EXPLANATORY NOTES


INTRODUCTION

1 This publication contains detailed estimates from the quarterly Building Activity Survey. Each issue includes revisions to the previous quarter. Therefore data for the latest quarter should be considered to be preliminary only.


SCOPE AND COVERAGE

2 The statistics were compiled using building approval details and returns collected from builders and other individuals and organisations engaged in building activity. Since the September quarter of 1990, the quarterly estimates have represented all approved public and private sector owned:

• residential building jobs valued at $10,000 or more.
• non-residential building jobs valued at $50,000 or more.

3 As of the June quarter 2006, the survey has consisted of:

• an indirect, modelled component comprising residential building work with approval values from $10,000 to less than $50,000 and non-residential building work with approval values from $50,000 to less than $250,000. The contributions from these building jobs are modelled based on their building approval details.
• a direct collection of all identified building work having approval values of $2,000,000 or more.
• a sample survey, selected from other identified building work.

4 For historical changes to the collection design see the Directory of Statistical Sources on the ABS website.

5 The use of sample survey techniques in the Building Activity Survey means that reliable estimates of private sector building activity are generally available only at state, territory and Australia levels. Although subject to higher relative standard errors (refer to paragraphs 18-21), a range of sub-state estimates of building activity may be available. For further information on the availability of Building Activity estimates, contact the inquiries contact officer on the front of this publication. Detailed data on Building Approvals, based on information reported by local government and other reporting authorities, are available for regions below state and territory level from the Building Approval series compiled by the ABS.

6 The statistics relate to building activity which includes construction of new buildings and alterations and additions to existing buildings. Construction activity not defined as building (e.g. construction of roads, bridges, railways, earthworks, etc.) is compiled from the ABS Engineering Construction Survey. Results from the Building Activity Survey, together with estimates from the Engineering Construction Survey, provide a complete quarterly picture of building and construction.

7 Building jobs included in each quarter in the Building Activity Survey comprise those jobs selected in previous quarters which have not been completed (or commenced) by the end of the previous quarter and those jobs newly selected in the current quarter. The population list from which jobs are selected for inclusion comprises all approved building jobs which were notified to the ABS (refer paragraph 3) up to but not including the last month of the reference quarter (i.e. up to the end of August in respect of the September quarter survey). This introduces a lag to the statistics in respect of those jobs notified and commenced in the last month of the reference quarter (i.e. for the month of September in respect of the September quarter survey). For example, jobs which were notified as approved in the month of June and which actually commenced in that month are shown as commencements in the September quarter. Similarly, building jobs which were notified in the month of September and which actually commenced in that month are shown as commencements in the December quarter.

8 From the September quarter 2002, building activity in the External Territories of Australia is included in these statistics. Jervis Bay is included in New South Wales, while Christmas Island and Cocos (Keeling) Islands are included in Western Australia.


TREATMENT OF GST

9 Statistics on the value of building work (current prices) show residential building on a GST inclusive basis and non-residential building on a GST exclusive basis. This approach is consistent with that adopted in the Australian National Accounts which is based on the conceptual framework described in the 1993 edition of the international statistical standard System of National Accounts (SNA93).

10 SNA93 requires value added taxes (VAT), such as the GST, to be recorded on a net basis where:

(a) both outputs of goods and services and imports are valued excluding invoiced VAT
(b) purchases of goods and services are recorded including non-deductible VAT.

11 Under the net system, VAT is recorded as being payable by purchasers, not sellers, and then only by those purchasers who are not able to deduct it. Almost all VAT is therefore recorded in the SNA93 as being paid on final uses - mainly on household consumption. Small amounts of VAT, may however, be paid by businesses in respect of certain kinds of purchases on which VAT may not be deductible.

12 Within building activity statistics, purchasers of residential structures are unable to deduct GST from the purchase price. For non-residential structures, the reverse is true. While the ABS collects all building activity data on a GST inclusive basis, it publishes value data inclusive of GST in respect of residential construction and exclusive of GST in respect of non-residential construction.

13 It is appropriate to add the residential and non-residential components to derive total building activity. Valuation of the components of the total is consistent, since, for both components, the value data is recorded inclusive of non-deductible GST paid by the purchaser. As such, total building activity includes the non-deductible GST payable on residential building.


CLASSIFICATION

14 Ownership. The ownership of a building is classified as either private sector or public sector, according to the sector of the intended owner of the completed building as evident at the time of approval. Residential buildings being constructed by private sector builders under government housing authority schemes whereby the authority has contracted, or intends to contract, to purchase the buildings on or before completion, are classified as public sector.

