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 >> Chapter 5: Price Index of Household Consumption Purchases


5.1. Section 4 developed the DFP price index model within the broader market transactions analytical framework. The DFP model is comprised of an aggregate, economy-wide DFP index along with its component indexes. The largest component index relates to HCP, estimated to account for 55–60% of the weight of the total DFP index (see Appendix 1). This component is illustrated in bold on diagram 5 in the previous section.

5.2. Under the DFP price index model, the ABS proposes to concentrate on the development of a price index of HCP designed specifically for the analysis of inflation.

5.3. Apart from the fact that HCP is the most important component of the DFP model, there are a number of other reasons for proposing this strategy, namely:

      • This component is of particular interest to those policy makers who view inflation from the household perspective and regard household consumption purchase prices as representing the ‘end of the chain’, capturing the final impact of price change within the economy.
      • Being based predominantly on existing CPI collections, early investigations indicate that this component would be the most viable to develop in the short to medium term. By contrast, the development of price indexes corresponding to the other consumption components and the capital components of the DFP model would be more complex. It would require data to be brought together from a number of different sources and modified, as well as involving additional data collection. An important area in which there is a lack of data is engineering construction activity, under capital purchases.
      • In terms of contributing to analysts’ knowledge and understanding of the inflation process, household consumption price measures are assessed as being likely to derive the most benefit from the application of statistical techniques designed to isolate underlying inflation (see section 6).

5.4. The development of indexes for the components other than HCP would be within a longer-term time frame. Nevertheless, for completeness, Appendix 1 contains indicative weighting estimates and contains the results of initial investigations into identifying possible data sources.

Relationship to the Consumer Price Index

5.5. Although widely used for other purposes, such as a measure of inflation, the CPI is designed primarily as a means of assessing changes in the purchasing power of wage and salary earner household incomes and, as such, has played an important role in the income adjustment process.

5.6. In recent years, there has been recognition of conceptual deficiencies in the CPI for inflation measurement, particularly for policy purposes. This has led to the development of alternative measures such as the index of underlying inflation as defined by the Commonwealth Treasury.

5.7. The main deficiencies in the CPI as a measure of inflation are the inclusion of interest rates, potential problems associated with the use of fixed weights in index construction and its restricted scope (metropolitan employee households).

5.8. The forthcoming periodic review of the CPI will provide an opportunity for, amongst other things, a review of the principal purpose of the index. A related ABS information paper to be released in April 1997 will identify the main issues to be addressed during the CPI review process.

5.9. The HCP price index would differ from the current CPI in the following ways:

      • interest rates (mortgage rates and consumer credit charges) would not be included as they do not directly represent market prices for actual transactions in goods and services;
      • the scope would be extended beyond the CPI target population to include all of the household sector;
      • the weights would be based on the national accounts rather than the Household Expenditure Survey;
      • the index would be likely to be an annually chained index; and
      • payments for goods and services which are purchased by households at non-market determined prices (e.g. public education and hospital services) would not be included.

5.10. The scope would be extended beyond the CPI target population by changing the weighting pattern so that it related to purchases by all Australian households, rather than just metropolitan wage and salary earner households (excluding the top decile). Weights corresponding to purchases by all householders would be obtained from the national accounts, with modifications to ensure consistency with the concepts of the DFP model. An index for HCP would then be calculated using appropriate disaggregated CPI price data.

5.11. Note that the HCP price index would differ from the national accounts measure, private final consumption expenditure, through the exclusion of notional transactions such as the imputed dwelling rent of owner occupiers, non-marketed goods and services provided to households and purchases by NPISH.

5.12. Below is a discussion on practical issues associated with the identification of market determined prices.

Market determined prices

5.13. As inflation is a phenomenon of markets, coverage of a HCP index should be confined to actual markets, where goods and services are exchanged at prices determined by the interaction of buyers and sellers. In comparison with private final consumption expenditure under the national accounting framework, this would require the exclusion of those transactions where, by convention, the SNA deems homeowners to rent dwellings from themselves as landlords.

5.14. Similarly goods and services which are acquired free of charge or at economically insignificant prices (generically referred to as non-market goods and services) should be excluded.

5.15. While the identification of wholly notional transactions included in the national accounts is relatively straightforward and unambiguous, the identification of non-market goods and services is more difficult. Therefore, in order to ensure consistency in treatment, a prerequisite is the development of suitably robust definitions of market and non-market determined prices. The following definitions, which concern economically significant prices within the context of market output, provide a useful starting point for assessing whether prices are market determined.

      • ‘Prices are said to be economically significant when they have a significant influence on the amounts that the producers are willing to supply and on the amounts purchasers wish to buy.’(System of National Accounts 1993, p. 128)
      • ‘By convention, in the ESA, an economically significant price (excluding taxes on products and including subsidies on products) should generally cover at least 50% of the production costs of the product.’ (Eurostat 1996, p.70.)

5.16. There are four general categories of goods and services which may not have market determined prices:
      • those which are provided by general government (e.g. defence, public administration, education and hospital services);
      • those whose production cost or prices are partly subsidised through government funding (e.g. pharmaceuticals);
      • those which have prices regulated or affected to a large degree by government policy (e.g. motor vehicle registration); and
      • those whose production costs are partly funded by unrequited donations or concessions from the private sector (in particular, goods and services produced by NPISH such as welfare services provided by charitable organisations).

5.17. The implications of differentiating between market and non-market goods and services are significant. Those non-market goods and services which are provided to households by government and NPISH would not be included in the HCP index. As the purchase of inputs to the production of non-market goods and services represents the final point at which these products are traded at market determined prices, the providers of non-market goods and services are regarded as the final purchasers, and their purchases would be within the scope of NPISH and General Government Consumption Purchases indexes. This represents a significantly different treatment between the HCP index and the CPI, which includes the actual costs incurred by households for all goods and services.

5.18. There are many goods and services where subsidies have played a role but the transaction price can still be viewed as being market determined. The prices at which these goods and services are purchased by households would therefore be reflected in the HCP index.

5.19. To illustrate some of the issues and complexities involved, consider the case of education services. Prices paid for education services provided by the public education system may not be market determined due to high levels of government subsidies. The cost of inputs into the public education system would therefore be used in the DFP measure and would be categorised as Government Consumption Purchases. Many non-government schools are NPISH, so if the prices of their services were assessed as being non-market determined, the prices of their inputs would feed into NPISH Consumption Purchases index. Other non-government schools, although in receipt of some government funding, may charge prices for their services which are market determined. These final transactions would then be included in the HCP index.

5.20. Areas which would need to be examined closely before decisions can be made include public transport fares, medical services, pharmaceuticals, rents for government-owned dwellings, local government services, tertiary and private education, and government provided child care. In some cases, these items may need to be studied at a commodity, regional and/or enterprise level, as there could be significant variation in the rate of subsidies paid when items are disaggregated.

Previous PageNext Page