Warning - this online document is now out of date. For the updated online version of this document see: Australian Consumer Price Index: Concepts, Sources and Methods - 2005 . For an acrobat version see the 'Details' tab for Australian Consumer Price Index: Concepts, Sources and Methods.
INTRODUCTION
2.1 Prior to the introduction of the CPI in 1960, there had been five series of retail price indexes compiled by the (then) Commonwealth Bureau of Census and Statistics. Those indexes were:
(i) The A Series Index, covering only food, groceries and house rents (all houses), which was first compiled in 1912, with index numbers going back to 1901, and was discontinued in 1938. Its main use was for wage adjustment purposes between 1913 and 1933.
(ii) The B Series Index, covering only food, groceries and house rents (4 and 5 roomed houses), which was first compiled in 1925 and was discontinued in 1953. It was introduced to replace the A Series Index for general statistical purposes but was never used for purposes of wage adjustment.
(iii) The C Series Index, covering food and groceries, house rents (4 and 5 roomed houses), clothing, household drapery, household utensils, fuel, lighting, urban transport fares, smoking and some miscellaneous items, which was introduced in 1921 and was discontinued in 1961. The food and rent component of the C Series Index was the same as that for the B Series Index. The C Series Index was used for the purposes of wage adjustment from 1934 to 1953.
(iv) The D Series Index, which was derived by combining the A and C Series Indexes and was compiled especially for wage adjustment purposes for a short period in 1933-34.
(v) The Interim Retail Price Index, covering food and groceries, house rent, clothing, household drapery, household utensils, fuel, lighting, urban transport fares, smoking and some services and miscellaneous items, was first compiled in 1954 and discontinued in 1960. The Interim Index was intended to serve as a transitional index, but to some extent it replaced the C Series Index for general statistical purposes for a number of years prior to 1960; it was never used for wage adjustment purposes.
C SERIES INDEX
2.2 By far the most important of these former retail price indexes was the C Series Index, which was the principal retail price index in Australia for close to forty years. It was first compiled in 1921, with index numbers being compiled retrospectively to 1914. C Series Index numbers were compiled for:
(i) the capital city in each of the six states
(ii) four of the larger towns in each of the six states
(iii) weighted average of five towns (including capital city) in each of the six states
(iv) weighted average of six state capital cities
(v) weighted average of thirty towns (including capital cities)
(vi) three additional towns - Whyalla, Port Augusta and Canberra.
2.3 The C Series Index was subject to a general review in 1936 and a slightly revised regimen was introduced following that review. The regimen then remained unchanged until the C Series Index was discontinued.
2.4 The main reason for the long interval without any review or change in composition of the C Series Index after 1936 was the recurrent changes in consumption patterns which occurred during and after World War II. It was considered impossible at the time to devise a revised weighting pattern which would be any more representative, on a continuous basis, of post-war consumption than was the existing weighting pattern of the C Series Index. The Commonwealth Statistician of the time, in successive editions of the Labour Report during the 1950s and 1960s, explained the absence of any reweighting of the C Series Index in the following words: “From the outbreak of war in 1939 to late in 1948, periodic policy changes in various wartime controls (including rationing) caused recurrent changes in consumption and in the pattern of expenditure. This rendered changes desirable but made it impracticable either to produce a new index, or to revise the old one, on any basis that would render the index more representative than it already was of the changing pattern of household expenditure in those years. When commodity rationing had virtually ceased in the latter part of 1948 action was taken by the Statistician to collect price data of about 100 additional items and to gather information as to current consumption and expenditure patterns. This was done to facilitate review of the component items and weighting system of the C Series Retail Price Index in the light of the new pattern of wage earner expenditure and consumption that appeared to be then emerging. But there supervened, in the next few years, conditions which caused wide price dispersion, coupled with a very rapid rise in prices and a new sequence of changes in consumption and in the pattern of wage earner expenditure. Under these conditions it was not possible to devise any new weighting pattern likely to be more continuously representative of conditions then current, than was the existing C Series Index on the 1936 revision.”
2.5 In 1953 the decision was made to continue to compile the C Series Index on its pre-war basis but to also compile an interim retail price index based as nearly as possible on the post-war pattern of consumer usage and expenditure. Nevertheless, the C Series Index continued to be regarded by the majority of users as the principal official index and was the one used in most indexation and escalation arrangements throughout the 1950s.
INTERIM RETAIL PRICE INDEX
2.6 The Interim Retail Price Index was based on post-war consumption weights. Compared with the C Series Index, the Interim Index covered an expanded range of items, including additional foods (such as packaged breakfast foods, soft drinks, ice cream and confectionery) and services (such as dry-cleaning and shoe repairs). It retained the same weighting pattern throughout the period of its compilation and no attempt was made to revise its weights to take account of major changes in expenditure patterns and lifestyles that were occurring during the 1950s. During that decade, house renting was substantially replaced by home ownership, the use of motor cars partially replaced the use of public transport, and a variety of electrical appliances, and subsequently television, became widely used in households. During the same period, widely disparate movements occurred in the prices of different items of household consumer expenditure. It was considered that the combined impact of these factors made it impracticable to successfully introduce a comprehensive new retail price index during the period to 1960.
