NAME OF ORGANISATION
Australian Bureau of Statistics (ABS)
The quarterly Building Activity Survey collects data relating to all stages (commenced, under construction, completed) of residential and non-residential building activity undertaken in the reference quarter. Other key data items are derived, and include the value of work done, work yet to be done and work not yet commenced.
The quarterly Building Activity Survey has been conducted since the September quarter 1980. Prior to this, building activity details were obtained from the Building Operations Census.
The statistics are compiled from returns collected from builders and other individuals and organisations engaged in building activity. The ABS Building Approvals collection provides the framework for the survey, from which a sample of newly approved building jobs is selected each quarter for inclusion in Building Activity Survey. As well as those jobs newly selected in the current quarter, other building jobs included in a quarterly Building Activity Survey comprise those jobs selected in previous quarters that have not been completed (or commenced) by the end of the previous quarter.
The Building Activity Survey currently consists of:
Since the estimates for private sector building activity (including alterations and additions) are based on a sample of approved building jobs, they are subject to sampling error, ie, 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. Furthermore, the estimation of smaller jobs from their approval value introduces further error. Relative standard errors (RSEs) of the estimates are published with the quarterly data to which they refer. These RSEs incorporate estimates of both components of estimation error.
- a sample of:
- all building jobs involving residential building jobs valued at $50,000 or more and non-residential building jobs valued at $250,000 or more,
- a complete enumeration of:
- all building jobs involving residential building jobs valued greater then $10,000 and less than $50,000 and non-residential building jobs valued greater than $50,000 and less then $250,000, with values for their data items modelled,
- new houses over $1million,
- A&A to residential over $500,000,
- all jobs over $5 million.
The use of sample survey techniques means that reliable estimates of building activity, including alterations and additions, are generally available only at the state/territory and Australian levels. For example, details of both the number and value of new houses and of new other dwellings and the value of alterations and additions to residential buildings can be made available for Capital City Statistical Divisions (Capital City SDs) and Rest of State. Similarly, details of the value of non-residential building activity are available at the Capital City SD and Rest of State level and perhaps other broad areas (eg, other SDs), depending on the RSEs involved. Estimates are produced to the Statistical Division level.
The predominant requirement of Building Activity Survey is to provide quarterly estimates of Value of Work Done and Value of Commencements data for national accounts purposes and economic policy departments. Building Activity Survey is the main source data for private sector Gross Fixed Capital Formation on new dwellings and alterations and additions to residential buildings as well as (private sector) new other buildings. Building Activity Survey also provides official national and state estimates of the number of dwellings commenced.
The statistics are compiled on the basis of returns collected from builders and other individuals and organisations engaged in building activity.
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 (eg, construction of roads, bridges, railways, earthworks, etc) is excluded.
The population list from which new jobs are selected comprises all approved building jobs which were notified to the ABS up to but not including the last month of the reference quarter (ie, up to the end of November with respect to the December quarter survey). Thus, given the coverage of the Building Approvals Collection, residential building jobs with a Value of Approval (VOA) of less than $10,000 and non-residential building jobs with a VOA of less than $50,000 are excluded from the Building Activity Survey.
The Building Activity Survey is a quarterly sample survey of approximately 34,000 - 43,000 building jobs in various stages of construction, identified largely through the building approval process. The statistics are compiled from returns collected from builders and other individuals and organisations engaged in building activity.
The population of approved jobs from which selections are made for inclusion in Building Activity Survey comprises all approved building jobs which were notified to the ABS up to but not including the last month of the reference quarter. This introduces a lag to the statistics in respect of those jobs notified and commenced in the last month of the reference quarter. For example, building jobs which were notified as approved in the month of December and which actually commenced in that month are shown as commencements in the March quarter.
Building jobs included in each quarterly 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. Once selected, a job is included in Building Activity Survey until it is completed or abandoned.
The data collected are Value of work done, Estimated completion value, Month of commencement, Month of completion and Floor area created. This data is linked to unit record data collected as part of the Building Approvals Collection.
Construction Work Done, Australia, Preliminary (cat. no. 8755.0) - issued from March quarter 2001 onwards and generally released in the last week of the second month after the reference period. (Previously called Building Activity, Australia: Building Work Done, Preliminary (cat. no. 8755.0)).
Dwelling Unit Commencements, Australia, Preliminary (cat. no. 8750.0) - released 54 working days (WDs) after the end of the reference period. (Previously called Building Activity, Australia: Dwelling Unit Commencements, Preliminary (cat. no. 8750.0)).
Building Activity, Australia (cat. no. 8752.0) - released 74 WDs after the reference period.
Contents of Publications
8755.0 - Up to the December quarter 2000 issue, contains preliminary estimates of the value of work done in Australia for both residential and non-residential building and for the private sector and total. From the March quarter 2001 issue onwards, includes estimates for work done on alterations and additions to residential building and for work done on engineering construction by sector. From the September quarter 2002 issue onwards, includes some state and territory data. From the June quarter 2004 issue onwards, includes estimates for work in the pipeline, ie number of dwelling units approved but not yet commenced at end of quarter, value of work approved but not yet commenced at end of quarter, and work yet to be done at end of quarter. Some of the data is available in seasonally adjusted and trend terms, as well as original, and some is available as chain volume measures.
