5216.0 - Australian National Accounts: Concepts, Sources and Methods, 2000  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 15/11/2000   
   Page tools: Print Print Page Print all pages in this productPrint All  
Contents >> Chapter 28: State accounts >> Chain volume estimates

Chain volume estimates

Government and household final consumption expenditure

28.62 Chain volume estimates of government and household final consumption expenditure for Australia are derived by aggregating the volume estimates for the States using a bottom-up approach. The State volume estimates are derived using State-specific price indexes. See Chapter 14 for details.

Private gross fixed capital formation

28.63 Chain volume estimates of private gross fixed capital formation for Australia are derived by aggregating the volume estimates for the States using a bottom-up approach. State-specific price indexes are used to derive the volume estimates for GFCF in dwelling and non-dwelling construction. National price indexes are used to derive the volume estimates for GFCF in machinery and equipment and intangibles, although account is taken of the different industry compositions within each State - industry patterns of GFCF by asset type vary. See Chapter 15 for details.

Public gross fixed capital formation

28.64 Quarterly chain volume estimates of public gross fixed capital formation for Australia are derived by aggregating the volume estimates for the States using a bottom-up approach. State-specific price indexes are used for GFCF in non-dwelling construction, but National price indexes are used to derive the volume estimates for the remainder of GFCF. Annual chain volume estimates of public GFCF by State are derived by aggregating the quarterly estimates and then constraining these estimates to be consistent with the national estimates. See Chapter 15 for details.

International trade in goods

28.65 Detailed estimates of current price exports of goods on a recorded trade basis at the two-digit level of the Standard International Trade Classification Revision 3 (SITC Rev 3) are revalued, generally using national rather than State-specific deflators, to produce volume estimates of these components. The exports data are on the basis of State of origin, i.e. the State in which the final stage of manufacture or production occurs. For most exports this should correspond to the basis required for State allocation, i.e. the State of final resident ownership. There is an assumption that the national deflators will usually provide a reasonable measure of change in price at the State level. Revaluation at the two-digit (division) level of SITC Rev 3 is a compromise giving a reasonable level of disaggregation by commodity type without increasing the magnitude of the revaluation exercise too significantly. In a very limited number of cases, more specific deflators have been used to replace the national deflator at the SITC Rev 3 division level. This has occurred in those cases where the composition of a particular division is known to be variable between States, and the division includes commodities with price movements which deviate significantly from the division average. Mineral and agricultural commodities are the most significant in this regard.

28.66 A similar approach has been adopted for imports of goods. The level of revaluation corresponds broadly to the commodity sub-group level of the balance of payments merchandise imports end-use classification of imports, i.e. two-digit Broad Economic Classification (BEC) by three-digit level of the SITC Rev 3. These data are on the basis of the State in which the imports were released from Australian Customs Service (ACS) control. In most cases this will correspond to the State of initial resident ownership, which is the required basis of State allocation. No information is available on which to base an adjustment for goods which are released from ACS control in a State other than that in which their owner resides.

28.67 The State estimates of exports and imports are benchmarked to the national estimates.

Compiling chain volume estimates of GSP

28.68 The methods used to obtain the chain volume estimates of GSP could be best described as 'indirect' because the only current price estimate of GSP available is obtained by aggregating the incomes accruing from production (i.e. the income approach). It is not possible to satisfactorily deflate such incomes to produce chain volume estimates because they do not comprise readily identifiable price and quantity elements. While it could be argued that the wages paid to an individual are equal to the product of the number of hours worked and the hourly wage rate, many employees are not paid according to the number of hours worked, and the supplements paid are generally independent of the number of hours worked. Moreover, from an employee's perspective, the value of wages and salaries has more to do with their purchasing power than with any hourly wage rate. Also, because gross operating surplus is the residual flow of income to the owners of capital, after the payment of labour costs and taxes less subsidies on production and imports, there is no way of defining it as the product of a price and a quantity. However, while it is not possible to construct chain volume estimates of GSP by deriving chain volume estimates of the various income components at the detailed level, it is possible to deflate the total current price estimate of GSP if a suitable aggregate deflator can be constructed.

28.69 Expenditure on GSP (i.e. the expenditure approach to deriving GSP) comprises the following:

State final demand
plus changes in inventories;
plus international exports of goods;
plus international exports of services;
less international imports of goods;
less international imports of services;
plus interstate exports of goods;
plus interstate exports of services;
less interstate imports of goods; and
less interstate imports of services.


28.70 Current price and chain volume estimates of SFD (the sum of private and public expenditures on consumption and gross fixed capital) and international trade in goods are compiled and published, but there are no estimates of GSP using the expenditure approach. The reason is the lack of changes in inventories data by State and, most importantly, the lack of data relating to inter-state trade in goods and services. Consequently, the approach taken has been to develop State-specific deflators to deflate the current price estimates of GSP derived using the income approach by calculating a deflator covering as much as possible of GSP using the expenditure approach.

