Sources and methods - Quarterly

The Quarterly Business Indicators Survey provides the data for estimating changes in private non-farm inventories. This survey collects estimates of the closing book value level of inventories from which changes are derived. The levels are also used in deriving the estimates of the inventories to sales ratio that is published in the quarterly national accounts.

The Quarterly Business Indicators Survey provides the quarterly movements in inventories for mining; manufacturing; wholesale trade; retail trade; electricity, gas; water and waste services; accommodation and food services; and telecommunication services.

The survey does not include some non-farm industries with only minor inventory holdings. For these industries, data are modelled from historical data, which was compiled from the periodic economic censuses and Taxation Statistics using the estimates for in-scope industries.

The most successful means of deriving chain volume changes in inventories has been found to be differencing chained estimates of the levels. The steps involved are as follows:

1. Re-value quarterly book value levels to levels valued in the prices of the previous year;
2. Sum to the required level of aggregation;
3. Calculate quarter to quarter indexes which show the volume growth in levels between the present and previous quarter;
4. Compound these indexes to form a chained index;
5. Reference the chained index to the June quarter book valuelevel of the reference year to give a chain volume series of
6. Difference the resultant values to derive the chain volume estimates of changes in inventories.

The price indexes that are used to re-value book value levels of inventories are formed by weighting together component price indexes from ABS publications: Consumer Price Index, Australia; Producer Price Indexes, Australia; and International Trade Price Indexes, Australia; and wage rate indexes from the publication, Wage Price Index, Australia. The regimen and weights for these price and wage rate indexes are derived using data from the various censuses and surveys conducted by the ABS.

Chain volume estimates of changes in private non-farm inventories are published in the following detail in the national accounts:

• mining;
• manufacturing;
• wholesale trade;
• retail trade; and
• other non-farm industries.

It is noteworthy that, unlike other national accounts aggregates, quarterly chaining and annual chaining of volumes of changes in inventories produce identical annual chain volume estimates of changes in inventories. This is because chain volume estimates of changes in inventories are derived by differencing the chain volume estimates of the levels of inventories which relate to the end of quarterly and annual periods and coincide for the June quarter.

Changes in farm inventories include changes in:

• inventories held on farms (including wool, wheat, barley, oats, maize, sorghum, hay, fertiliser, apples and pears, and livestock);
• produce (e.g vegetables) held in cold store where ownership remains with the primary producer.

Quarterly changes in the book value of inventories of wool are estimated as the difference between inventory levels derived from available information obtained from the ABARES publication, Agricultural Commodities; the Australian Wool Exchange; and from ABS publications, Value of Agricultural Commodities Produced, Australia and Livestock Products, Australia. Quarterly changes in the book value of inventories of apples and pears are estimated as the difference between inventory levels, which was modelled from historical data, provided monthly by the Tasmanian Department of Primary Industries, Parks, Water and Environment.

Animals reared for slaughter are regarded as work-in-progress. The estimates are derived from numbers of animals collected annually by the ABS, and prices from the Australian Bureau of Agricultural and Resource Economics and Sciences. Animals reared for breeding purposes or recurrent production (such as dairy cattle and sheep reared for wool production) are treated as gross fixed capital formation.

The techniques used to calculate chain volume estimates of changes in farm inventories are only slightly different to those shown above for private non-farm inventories. The difference is that for many of the detailed components of the former it is difficult to obtain true book value levels. Therefore, constant price estimates of changes in inventories that preceded the introduction of chain volume estimates are used in the calculations.

The steps followed are:

1. Derive constant price levels of inventories for each component by accumulating the constant price changes over time and add these to a base level (i.e. the level at a particular time for which there is an estimate). The base level is often only an approximation of the true level and is sometimes only derived as a figure which will ensure that subsequent levels remain positive. These constant price levels are then converted to levels valued in the prices of the previous year;
2. Sum to the required level of aggregation;
3. Calculate quarter to quarter indexes which show the volume growth in levels between the present and previous quarter;
4. Compound these indexes to form a chained index;
5. Reference the chained index to the June quarter book value level of the reference year to give a chain volume series of levels; and
6. Difference the resultant values to derive the chain volume estimates of changes in inventories.

For farm commodities, the price indexes used to convert constant price levels into levels valued in the prices of the previous year are calculated using production unit values.

Changes in public authorities’ inventories include estimates for general government, public non-financial corporations and public financial corporations.

Quarterly estimates of changes in the book value of marketing authorities’ inventories are derived from information supplied by the authorities concerned.

Recorded inventories include demonetised gold transactions (gold sales and gold loans) by the Reserve Bank of Australia and the construction of military equipment for export.

Quarterly data are obtained from ABS Government Finance Statistics and Balance of Payments quarterly collections covering all significant public corporations/organisations and from the Department of Finance’s Quarterly Ledger.

The techniques used to calculate chain volume estimates of changes in public authorities’ inventories are only slightly different to those shown above for private non-farm inventories. The difference is that for many of the detailed components of the former it is difficult to obtain true book value levels. Use is therefore made of the constant price estimates of changes in inventories that preceded the introduction of chain volume estimates and which are still calculated.

The steps followed are:

1. Derive constant price levels of inventories for each component by accumulating the constant price changes over time and add these to a base level (i.e. the level at a particular time for which there is an estimate). The base level is often only an approximation of the true level and is sometimes only derived as a figure which will ensure that subsequent levels remain positive. These constant price levels are then converted to levels valued in the prices of the previous year;
2. Sum to the required level of aggregation;
3. Calculate quarter to quarter indexes which show the volume growth in levels between the present and previous quarter;
4. Compound these indexes to form a chained index;
5. Reference the chained index to the June quarter book value level of the reference year to give a chain volume series of levels; and
6. Difference the resultant values to derive the chain volume estimates of changes in inventories.

For the other public authorities’ inventories component, a price index is constructed in a similar way to that described above for private non-farm inventories.

The book values of private non-farm inventories are disaggregated into 30 industry groups: mining; 15 groups within manufacturing; eight within wholesale trade; retail trade; accommodation and food services; electricity; gas; water and waste services; construction; transport and storage; and telecommunication services. The inventories held by manufacturing industries are further split into Materials, and work-in-progress or finished goods. An IVA is derived using the assumptions described in above paragraphs.

An IVA is generally not necessary for farm inventories because the values of changes in inventories at average current quarter prices can be estimated directly from detailed quantity and price data. However, the value of changes in inventories for wheat and wool is calculated by subtracting from their respective sales the value of receivals. A production valuation adjustment (PVA) was previously required for the estimates of changes in inventories of wheat (ceased in June quarter 2010) and wool (ceased in March quarter 2011). This was due to the differences in current quarter sale price and the price at which receivals were valued. The PVA adjustment was deducted from the value of the receivals when calculating gross value of farm production (estimated in deriving farm income) on a national account’s basis.