Coverage and classifications

5.1 The set of goods and services included in a Consumer Price Index (CPI) is called its item coverage, or more commonly, the CPI basket. In concept, all consumer goods and services are within scope of the index.

5.2 As households acquire many different goods and services, it is not practical or necessary to price all the goods and services that they buy. Many related items have similar price movements, and households acquire more of some items than others. The items selected for pricing in the CPI are purchased by a large proportion of households, and are likely to have price movements that are representative of a wider range of goods and services.

5.3 When determining what items to price, these factors are taken into consideration. The items: 

• must be representative of purchases made by the CPI population group (see paragraphs 5.19-5.22, Demographic coverage);
• must be identifiable and specific commodities or services (e.g. 420g can of baked beans from a supermarket, or adult general admission to a club football game); and
• are not excluded because of moral or social judgements.

5.4 The exclusions and inclusions are discussed more fully in paragraph 5.10 Illegal or undesirable goods and services.

5.5 As a general principle, a CPI only includes goods and services that are purchased by households for consumption. A consumption good or service is one from which households directly derive utility or satisfaction. Any business-related purchases by households are excluded from the basket, as are those items that have a significant savings or investment component, such as land and capital goods. All types of income are generally excluded as well, except those which directly offset a specific purchase, such as subsidies or trade-ins.

5.6 The prices of consumer goods and services, and the ability of households to purchase those items, are affected by a wide range of taxes, regulatory processes, and assistance measures. The treatment of these under the acquisitions and outlays approaches are similar, but there are differences under a cost-of-use approach.

5.7 As a general principle, the acquisitions and outlays approaches only include taxes and subsidies whenever they are tied to the level of consumption of a specific good or service. Thus any taxes based solely on income will generally be out of scope, whereas the prices of goods and services will be inclusive of indirect taxes and commodity-specific subsidies. In some cases, taxes and governmental charges may not be directly related to the level of consumption of a good or service. However, they may still be included if they are an inescapable cost of other decisions made by households about consumption. For example, local government rates and charges are an inescapable cost of home ownership, and so are included in a CPI.

5.8 A cost-of-use approach is concerned with the true value of goods and services consumed. For example, it will value subsidised items at their full market value. It will also exclude income taxes.

5.9 In concept, both the purchases and sales of second hand goods should be included in a CPI. The purchases of second hand goods by households are regarded as positive expenditures, while the sales of second hand goods by households are regarded as negative expenditures. The exact treatment of second hand goods will also depend on the nature and extent of transactions with other sectors of the economy. In practice, all transactions involving second hand goods are assumed to occur within the household sector, with purchases and sales cancelling out to give an effective weight of zero. The exception for this is motor vehicles where household expenditure on ex-business and ex-government motor vehicles, plus the dealer margin on used motor vehicles are included in the weights for motor vehicles. As there is difficulty in obtaining ongoing prices to constant quality for second hand goods, prices for second hand goods are not collected for the CPI.

5.10 In principle, all purchases of goods and services for household consumption are in scope of a CPI. They include goods or services that are either illegal or may be considered socially or morally undesirable, such as alcohol and tobacco, gambling, prostitution, and so-called recreational drugs. However, decisions regarding the composition of the CPI basket are not based on moral grounds, but rather on practical considerations. In the Australian CPI, gambling is excluded as it is difficult to establish the service or utility that households derive from gambling, and thus to determine an appropriate price measure. Recreational drugs and prostitution are both excluded as it is very difficult and indeed dangerous to obtain estimates of prices and expenditures, or to measure quality change.

5.11 All price indexes have a geographical dimension such as city, rural area, state, region or country. A further aspect to the geographical coverage that is important for CPI price collection is whether the objective is to measure price changes for:

• sales within a particular geographical area; or
• purchases by the residents of a geographical area.

5.12 If the aim of the index is to measure the prices of items sold in an area then the CPI basket will comprise all consumer goods and services sold in that region to households for final consumption. These sales can be made to households that are residents of that region, or to visitors to the region including overseas visitors.

5.13 On the other hand, if the index is to measure prices of items purchased by residents of a region, then it will comprise all consumer goods and services purchased by those households regardless of where they are purchased. So, in addition to purchases made in that region, it will include any purchases those households make whilst visiting other domestic regions and foreign countries, as well as items they order online or by post from suppliers outside the region.

5.14 The geographical dimension becomes more important the smaller the region to which the index relates.

5.15 The Australian CPI is compiled separately for each capital city. For general statistical purposes, the equivalent of a national index is the series published as the weighted average of eight capital cities. Each capital city index is compiled from data about acquisitions of goods and services by the resident population of that city, and includes their purchases from local outlets, purchases made in other capital cities and regions of Australia, and overseas purchases.

5.16 The CPI geographical classification is mainly based on the capital city average household expenditure data obtained from the latest available Household Expenditure Survey (HES). In line with the 2015-16 HES, the geographical coverage of the CPI is the Australian Statistical Geography Standard (ASGS) 2011, with the capital cities defined by Greater Capital City Statistical Areas (GCCSAs). For more information, refer to https://www.abs.gov.au/geography or Australian Statistical Geography Standard (ASGS): Volume 1 - Main Structure and Greater Capital City Statistical Areas, July 2011 (cat. no. 1270.0.55.001).

5.17 By and large, the ABS expects that most of the acquisitions made by capital city households will be from suppliers that are located in the same city. The most obvious exception is holiday accommodation services. Where online purchases are known to be significant (as with airfares and holiday accommodation), prices are collected from these sources.