28.71 The approach adopted essentially involves deriving the best possible current and volume estimates of significant subsets of the above data items. For each State, current price estimates of the identified components are added to the current price estimates of SFD and international trade in goods and, similarly, the volume estimates of those components are calculated. The quotient derived by dividing the aggregate chain volume estimates into the aggregate current price estimates is a chain Paasche price index. This price index is used to revalue the current price estimates of GSP derived using the income approach. It is crucial to identify separately those components for which the deflators deviate significantly from the average, since the method adopted effectively attributes the weighted average deflator to the unidentified components. Available information on the State distribution of these data items has been drawn on where possible and alternative indicators, as detailed below, have been used in their absence. Nevertheless, the aggregate current and chain volume estimates used in deriving this deflator are not considered to be complete measures. They merely serve to produce the best deflators for the income based measure of GSP that the available data and resources allow.

28.72 In some instances, both the current and volume estimates of a component, or some approximation to them, are split by State using available indicators. In other instances, a split is derived for either a current or a volume estimate, and the corresponding volume or current price estimates are derived by deflation if the initial split is in terms of current price estimates, or by inflation if the initial split is in terms of volume estimates. There is limited price information available at the State level, especially in the area of producers' prices and prices of internationally traded goods and services, which are the prices most relevant to the data items comprising the difference between GSP and SFD. There is an assumption underlying the approach adopted in deriving the aggregate State deflators, namely that price competition will ensure that the available price indicators are reasonably indicative of the changes at the State level. This is a less distorting assumption if the price indicators are weighted together at a reasonable level of commodity disaggregation, and implies deriving current and chain volume estimates at as fine a level of detail as possible. The indicators used in generating the State component splits are necessarily partial, which leads to the need for caution about the accuracy of the derived GSP deflators. A description follows of the available data sources in respect of each of the component categories and the way in which they are used.

Changes in inventories

28.73 There is no available State dimension to the existing quarterly survey estimates for inventories. Available information on inventory holding, production or sales has been used to give a State allocation of the national totals, in current price and volume terms, for a number of sectors for which price variation in the component commodities is considered likely to have a significant impact on the individual aggregate State deflators or for which indicator data are readily available.

28.74 In the case of private non-farm inventories:

      • Mining - annual estimates of inventory levels, by State, are available from mining census data for each of the two categories: non-metallic minerals and metallic minerals (see Australian Mining Industry (Cat. no. 8414.0)). These are used to split inventories estimates from the quarterly survey of inventories onto a State basis;
      • Manufacturing - annual factor cost estimates of manufacturing production, by State, based on manufacturing survey data (see Manufacturing Industry, Australia (Cat. no. 8221.0)) are used to produce State splits of the national estimates of the changes in inventories;
      • Retail trade - quarterly estimates of retail turnover, by State, based on estimates from the monthly retail turnover survey, are used to allocate, by State, estimates of the changes in inventories; and
      • Wholesale trade - a fixed relationship between States has been derived from wholesale turnover estimates from the 1991-92 survey of wholesalers (see Wholesale Industry: Details of Operations (Cat. no. 8638.0)). This fixed relationship is used to allocate, by State, changes in wholesalers' inventories. For those marketing authorities privatised in July 1999, changes in inventories are allocated to the States using indicators such as production or export data.

28.75 In the case of farm and public authorities inventories, changes in inventories are allocated to the States using annual indicators such as production or exports data.

International trade in services

28.76 Activity indicators are used to derive State contributions to the national current price and chain volume estimates of exports and imports of services. Since there are no available indicators of variations in the component deflators between States, the same splits have been applied in current and volume terms:

  • Transport -
      Exports: allocated using merchandise exports, recorded trade State data;
      Imports: allocated using merchandise imports, recorded trade State data.
  • Travel and Government services -
      Exports: allocated using State estimates of expenditure on travel services by non-residents in Australia (these data are used in the derivation of household final consumption expenditure);
      Imports: allocated using State estimates of expenditure on travel services by Australians overseas (these data are used in the derivation of household final consumption expenditure).
  • Insurance services, Freight -
      Exports: allocated using merchandise exports, recorded trade State data;
      Imports: allocated using merchandise imports, recorded trade State data.
  • Other, balance of above -
      Exports: allocated using estimates of State final demand;
      Imports: allocated using estimates of State final demand.

Interstate trade in goods

28.77 Interstate exports and imports of goods pose a particular problem. These data items are different from those above in that the flows must net to zero at the national level, and hence there are no equivalent current price estimates at the national level which can be split and no relevant price indicators derived in relation to these items at the national level. The strategy adopted is to generate volume estimates, derive price indexes relevant to these flows, and then inflate the volume estimates to yield current price estimates.

