue of 5220.0 primarily due to updated Australia level benchmarks from the annual supply and use tables and the availability of additional data sources.


The GSP(P) and GSP(A) estimates shown in this paper should be regarded as indicative. They are not official ABS estimates. Their purpose is to demonstrate the GSP(P) approach. While the ABS has taken care in developing these estimates, they are likely to be revised in the 2006-07 issue of 5220.0, at which stage they will become official estimates.



STRUCTURE OF PAPER

The remainder of the paper is structured as follows:

  • Section 1 outlines the methodologies used to derive GVA at the Australia level compared with the states and territories, and a comparison of the various GSP measures. It includes a comparison of the levels and growth rates of the three measures of GSP (GSP(P), GSP(I/E) and GSP(A)) and other additional analysis for each state/territory.
  • Section 2 presents a discussion of the structure of the industries under ANZSIC93 and a detailed description of methods and results for each industry.
  • Section 3 contains some concluding remarks and outlines changes to the format of the 2006-07 publication of Australian National Accounts: State Accounts. Areas requiring further research and development are also presented.

The inclusion of the GSP(A) and GSP(P) estimates is a major change in State Accounts and, as such, the ABS would be interested in receiving any feedback on these estimates and their inclusion in Australian National Accounts: State Accounts. Please contact Donna Grcman (email donna.grcman@abs.gov.au or telephone (02) 6252 5892) if you have any comments or inquires about the proposed approach.