5.18 The expenditures or quantity weights applied to the index basket reflect the expenditures of a reference population. Typically the basic unit of this reference population is the household¹.  The household is an appropriate unit because all members of the household jointly consume or use many items, such as food, motor vehicles, and housing, and it is not practicable to determine expenditure for each member of a household.

5.19 A CPI can be constructed for all households or for a subset of households (e.g. age pensioners, wage and salary earners, self-funded retirees). Even if the purpose of a CPI requires the broadest possible reference population, some types of households whose consumer expenditures are minimal or atypical may be excluded; for example, those living in institutions such as hospitals, barracks, prisons, and on board ships.

5.20 The reference population for the Australian CPI is private households in the eight capital cities. The eight capital cities are the six state capitals and the territory capitals of Canberra and Darwin. This is referred to as the CPI population group and it represents about two thirds of Australian private households. Ideally, the CPI population group should encompass all Australian households, but this is not feasible because of the substantial additional cost of collecting prices outside the capital cities.

5.21 In Australia, few people are not part of a private household; that is; do not reside in a private dwelling. These people live in public dwellings such as hotels, boarding houses, prisons, and university residences. Expenditure by people who reside in public dwellings is excluded from the Australian CPI.

5.22 A diagrammatic overview of the structure of the Australian CPI is provided in Figure 5.1. The structure can best be thought of from the top down. At the top is the total expenditure or pool of items purchased by the reference population. This is known as the All groups CPI index, and the CPI number produced at this level is commonly referred to as the headline rate of inflation. Below this, the index branches into finer and 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 expenditure class as this is the level at which the structure and weights are fixed for the life of a CPI series.

5.23 For the 28 ECs which are compiled using multilateral methods, the aggregation structure below the published (EC) level is modified to maximise the use of transactions data. Further details can be found in Use of transactions data in the Australian CPI of this manual.

5.24 This same structure is used for each of the eight capital cities. A full list of groups, sub-groups and expenditure classes is provided in Appendix 1.

5.25 The division of the groups and sub-groups into product classes is intended to reflect increasing levels of substitutability of the items consumed by households in response to changes in relative prices. For example, at the group level there are unlikely to be any substitution effects between Food and non-alcoholic beverages and Transport in response to changes in their relative prices. However, within the Oils and fats expenditure class it would be expected that households are more likely to substitute between margarine and butter in response to changes in their relative prices.

5.26 The commodity classification used in the Australian CPI is a demand-based classification that broadly aligns with the international standard Classification of Individual Consumption According to Purpose (COICOP). This classification is based on the concept of household utility. A significant advantage of using a COICOP-based classification is to allow greater international comparability of price inflation.

5.27 The 17th series CPI basket is divided into eleven major groups, each representing a broad set of commodities:

• Food and non-alcoholic beverages;
• Alcohol and tobacco;
• Clothing and footwear;
• Housing;
• Furnishings, household equipment and services;
• Health;
• Transport;
• Communication;
• Recreation and culture;
• Education; and
• Insurance and financial services.

5.28 These groups are divided in turn into 33 sub-groups, and the sub-groups into 87 expenditure classes. Presentation of the CPI in the form of groups and sub-groups provides the user with quite a degree of versatility in interpreting the results. Index numbers for individual groups and sub-groups can be analysed separately as can their individual effects on the whole index.

5.29 As described in Weights and their sources, it can be seen that the ABS HES is the most important source of CPI weights. The expenditures recorded in the HES were coded by the ABS according to the Household Expenditure Classification (HEC). To derive expenditures for the CPI expenditure classes, a correspondence was established with the HEC codes at their most detailed ten-digit level. Establishing the correspondence involved examining detailed listings of commodities coded to each HEC code. The correspondence is available as an Excel spreadsheet in Consumer Price Index; Correspondence with 2015-16 Household Expenditure Classification, Australia (cat. no. 6446.0.55.001) on the ABS website.

5.30 The majority of HEC codes could be exclusively allotted to a CPI expenditure class. For example, all of HEC code 0301030201 Biscuits expenditure is allotted to the CPI expenditure class Cakes and biscuits. However, there are some HEC codes where a one-to-one correspondence could not be established. There are just over 700 HEC codes at the ten-digit level, but only 87 CPI expenditure classes. The reasons why unique correspondences could not be established are as follows.

• The HEC code may not be sufficiently detailed. For example, HEC 1001090201 Supermarket and alcohol delivery charges could be for a variety of food and beverage items, so it was decided to spread household expenditure on supermarket and alcohol delivery charges across the various food, non-alcoholic beverage and alcoholic beverage expenditure classes in the CPI.
• Information provided by households does not allow a commodity to be clearly identified. These expenditures are reported in HEC codes such as 0302000000 Meat (excluding fish and seafood) nfd (where nfd is an abbreviation for not further defined). Again these expenditures were spread across appropriate CPI expenditure classes.
• Households cannot or do not separately identify some expenditures. For example, some state governments operate compulsory third-party vehicle insurance schemes, and the amount of insurance paid is included with the vehicle registration charge, often resulting in households reporting the combined amount only. In this case, a split was derived from average registration and insurance charges collected for the CPI.

5.31 Where HEC codes were split across CPI expenditure classes, the splits were determined using any industry or other data available or, as a last resort, subjectively. Mostly the expenditures concerned were small.

5.32 For inter-HES years the annually re-weighted CPI uses data from the National Accounts expenditure estimates, these are derived from Household Final Consumption Expenditure (HFCE) data. A classification concordance enables expenditure to be mapped between HFCE and CPI data. HFCE data, in it's most disaggregated form, is based on the Input-Output Product Classification (IOPC). IOPC to CPICC mapping is based on the classification concordance found in table 2 of Australian National Accounts: Input-Output Tables, Product Details (cat. no. 5215.0.55.001).