6421.0 - Information Paper: An Analytical Framework for Price Indexes in Australia, 1997  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 24/02/1997  Ceased
   Page tools: Print Print Page Print all pages in this productPrint All  
Contents >> Appendix 1: Indicative weighting pattern and potential data sources for other Domestic Final Purchases component indexes


A1.1. As noted in section 5, the ABS proposes to focus on the development of an index relating to the HCP component of DFP. However, for completeness, this Appendix contains a summary of the results of initial investigations undertaken to derive indicative weighting estimates and identify potential data sources for the possible future development of indexes for the other DFP components, namely NPISH Consumption, General Government Consumption and Capital Purchases.

Indicative weighting pattern

A1.2. The weights for the DFP measure would be based on the share of the total value of purchases for each purchase type. They would be based primarily on national accounts data valued at purchasers’ prices (i.e. including taxes, and transport and trade margins), but significant adjustments would need to be made to account for the conceptual differences between the national accounts and DFP approaches. The different treatment of notional transactions and non-market output would be particularly important.

A1.3. Graph 8 provides crude estimates of these weights. The estimates are based on the national accounts measures; private final consumption expenditure, government consumption expenditure, and gross fixed capital expenditure, adjusted to exclude imputed dwelling rent, imputed bank/insurance service charges and depreciation. Most significantly, the exclusion of the imputed dwelling rent of owner-occupiers reduces private final consumption expenditure by 10–15%. Note that NPISH Consumption Purchases are included as part of HCP in graph 8, in accordance with current national accounts practices. NPISH Capital Purchases are included in Private Corporate Capital Purchases.

A1.4. Note that a large number of conceptual differences between the national accounts and the DFP approaches have not been accounted for in developing these weighting estimates.

A1.5. It can be seen that HCP would be the most significant component of the DFP index, having a weight of 55–60%. The weight of input purchases by NPISH is very small (2–3%), while the Government Consumption Purchases component has an estimated weight of 15–20%. The Private Corporate Capital Purchases price index would have a weight of 10–15%, while Household Capital Purchases and Public Sector Capital Purchases would each have a weight of about 5%. The fluctuations in the shares of these components, especially the capital components, over the business cycle indicate that a chain index is likely to be preferable to a long-term, fixed-weighted index.

        Weighting Estimates
        Source: Based on published and unpublished ANA data.
A1.6. As previously noted, the DFP price index would preferably incorporate weights based on very recent purchasing patterns. Determining the precise form the index should take would require further investigation, but in general terms a chained measure would be the preferred solution.

Potential data sources for other DFP component indexes

NPISH Consumption Purchases

A1.7. The second consumption component within the DFP model is the purchases of inputs by private NPISH such as charities, conservation groups, unions, churches, professional and sporting associations. NPISH provide educational, health, cultural, recreational and other social community services to households free of charge, or at prices which are not economically significant.

A1.8. The market transactions measures would only cover the purchases of inputs by those non-profit institutions engaged in non-market production. Purchases by non-profit institutions controlled and mainly financed by government would be included in the Government Consumption Purchases component, while the input purchases of non-profit institutions serving businesses are intermediate purchases, not final transactions.

A1.9. There are no existing price indexes designed to measure input prices for purchases by NPISH. In fact, the ABS publishes very little information about NPISH. This lack of data may be addressed in the future due to the System of National Accounts 1993 (SNA93) recommendation of a separate institutional sector for NPISH within the national accounts framework. However, there are serious practical concerns regarding both the availability of data and the likely data quality for what is a very small sector (2–3% of DFP).

A1.10. In the medium term, it may be possible to develop an input price index for NPISH based on existing price indexes and the LCI, with weighting information derived from the national accounts.

General Government Consumption Purchases

A1.11. A Government Consumption Purchases price index would represent all government input purchases associated with non-market goods and services. The main subcomponent would be a government LCI, while material and service input purchases (e.g. office supplies, electricity, consultancy fees) would also be covered. The national accounts could be the source of weights for this price index with modifications to ensure consistency with the market transactions approach.