15 Functional classification of buildings. A building is classified according to its intended major function. Hence a building which is ancillary to other buildings, or forms a part of a group of related buildings, is classified to the function of the building and not to the function of the group as a whole. An example of this can be seen in the treatment of building work approved for a factory complex. In this case, a detached administration building would be classified to Offices, a detached cafeteria building to Retail/wholesale trade, while factory buildings would be classified to Factories. An exception to this rule is the treatment of group accommodation buildings where, for example, a student accommodation building on a university campus would be classified to Educational. The categories included under type of building classifications are defined in the Glossary.

16 In the case of a large multi-function building which, at the time of approval, is intended to have more than one purpose (e.g. a hotel/shops/residential apartments project), the ABS endeavours to split the details according to each main function. Where this is not possible because separate details cannot be obtained, the building is classified to the predominant function of the building on the basis of the function which represents the highest proportion of the total value of the project.

17 Building jobs are classified both by the Type of Building (e.g. 'house', 'factory') and by the Type of Work involved (e.g. 'new', 'alterations and additions' and 'conversions, etc.'). These classifications are used in conjunction with each other and are defined in the Glossary.


RELIABILITY OF THE ESTIMATES

18 Since the estimates for building activity (including alterations and additions) are based on a sample of approved building jobs, they are subject to sampling error; that is, they may differ from the figures that would have been obtained if information for all approved jobs for the relevant period had been included in the survey. One measure of the likely difference is given by the standard error (SE), which indicates the extent to which an estimate might have varied by chance because only a sample of approved jobs was included. There are about two chances in three that a sample estimate will differ by less than one SE from the figure that would have been obtained if all approved jobs had been included, and about nineteen chances in twenty that the difference will be less than two SEs. Another measure of sampling variability is the relative standard error (RSE), which is obtained by expressing the SE as a percentage of the estimate to which it refers. The RSEs of estimates provide an indication of the percentage errors likely to have occurred due to sampling, and are shown in tables 38 and 39.

19 An example of the use of RSEs is as follows. Assume that the estimate of the number of new private sector houses commenced during the latest quarter is 30,000 (for actual estimate see table 18) and that the associated RSE is 1.5% (for actual percentage see table 38). There would then be about two chances in three that the number which would have been obtained if information had been collected about all approved private sector house jobs would have been within the range 29,550 to 30,450 (1.5% of 30,000 is 450) and about nineteen chances in twenty that the number would have been within the range 29,100 to 30,900.

20 Estimates that have an estimated relative standard error between 10% and 25% are annotated with the symbol ‘^’. These estimates should be used with caution as they are subject to sampling variability too high for some purposes. Estimates with an RSE between 25% and 50% are annotated with the symbol ‘*’ indicating that the estimate should be used with caution as it is subject to sampling variability too high for most practical purposes. Estimates with an RSE greater than 50% are annotated with the symbol ‘**’ indicating that the sampling variability causes the estimates to be considered too unreliable for general use.

21 The imprecision due to sampling variability, which is measured by the RSE, should not be confused with inaccuracies that may occur because of inadequacies in the source of building approval information, imperfections in reporting by respondents, and errors made in the coding and processing of data. Inaccuracies of this kind are referred to as non-sampling error, and may occur in any enumeration whether it be a full count or only a sample. Every effort is made to reduce the non-sampling error to a minimum by the careful design of questionnaires, efforts to obtain responses for all selected jobs, and efficient operating procedures. Some non-sampling error is introduced by the estimation process for smaller jobs (see paragraph 3). The impact of this component of error has been estimated and included in the RSE measures presented in this publication.


SEASONAL ADJUSTMENT

22 Seasonally adjusted building statistics are shown in tables 1-10, 13-21, 23 and 24. In the seasonally adjusted series, account has been taken of normal seasonal factors, ‘trading day’ effects arising from the varying numbers of working days in a quarter and the effect of movement in the date of Easter which may, in successive years, affect figures for different quarters.

23 Since seasonally adjusted statistics reflect both irregular and trend movements, an upward or downward movement in a seasonally adjusted series does not necessarily indicate a change of trend. Particular care should therefore be taken in interpreting individual quarter-to-quarter movements. Some of the component series shown have been seasonally adjusted independently. As a consequence, while the unadjusted components in the original series shown add to the totals, the adjusted components may not add to the adjusted totals. (For example, the sum of the adjusted state series - for both work done and number of dwelling unit commencements - may not add to the adjusted Australian total). Therefore, figures should not be derived using the adjusted totals.