CONSUMER PRICE INDEX
2.7 In 1960, a new approach was implemented. Instead of the former emphasis on long-term fixed-weight indexes, the aim was to compile series of shorter-term indexes that would be chain linked together to form long-term series. The Consumer Price Index, commonly referred to as ‘the CPI’, was the first price index of this kind constructed in Australia.
2.8 The Consumer Price Index was first compiled in 1960, with index numbers being compiled retrospectively to mid 1948. Like its predecessor indexes, the CPI was designed to measure quarterly changes in retail prices of goods and services purchased by metropolitan wage earner households.
2.9 The CPI has been reviewed and reweighted thirteen times. At its inception in 1960, the CPI consisted of three original linked series, with changes in weights in 1952 and 1956. Weights were changed in 1960 and subsequently in 1963, 1968, 1973, 1974, 1976, 1982, 1987, 1992, 1998 and 2000. The method of linking the sequence of short-term price indexes to form one continuous series is described in Chapter 9.
LONG-TERM LINKED SERIES
2.10 To provide an approximate long-term measure of retail/consumer price change for the period since the first Australian retail price index was compiled, the ABS has constructed a single series of index numbers by linking together selected retail/consumer price index series from amongst those described above (see table 2.1). The index numbers are expressed on a reference base 1945 = 100.0, which was the end of a period of relative price stability during World War II. The successive series linked together to produce this long-term series of index numbers are:
- from 1901 to 1914, the A Series Retail Price Index
- from 1914 to 1946-47, the C Series Retail Price Index
- from 1946-47 to 1948-49, a combination of the C Series Index, excluding rent, and the housing group of the CPI
- from 1948-49 onwards, the CPI.
2.11 This long-term series of index numbers is updated each year and published in Year Book Australia (cat. no. 1301.0). A graph of the series (from table 2.1) is presented in figure 2.1.
FIGURE 2.1: GRAPH OF LONG-TERM RETAIL PRICE INDEX/CPI
TABLE 2.1 RETAIL/CONSUMER PRICE INDEX NUMBERS (a)
Year | Index | Year | Index | Year | Index | Year | Index |
![](/icons/ecblank.gif) | ![](/icons/ecblank.gif) | ![](/icons/ecblank.gif) | ![](/icons/ecblank.gif) | ![](/icons/ecblank.gif) | ![](/icons/ecblank.gif) | ![](/icons/ecblank.gif) | ![](/icons/ecblank.gif) |
1901 | 47 | 1931 | 78 | 1961 | 252 | 1991 | 1898 |
1902 | 50 | 1932 | 74 | 1962 | 251 | 1992 | 1917 |
1903 | 49 | 1933 | 71 | 1963 | 252 | 1993 | 1952 |
1904 | 46 | 1934 | 73 | 1964 | 258 | 1994 | 1989 |
1905 | 48 | 1935 | 74 | 1965 | 268 | 1995 | 2082 |
1906 | 48 | 1936 | 75 | 1966 | 276 | 1996 | 2136 |
1907 | 48 | 1937 | 78 | 1967 | 286 | 1997 | 2141 |
1908 | 51 | 1938 | 80 | 1968 | 293 | 1998 | 2159 |
1909 | 51 | 1939 | 82 | 1969 | 302 | 1999 | 2191 |
1910 | 52 | 1940 | 85 | 1970 | 313 | 2000 | 2289 |
1911 | 53 | 1941 | 89 | 1971 | 332 | 2001 | 2389 |
1912 | 59 | 1942 | 97 | 1972 | 352 | 2002 | 2462 |
1913 | 59 | 1943 | 101 | 1973 | 385 | 2003 | 2530 |
1914 | 61 | 1944 | 100 | 1974 | 443 | | |
1915 | 70 | 1945 | 100 | 1975 | 510 | | |
1916 | 71 | 1946 | 102 | 1976 | 579 | | |
1917 | 75 | 1947 | 106 | 1977 | 650 | | |
1918 | 80 | 1948 | 117 | 1978 | 702 | | |
1919 | 91 | 1949 | 128 | 1979 | 766 | | |
1920 | 103 | 1950 | 140 | 1980 | 844 | | |
1921 | 90 | 1951 | 167 | 1981 | 926 | | |
1922 | 87 | 1952 | 196 | 1982 | 1028 | | |
1923 | 89 | 1953 | 205 | 1983 | 1132 | | |
1924 | 88 | 1954 | 206 | 1984 | 1177 | | |
1925 | 88 | 1955 | 211 | 1985 | 1257 | | |
1926 | 90 | 1956 | 224 | 1986 | 1370 | | |
1927 | 89 | 1957 | 229 | 1987 | 1487 | | |
1928 | 89 | 1958 | 233 | 1988 | 1594 | | |
1929 | 91 | 1959 | 237 | 1989 | 1714 | | |
1930 | 87 | 1960 | 245 | 1990 | 1839 | | |
a. The index numbers relate to the weighted average of six state capital cities from 1901 to 1980 and to the weighted average of eight capital cities from 1981. Index numbers are for calendar years. |
|