A1.12. While the national accounts measure of government final consumption expenditure aims to value government output, in practice it is expenditure on inputs which is measured. The coverage of the Government Consumption Purchases component would differ from that of government final consumption expenditure in that the former measure would also include inputs associated with the production of non-market capital goods by the government. Consumption of fixed capital (a depreciation allowance) would also be excluded from the Government Consumption Purchases component. Presently, the national accounts treats all defence expenditure as part of government final consumption expenditure. Ideally, and in accordance with SNA93, purchases of some durable military equipment and structures would be considered capital items.

A1.13. The Government Consumption Purchases price index would be based on input price indexes similar to those used to deflate government final consumption expenditure in the national accounts. The government LCI would be the most important contributor.

A1.14. Price indexes corresponding to other government input purchases could be obtained from existing consumer and producer price data. However, there is little data on the prices of services purchased by government. While the ABS has commenced developing a more comprehensive set of producer price indexes for services, this data gap will only be filled in the long term. Price movements for many service inputs would need to be imputed.

Capital Purchases

A1.14. A price index covering all purchases of newly produced and imported capital goods by domestic institutional units would be an important component of the DFP measure. An existing capital item is one which has already been acquired by a domestic user and whose value has already been included as part of Capital Purchases. All other capital items are considered ‘new’, including purchases of second-hand imported capital goods which are being purchased by a domestic resident for the first time. Price indexes corresponding to purchases of new capital goods could be developed for the following four institutional sectors:

      • Households;
      • Private Corporations;
      • NPISH; and
      • General Government and Public Corporations (Public Sector).

A1.15. National accounts capital expenditure data could be used as a weighting source with modifications to ensure the weights were in accordance with the DFP approach. It would be desirable for the Capital Purchases component to be able to be broken down by asset type, as itemised in SNA93.

A1.16. The Household Capital Purchases price index would consist mainly of purchases of new dwellings. Purchases of motor vehicles, computers and other durable goods by households are treated as consumption purchases rather than capital purchases.

A1.17. The Private Corporate Capital Purchases component is the most significant contributor to Capital Purchases and would include most of the asset types. The NPISH Capital Purchases component accounts for less than 1% of GDP but, like the Public Sector Capital Purchases component, would comprise a range of asset types.

A1.18. The concept of a Capital Purchases measure is broader than the national accounts gross fixed capital expenditure which currently excludes net acquisitions of valuables. The Capital Purchases price index would also differ from gross fixed capital expenditure in the treatment of non-market (including own-account) capital goods production.

A1.19. Even in the long term, price indexes may not be available for some types of capital purchases, and indexes may need to be imputed.

A1.20. A price index for the purchases of new dwellings would be based on the Project Homes Price Index published in ABS Cat. No. 6416.0, which would be the main contributor to the Household Capital Purchases component. Price indexes derived by Australian Construction Services relating to non-house dwellings such as townhouses, duplexes and apartments could also contribute to the Household Capital Purchases price index.

A1.21. The price index for purchases of new non-dwelling buildings could be based on output price indexes derived by Australian Construction Services for each State. This data is used to deflate the relevant components of the national accounts gross fixed capital expenditure.

A1.22. There is a significant gap in the availability of price data for engineering construction. The ABS is undertaking a study to assess the feasibility of establishing price indexes relating to the output of this industry. In the meantime, some relevant input price indexes relating to the construction of roads, railways, dams, airports, etc. are available, and are currently used to deflate this component in the national accounts.

A1.23. The price index for purchases of new machinery and equipment could be based on data from the IPI and APMI. The index would be a weighted average of the prices of domestically produced and imported machinery and equipment. Ideally, the valuation basis of the producer price data would be adjusted from basic prices to purchasers’ prices. However, this would not be a simple procedure and considerable time and effort would be required before such measures could be developed, if indeed they were warranted.

Previous PageNext Page