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 3: An analytical framework for inflation measurement


3.1. As described in section 2, there is a wide range of partial price indexes available in Australia, presented separately as specialised, stand-alone measures. This approach places a constraint on optimising the analytical use that can be made of the large volume of price data that is collected on an ongoing basis.

3.2. The analytical value of this price information could be significantly enhanced through the ABS drawing the series together and presenting them as a system or family of price indexes. This system would be designed around a cohesive statistical framework developed specifically to support the study of inflation. The use of such a framework would facilitate the examination of the relationships between different price measures and assist analysts to understand and interpret inflationary signals.

3.3. The framework would also accommodate future developments in the field of price statistics such as extensions in coverage of the economy and the presentation of data under alternative classification systems.

Ideal properties

3.4. Despite universal usage of the term ‘inflation’, there is no generally agreed definition that is sufficiently precise to support the development of a statistical framework. The measurement of inflation is a complex issue and it is generally accepted that no single price index can measure all aspects of inflation or meet the needs of all users.

3.5. Nevertheless, despite the lack of a strict macro-economic definition of inflation, there is some convergence of views as to the ideal conceptual properties that would be possessed by a system of price indexes designed for the analysis of inflation. These can be summarised as below:

      • it would encompass only market transactions. That is, government services which are not marketed, and notional transactions such as those where homeowners are deemed to rent dwellings from themselves as landlords, would not be included;
      • it would capture the inflationary trend in prices associated with transactions in goods and services; accordingly, it would not include interest rates;
      • conceptually, the framework would embrace the entire economy and not be restricted to particular segments;
      • it would provide for broad, economy-wide price indexes as well as component indexes, potentially presented under alternative classifications;
      • the framework would provide for taking alternative, complementary views of the economy;
      • it would relate to pure price change and incorporate very recent weighting information;
      • from an analytic viewpoint, the effects of changes in government charges and taxes would be capable of separate analysis, and the effects of erratic price fluctuations would be identifiable; and
      • the price indexes would be non-revisable, and provide certainty to users.

3.6. Two possible broad approaches for developing an economy-wide system of price indexes are the national accounts approach and the market transactions approach.

National accounts approach

3.7. As described in section 2, the national accounts approach utilises price information generated during the construction of the national accounts.

3.8. The strength of the national accounts approach lies in the fact that the System of National Accounts (SNA) represents the only comprehensive and detailed framework for the systematic and integrated recording of the stocks and flows of an entire economy. As such, most users of economic statistics are familiar with its underlying principles and the major aggregates.

3.9. The national accounts approach involves two sets of price measures: IPDs and FWIs, both of which represent economy-wide price measures. However, they possess conceptual limitations as described below.

3.10. The IPDs do not compare the price of a constant basket of goods and services between any two periods (except when comparing the base period and any other period) because the weights change from period to period. As a result, period-to-period movements in the IPDs do not measure pure price change.

3.11. The FWIs of GDP, exports of goods and services, imports of goods and services, and of DFD and its four major components provide a measure of pure price change and generally conform with several of the conceptually desirable properties for an economy-wide system of price measure.

3.12. However, they do include notional transactions, they do not directly provide for the analysis of government taxes and charges and they have weights which remain fixed for a long period of time. Further, both the IPDs and FWIs are restricted to a single view of the economy, that is, from the final expenditure perspective.

3.13. Overall, the national accounts have not been designed with the specific intention of price measurement. An additional practical problem with this approach is that the price measures are subject to revisions as firmer data are included in the accounts.

Market transactions approach

3.14. A market transactions approach represents a relatively new proposal being considered by a number of international experts as a means of overcoming the perceived deficiencies of the national accounts approach to the measurement of price change.

3.15. The fundamental rationale for a market transactions approach is the view that inflation in an economy is a phenomenon peculiar to the operation of markets and thus price measures designed for the analysis of inflation should be confined to market activity; that is, non-market transactions (e.g. non-traded goods and services) have no role. The significance of this concept compared with the strict national accounting approach lies in its focus on only actual transactions and their respective prices.

3.16. The market transactions approach has been developed by considering the issue of price measurement from a micro-economic perspective. From this perspective, it can be argued that it is possible to define a price measure for each institutional unit (e.g. each household, corporation, or government body) in terms of the goods and services actually purchased by each unit in the market place. In other words, each unit’s inflationary experience is determined by its own unique purchasing patterns. Individual price indexes could therefore be computed as a function of prices and quantities of the goods and services purchased by each unit. In concept, an economy-wide system of price indexes could involve aggregating all the individual measures, using weights which reflect the aggregate value of their purchases, to build up a total economy-wide price measure.

3.17. While this micro-economic perspective represents the theory underpinning the market transactions approach, the notion of actually computing measures for each institutional unit is obviously not feasible. Even if it could be done, the interpretation and usefulness of a single, summary price measure based on aggregation across the units which included both intermediate and final transaction prices would be dubious because of multiple counting.

3.18. In order to provide a better framework for the analysis of price change, the individual transactions are conceptually allocated to specific ‘markets’, rather than specific institutional units, to arrive at the classification presented schematically in diagram 4. The diagram provides a broad level representation of all market transactions where one or both of the parties to the transaction are located in Australia. Each of the specific markets is shown in the lower part of the diagram, with summary aggregations of the markets in the upper part.

        Market Transactions View of the Economy
(a) Non-Profit Institutions Serving Households.
3.19. In theory, each of the markets shown on the diagram could be represented by a price index. However, the economic meaning of price indexes corresponding to such aggregations of market activity as ‘Total domestic purchases’ and ‘All transactions’ is questionable and they are shown here for illustrative purposes only.

3.20. Rather, it is more meaningful to consider the market transactions framework as providing alternative, but complementary views of the economy through different markets.

3.21. Section 4 discusses a DFP price index model corresponding to the ‘Domestic final purchases’ market shown in the analytical framework illustrated in diagram 4.

Previous PageNext Page