28.78 For those States for which survey data by commodity on the value of interstate imports and exports of goods are available (Western Australia (annual up to 1991-92) and Queensland (quarterly)), the data are deflated to obtain volume estimates. For the remaining States, and for Western Australia from 1992-93, indicators of interstate exports and imports are derived using broad (chain volume) activity indicators to extrapolate interstate trade flow estimates which have been derived using estimates from a model produced by the Monash University Centre of Policy Studies, as described in Notes on the Construction of the 1986-87 Input-Output Database for the MONASH-MR Model by G.A. Meagher. The model yields estimates of interstate trade in goods by industry and State of supply and by State of destination. The relationships between the estimates by State and by industry are assumed to be fixed over time. The Monash model data used at present are in respect of 1986-87. The model identifies interstate trade flows only with respect to three industries: agriculture, mining and manufacturing. An internationally sourced component of interstate exports and imports is also incorporated. This component is not derived directly from the Monash data, but by comparison of the Monash net interstate trade estimates and the net interstate trade balances after identifying all other items, by State, in 1986-87.

28.79 The export and import flow data from the model are used to weight together price indexes for the three industries to form an interstate import price index and an interstate export price index for each State. The industry price indexes are derived separately for each State. They are fixed weighted indexes for agriculture and chain Paasche indexes for mining and manufacturing, on the basis of production or turnover estimates for the three industries. Thus the price indexes are fixed weighted between industries and between supplying States on the basis of the composition of interstate trade in 1986-87 as identified in the Monash model, but variably weighted within the mining and manufacturing industries.

28.80 State by industry deflators have been derived by the following means:

      • Agriculture, etc. - Australian Bureau of Agricultural and Resource Economics (ABARE) indexes of prices received by farmers, by State, have been used for this industry (see the ABARE quarterly publication Indexes of Prices Received and Paid by Farmers). These price indexes are fixed-weighted. It would be possible to produce chain Paasche indexes using component commodity prices and production details by commodity, given the relative importance of the flows in this sector and the possibly low correlation between the composition of production and the composition of the interstate trade component. However, this additional degree of complexity has not been pursued.
      • Mining - State chain Paasche price indexes are compiled by weighting together price indicators for a variety of mineral commodities using monthly mineral production data. Both the price and quantity estimates are consistent with those published in the quarterly ABARE publication Quarterly Mineral Statistics.
      • Manufacturing - State chain Paasche price indexes are derived using turnover data from the annual manufacturing census as weights, and producer price indexes. These annual price indexes are extrapolated, when necessary, using Paasche price indexes derived using production indicators from the monthly ABS surveys of manufacturing production and ABARE production data to weight relevant component producer price indexes.

28.81 The deflators for the internationally sourced component of interstate trade reflect the fact that only estimates of the net flows are made for this component. A chain Paasche price index of international imports into New South Wales and Victoria, the net exporter States, is used for revaluation of flows into the net importer States, and the respective aggregate State international import deflators are used for flows out of New South Wales and Victoria. Current price estimates of interstate exports and imports flows are derived by inflating the extrapolated volume estimates using the deflators derived as described above.

Interstate trade in services

28.82 Interstate trade in services, which is not accounted for in the model used for goods, is likely to be significant in the case of at least some of the States, with the Australian Capital Territory likely to be particularly affected. State estimates of household final consumption expenditure include an adjustment to account for net expenditures interstate on services by households. This item is added to household final consumption expenditure to adjust it from a territorial to a resident basis. In order to ensure consistency of treatment, the current and volume estimates of household expenditures interstate are included here, but with the opposite sign. It has not been possible to include estimates for the net expenditures interstate on services by the business and government sectors. The net effect of making only partial allowance for the interstate flow of services is that the weighted average deflator derived for the identified components of GSP, including SFD, will be attributed to the balance of this item.

Qualifications attaching to chain volume GSP estimates

28.83 The incomplete nature of the price and expenditure data available at the State level for the derivation of the GSP deflators, and the generally lower level of accuracy of State data, referred to earlier, mean that there is a greater scope for inaccuracy than in comparable estimates at the Australian level. The chain volume measures of GSP are derived measures, calculated by adjusting the current price estimates using specially constructed State deflators. This means that the chain volume measures incorporate the combined effects of the inaccuracies in the current price estimates and those in the deflators. These inaccuracies reflect a number of factors, including: higher relative sampling errors; the 'fuzziness' of some State data arising from difficulties that businesses and government agencies operating across the country have in allocating their activities by State; and the limited availability of source data on a State basis in a number of areas. The inaccuracies in the component value and price estimates may be compounded or offset in the process of deriving the chain volume measures.

28.84 In analysing the chain volume measures it is important to recognise the data limitations at the State level and to be aware that the accuracy of the estimates will not be as high as that of the corresponding national estimates.



Previous PageNext Page