8750.0 - contains preliminary estimates of the number of dwelling units (new houses, new other residential and conversions, etc) commenced for private sector and public sector, by state and territory. Total number of dwelling units commenced by state and territory in seasonally adjusted and trend terms.
8752.0 - contains information for each state and territory, and for private and public sectors for Australia. Includes number of dwelling units commenced and completed and value of building work commenced, completed, done and yet to be done, all by state and territory. Information is provided for new residential buildings (houses and other residential), alterations and additions to residential buildings and non-residential buildings by function of building (eg, offices, warehouses, accommodation, etc.). In addition, many series are shown in seasonally adjusted and trend terms, while the value of building work commenced and done is also shown in chain volume measures terms, with the reference year being the year prior to the latest complete financial year. It includes more comprehensive and updated information to that contained in cat. nos. 8755.0 and 8750.0. It also includes tables of Relative Standard Errors.
Data from the above national publications are available on AusStats within minutes of release of the publications, as well as most state and territory series previously published in 8752.1-8. Some series no longer published in 8752.0 (from and including the March quarter 2003 issue), together with data previously included in Construction Activity at Constant Prices, Australia (cat. no. 8782.0), which was withdrawn after the June quarter 1995 issue, are also available on AusStats.
A data cube containing state and territory data from September quarter 2001 to the current quarter by type of work, type of building, sector, number of dwelling units commenced and completed and value of work done, commenced and completed, is available from the ABS website.
Special Data Service
We can prepare tables from the Information Warehouse to suit the needs of clients.
A Job Specific
These items are generally the same as collected in the Building Approvals collection. Their values are carried through to Building Activity Survey. However, the values for Building Activity Survey can be amended during processing. Normal survey outputs are not necessarily available across all of the classifications presented here.
The following items are recorded using the classifications listed:
Type of Work (WRK)
From and including the December quarter 2000 cycle, a WRK code of "7" was introduced to be consistent with the Building Approvals Collection and to uniquely represent the conversion of a non-residential building to a residential building. The Type of Work classification thus became:
1 New building
2 Alterations and/or additions to an existing building
3 Refurbishment (a special category of alterations and additions primarily for non-residential building)
7 Conversions (of non-residential buildings to residential buildings).
Data published for the number of dwelling units under the heading of "conversions, etc", is obtained by a combination of the above codes 1, 2 or 7 with relevant Type of Building (BLD) codes below. Thus, if jobs with a WRK code of 2 or 7 involve 1 or more dwellings, the dwellings are counted in "Conversions, etc" no matter what the BLD code, but if a job with a WRK code of 1 involves 1 or more dwellings, it's only counted as a "Conversion, etc" if it has a BLD code of non-residential.
Note also that the number of dwelling units published under the heading of "conversions, etc", is the equivalent of the total number of dwelling units approved as published in Building Approvals, Australia (cat. no. 8731.0) under the 3 headings "Alterations and additions to residential buildings", "Conversions" and "Non-residential building".
Type of Building (BLD) -
Prior to September quarter 2001 type of building was coded to the 1986 Functional Classification of Buildings. From September quarter 2001 the type of building has been coded to the 1999 Functional Classification of Buildings.
The 1999 Functional Classification of Buildings (Cat no. 1268.0.55.001) is available on the ABS website.
Sector of ownership (OWN)
1 Private sector
5 Public sector (Federal, State and Local)
The ownership of a building is classified as either public sector or private sector according to the sector of the intended owner of the completed building as evident at the time of approval. Data is published for private/public/total sectors.
State and Statistical Local Area (SLA) are recorded using the version of Australian Standard Geographical Classification (ASGC) that is relevant to the time period.
Details of new houses and total new residential building (number of dwellings and value) and of alterations and additions to residential buildings (value only) and total residential building (value only) are available for Capital City SDs and Rest of State, subject to a warning regarding the level of accuracy. Data for Capital City SDs and Rest of State continues to also be available for non-residential and other residential building, despite the introduction of sampling for these types of building from the December quarter 2001 and March quarter 2002 respectively.
Other concepts (summary)
From the September quarter 1995 (for cat. no. 8750.0) and the June quarter 1996 (for cat. no. 8752.0) publications, dwelling units created as part of alterations and additions to existing buildings (including conversions to dwelling units) and as part of the construction of non-residential buildings are shown separately under the heading of "Conversions, etc" and included in the total number of dwelling units. Previously, such dwellings were included only as a footnote.
A building is defined as commenced when the first physical building activity has been performed on site in the form of materials fixed in place and/or labour expended (this includes site preparation but excludes delivery of materials, the drawing of plans and specifications and the construction of non-building infrastructures such as roads).
