|Page tools: Print Page Print All|
5.17 Households acquire a large number of different goods and services. It is not practical or necessary to price all the goods and services acquired by the CPI population group. Many related items are subject to similar price changes and households acquire more of some items than others. Rather, the items selected for pricing in the CPI are the more significant ones and are likely to have price movements that are representative of a wider range of goods and services.
5.18 Items are selected using purposive (or judgmental) sampling. To be included in the basket, goods and services must be representative of purchases made by the CPI population group and have prices that can be associated with an identifiable and specific commodity. Items are not excluded from the CPI basket on the basis of moral or social judgements. Some commodities that may be considered socially undesirable (e.g. tobacco) are included in the CPI basket because they are significant items of household expenditure and their prices can be accurately measured.
5.19 The CPI is compiled separately for each capital city. For general statistical purposes the equivalent of an all Australia index is the series published as the weighted average of eight capital cities.
5.20 The geographical coverage of each capital city index is the acquisitions of goods and services by the resident population of that city. Hence, in concept the CPI includes all acquisitions by capital city residents, from outlets in their capital city, other capital cities and regions of Australia and abroad.
5.21 By and large, it is expected that most of the acquisitions made by households in the capital cities will be from providers of the goods and services that are located in the same city. The most obvious exception is holiday accommodation services. The acquisition of goods by mail order or over the phone or internet from outlets within and outside the capital city of residence is considered to be relatively small. However, where transactions made by such methods are known to be significant (as is the case with airfares and holiday accommodation purchased on the internet), prices are collected from these sources.
5.22 For the main part the CPI captures expenditures of capital city households regardless of where those expenditures are incurred, but uses prices applying in the capital city of residence. However, for domestic holiday accommodation, prices are included from outside the capital city of residence of the household.
STRUCTURE OF THE AUSTRALIAN CPI
5.23 A diagrammatic overview of the structure of the CPI is provided in figure 5.1. The structure of the CPI can best be thought of from a top-down approach. At the top is the total expenditure or pool of items purchased (the All groups level). This is progressively divided into finer commodity groupings until, at the lowest level, there are samples of prices for the individual items (elementary aggregates). Indexes are only published down to the expenditure class level as this is the level at which the structure and weights are fixed for the life of a CPI series.
5.1 CPI STRUCTURE
5.24 This same structure is used for each of the eight capital cities. A full list of groups, subgroups and expenditure classes is provided in the Appendix 1.
5.25 The division of the items into finer product groupings, is intended to reflect increasing levels of substitutability of the items in consumption by households in response to changes in relative prices. For example, at the group level there is likely to be little substitution between, say, food and transportation groups in response to changes in their relative prices. However, within the Fats and oils expenditure class it would be expected that households are more likely to substitute between, say, margarine and butter in response to changes in their relative prices.
5.26 The structure of the Australian CPI in large part follows that of the Classification of Individual Consumption by Purpose (COICOP) which is also the basis of the European Union’s Harmonised Indices of Consumer Prices (HICP). (footnote 10) However, there are some important differences.
5.27 The 14th series CPI basket is divided into 11 major groups, each representing a specific set of commodities:
5.28 These groups are divided in turn into 34 subgroups, and the subgroups into 89 expenditure classes.
5.29 Presentation of the CPI in the form of groups and subgroups provides the user with quite a degree of versatility in interpreting the results. Index numbers for individual groups and subgroups can be analysed separately as can their individual effects on the overall index.
5.30 The All groups index represents the highest level of the index and it is commonly referred to as the ‘headline’ rate of inflation.
WEIGHTING THE INDEX
5.31 The overall (or All groups) CPI is compiled by weighting price movements (or price relatives) between the base and current period by their shares of total household expenditure in the base period. This is simply the alternative way of calculating a Laspeyres index as described in paragraphs 3.24 and 3.25.
5.32 In practice the CPI is compiled using ‘expenditure aggregates’ rather than expenditure shares. The expenditure aggregate for an item in period t is calculated by multiplying the base period expenditure aggregate ( ) by the price relative for period t ( ). In effect this is simply the product of the (unobserved) base period quantity and the period t price. Summing the expenditure aggregates in period t and dividing by the sum of the expenditure aggregates in the base period yields the Laspeyres price index.
5.33 Hence the CPI is described as measuring the change over time in the total price of a fixed basket of goods and services purchased by households in aggregate. It is important to note that the use of the term ‘fixed’ relates to the quantities underlying the base period expenditures - it is, after all, the base period quantities that are fixed in a Laspeyres index. Weights are usually expressed in terms of expenditure shares because quantities are not meaningful or consistent across commodities and services. Further, the expenditure shares will change over time as the rate of price change varies across commodities.
5.34 The description of the CPI as a fixed-weight index requires qualification reflecting the practices adopted by ABS.
5.35 Holding the weights of goods and services in the CPI basket fixed permanently is neither realistic nor entirely possible. If held constant on a permanent basis, the weights would become less representative of the relative importance of goods and services purchased by households the further the series moved away from the base period. There would also be the problem of items that cease to exist and the entry of new goods and services. Furthermore, the finer the level of detail the less information that exists about the relative importance of items in the basket, which makes it difficult to calculate weights at lower levels of the index.