24 From the June quarter 2003, the seasonally adjusted estimates are produced by the concurrent seasonal adjustment method which takes account of the latest available original estimates. The concurrent method improves the estimation of seasonal factors and, therefore, the seasonally adjusted and trend estimates for the current and previous quarters.

25 A more detailed review of concurrent seasonal factors will be conducted annually, generally prior to the release of data for the December quarter.

26 The revision properties of the seasonally adjusted and trend estimates have been improved by the use of autoregressive integrated moving average (ARIMA) modelling. ARIMA modelling relies on the characteristics of the series being analysed to project future period data. The ARIMA model is assessed as part of the annual reanalysis. For more information on the details of ARIMA modelling see feature article: Use of ARIMA modelling to reduce revisions in the October 2004 issue of Australian Economic Indicators (cat. no. 1350.0).

27 As a general rule, caution should be exercised in using the seasonally adjusted series for dwelling unit commencements in Northern Territory and Australian Capital Territory. The small numbers and volatile nature of these data makes reliable estimation of the seasonal pattern very difficult.


TREND ESTIMATES

28 Seasonally adjusted series can be smoothed to reduce the impact of the irregular component in the adjusted series. This smoothed seasonally adjusted series is called a trend estimate.

29 The trend estimates are derived by applying a 7-term Henderson moving average to the seasonally adjusted series. The 7-term Henderson average (like all Henderson averages) is symmetric but, as the end of a time series is approached, asymmetric forms of the average are applied. Unlike weights of the standard 7-term Henderson moving average, the weights employed here have been tailored to suit the particular characteristics of individual series.

30 While the smoothing technique described in paragraphs 28 and 29 enables trend estimates to be produced for recent quarters, it does result in revisions to the estimates for the most recent three quarters as additional observations become available. There may also be revisions because of changes in the original data. For further information, see Information Paper: A Guide to Interpreting Time Series - Monitoring Trends, 2003 (cat. no. 1349.0) or contact the Assistant Director, Time Series Analysis on Canberra (02) 6252 6540 or email <time.series.analysis@abs.gov.au>.


CHAIN VOLUME MEASURES

31 Chain volume estimates of the value of commencements and work done are presented in original, seasonally adjusted and trend terms for Australia and for each state and territory.

32 While current price estimates of the value of commencements and work done reflect both price and volume changes, chain volume estimates measure changes in value after the direct effects of price changes have been eliminated and therefore only reflect volume changes. The direct impact of the GST is a price change, and hence is removed from chain volume estimates. The deflators used to revalue the current price estimates in this publication are derived from the same price data underlying the deflators compiled for the dwellings and new other building components of the national accounts aggregate ‘Gross fixed capital formation’.

33 The chain volume measures of commencements and work done appearing in this publication are annually reweighted chain Laspeyres indexes referenced to current price values in a chosen reference year. The reference year is updated annually in the September quarter publication. Each year’s data in the value of commencements and work done series are based on the prices of the previous year, except for the quarters of the latest incomplete year which are based upon the current reference year. Comparability with previous years is achieved by linking (or chaining) the series together to form a continuous time series.

34 Chain volume measures do not, in general, sum exactly to the total value of the components. Further information on the nature and concepts of chain volume measures is contained in the ABS Information Paper: Australian National Accounts, Introduction of Chain Volume and Price Indexes (cat. no. 5248.0).

35 The factors used to seasonally adjust the chain volume series are identical to those used to adjust the corresponding current price series.


ACKNOWLEDGMENT

36 ABS publications draw extensively on information provided freely by individuals, businesses, governments and other organisations. Their continued cooperation is very much appreciated: without it, the wide range of statistics published by the ABS would not be available. Information received by the ABS is treated in strict confidence as required by the Census and Statistics Act 1905.


RELATED PRODUCTS

37 Users may also wish to refer to the following publications:

Building Approvals, Australia, cat. no. 8731.0
Construction Work Done, Australia, Preliminary, cat. no. 8755.0
Dwelling Unit Commencements, Australia, Preliminary, cat. no. 8750.0
Engineering Construction Activity, Australia, cat. no. 8762.0
House Price Indexes: Eight Capital Cities, cat. no. 6416.0
Housing Finance, Australia, cat. no. 5609.0
Private Sector Construction Industry, Australia, cat. no. 8772.0
Producer Price Indexes, Australia, cat. no. 6427.0.

ABS DATA AVAILABLE ON REQUEST

38 As well as the statistics included in this and related publications, the ABS may have other relevant data available on request. Inquiries should be made to the National Information and Referral Service on 1300 135 070.