5.36 To reduce these problems, weighting practices vary by the level of aggregation:
Derivation of the CPI weights
5.37 The underlying quantity weights for the CPI expenditure classes are updated at approximately five yearly intervals with timing generally linked to the availability of Household Expenditure Survey (HES) data. The 14th series CPI was introduced in the September quarter 2000 using expenditure weights from the 1998-99 HES.
5.38 The weights employed below the expenditure class level are adjusted at irregular intervals according to information obtained from a variety of sources. These sources include data obtained from both within and outside the ABS (e.g. the publication Retail World provides information about food lines and supermarkets). Industry consultation provides a rich source of information for weight adjustment purposes. Manufacturers, importers and retailers are interviewed to obtain information about current turnover, expected future trends and new products introduced. The media also represents a major source of information for monitoring trends.
Price index formulas
5.39 The index formulas used in the CPI can be considered at two levels: the formulas applied to calculate the average price at the price sample or elementary aggregate level, and the formulas used to derive index numbers at higher levels.
5.40 As already noted, there are no explicit weights for the prices in each elementary aggregate. The primary concern, is to establish the ‘best’ average price for the sample. The (unweighted) geometric mean (GM) formula is the preferred formula for calculating the lowest level indexes primarily because it models, better than alternative formulas, substitutions that consumers make in their consumption of goods and services in response to changes in relative prices.
5.41 The geometric mean cannot be used to calculate the average price in all elementary aggregates. It cannot be used in cases where the price could become zero (e.g. the cost of a good or service becomes fully subsidised by the government). It is also not appropriate to use the geometric mean in elementary aggregates covering items between which consumers are unable to substitute (e.g. local government rates where it is not possible to switch from a high rate council to a low rate council without physically moving locations). For these elementary aggregates the Relative of Arithmetic Mean Prices (RAP) formula is used. For further information on these formulas, refer to paragraphs 3.37 to 3.41.
5.42 It is important to distinguish between the weighting period and the reference base of the CPI. The weights for the CPI are derived from household expenditure surveys and other sources. As it takes a long time to collect and process expenditure data, there is usually a considerable lag before such data can be introduced into the calculation of the CPI. The current weighting base period (the underlying quantities) is 1998-99, while the reference base (the year in which the index value is set to 100.0) is 1989-90. (footnote 11) Further, because of the time lags involved in collecting the data required for reweighting, the 1998–99 expenditure weights were introduced in respect of the June quarter 2000.
5.43 The practical use of index formulas in the compilation and aggregation of the CPI is discussed further in Chapter 9.
1. Although there is no universally agreed definition of inflation, it is typically described as the phenomenon of an increase in the general level of prices over time. Strictly speaking the CPI measures price change, that is, both rises and falls in prices.< Back
2. A fundamental change was made in the principal purpose of the CPI with the introduction of the 13th series in the September quarter 1998. Prior to this period, the principal purpose of the CPI was as an input to wage determination processes. After extensive public consultation for the 13th series CPI review, it was determined, on balance, that the needs of the Australian community would best be served by changing the principal purpose to a measure of price inflation. In practical terms, the major effect of this change was dropping interest charges and including house purchase in the CPI. For further information refer to Information Paper: Issues to be Considered during the 13th Series Australian Consumer Price Index Review, (cat. no. 6451.0) and Information paper: Outcome of the 13th Series Australian Consumer Price Index Review, (cat. no. 6453.0). < Back
3. Between 1993 and the introduction of the 13th series, a subset of the CPI was used as the target. Commonly referred to as the ‘Treasury measure of underlying inflation’, this index excluded items from the CPI that were considered by the Treasury and RBA to be unrepresentative of the ‘core’ or underlying rate of inflation in that their prices were highly volatile, exhibited marked seasonal patterns or were largely affected by policy decisions. < Back
4. For a discussion of the use of the CPI in the conduct of monetary policy, refer to the Reserve Bank Bulletin (October 1998) article ‘The Implications of Recent Changes to the Consumer Price Index for Monetary Policy and the Inflation Target’. < Back
5. See paragraph 26 of Information paper: Outcome of the 13th Series Australian Consumer Price Index Review (cat. no. 6453.0). < Back
6. See ‘Analytical Living Cost Indexes for Selected Australian Household Types’, Australian Economic Indicators, June 2001 (cat. no. 1350.0). < Back
7. Source: Household Expenditure Survey, Australia 1998-99 (cat. no. 6530.0). < Back
8. In addition, information would be required on the locations where the households purchased the goods and services (often referred to as ‘point of purchase’).
9. The ABS has developed experimental price indexes for deposit and loan facilities and real estate. They are published each quarter in Producer Price Indexes, Australia (6427.0). The ABS plans to incorporate these measures in the 15th series CPI, scheduled for release in October 2005. < Back
10. For information on the HICPs, refer to Report from the Commission to the Council on Harmonisation of Consumer Price Indices in the European Union, Eurostat, Catalogue number KT-CO-00-176-EN-C. < Back
11. Some components have a reference base later than 1989-90, reflecting their introduction to the CPI since the 11th series CPI. < Back