Value of building commenced or under construction
This represents the anticipated completion value or Builders Anticipated Value based, where practicable, on the estimated market or contract price of building jobs excluding the value of land and landscaping. Site preparation costs are included. Where jobs proceed over several quarters the anticipated completion value reported on the return for the first (commencement) quarter may be amended on returns for subsequent (under construction) quarters as the job nears completion.
Value of job
Initially this is the Value of Approval, but it can be amended any quarter by the respondent.
Value of work done at the end of the reference quarter
This is the total value of work done on the building job to date as reported by the respondent.
Value of work done in the reference quarter
This is calculated by subtracting the total work done to the end of the previous quarter from the total work done to the end of the current quarter.
New South Wales
Part of State Metropolitan
Part of State Extra-Metropolitan
Capital City Statistical Division
Comments and/or Other Regions
BACS has been a quarterly survey since its commencement in the September quarter 1980.
The Building Activity Survey produces estimates at various stages of the quarterly collection cycle, each set of estimates being based on successively higher response rates. These stages (or "cut-offs") are:
|Referred to as:||and used to produce:|
|National Accounts (NAs) "stage" or "estimates"||Construction Work Done, Australia, Preliminary (ABS cat. no. 8755.0) and estimates for the ABSs quarterly National Accounts |
|Version 1 (V1) "stage" or "estimates"||Dwelling Unit Commencements, Australia, Preliminary (ABS cat. no. 8750.0)|
|Version 2 (V2) "stage" or "estimates"||Building Activity, Australia (ABS cat. no. 8752.0) and revised data for 8755.0 (up to September quarter 2003 issue)|
|Version 3 (V3) "stage" or "estimates"||revised data for 8750.0|
|Version 4 (V4) "stage" or "estimates"||revised (and final) data for all 3 publications|
The Building Activity Survey has been conducted since the September quarter 1980. Prior to this, building activity details were obtained from the Building Operations Census.
From the September quarter 1990, non-residential building jobs (both new and alterations and additions) with an approval value of $50,000 or more were included in the survey. Prior to this quarter, this "cut-off" was $30,000.
From and including the December quarter 1996:
- a new sampling methodology was used with the aim of keeping the number of newly selected sample sector jobs constant, and
- estimates were made at the V1 stage to impute commencements for non-responding uncommenced residential jobs.
From and including the September quarter 1998:
A new publication Building Activity, Australia: Building Work Done, Preliminary (ABS cat. no. 8755.0), contained preliminary estimates of the value of building work done, to provide an early indication of trends in building activity. The estimates are based on a response rate of approximately 80% of the value of building work done during the quarter.
For the June quarter 2000:
Three variations of the collection forms (BACS1 to BACS3) had the words "Note: Please include the value of the Goods and Services Tax (GST)" added after Question 1 "Estimated value of job when completed". The BACS4 form simply had the words "(Including GST)" added after the corresponding Question 6, due to space constraints.
For the September quarter 2000:
All respondents received collection forms for which an instruction was added to the question on "work done" to include the GST component, similar to that added to the June quarter 2000 forms for the question on the "completion value" of the job.
From and including the September quarter 2000:
Not all building jobs are reported to ABS on the preferred (value data to be inclusive of GST) basis. Where this is indicated to be the case, appropriate adjustments are made to the data for those jobs. An adjustment is also made to non-residential data such that published estimates of non-residential building are on a GST exclusive basis.
From and including the December quarter 2000:
A new sampling methodology and imputation and estimation system was introduced.
From and including the March quarter 2001
A revised publication Construction Work Done, Australia, Preliminary (ABS cat. no. 8755.0) was issued. This continued data previously included in Building Activity, Australia: Building Work Done, Preliminary (ABS cat. no. 8755.0) and presented some additional series and also included some preliminary figures on engineering construction work done. The latter are based on a response rate of approximately 80% of the value of engineering work done during the quarter.
From and including the June quarter 2001
The stratum boundaries for sampling new private sector houses and alterations and additions to private sector house jobs were changed. They were generally increased in value, resulting in fewer jobs being in the completely enumerated strata and more being selected from the sampled strata.
From and including the December quarter 2001
Sampling was introduced for private sector non-residential building, for both new and alteration and addition jobs. Jobs above a cut-off level of Value of Approval (VOA) - which varies for each state - continued to be completely enumerated.
Thus, although complete data by SLA was no longer available for non-residential building, aggregate details of large (VOA more than $1million) private sector non-residential jobs were still available by SLA, as was complete public sector non-residential building.
From and including the March quarter 2002
Sampling was introduced for private sector other residential building, for both new and alteration and addition jobs. In the five mainland states, jobs were either sampled or completely enumerated, depending on a combination of the VOA and number of dwelling units involved. In Tasmania and the two territories, all jobs continued to be completely enumerated.
Thus, although complete data by SLA was no longer available for other residential building for the five mainland states, aggregate details of large (VOA more than $0.5 million for new buildings and more than $0.3 million for alterations and additions) private sector other residential jobs were still available by SLA. Complete private sector other residential buildings data for Tasmania and the two territories continued to be available by SLA, as did complete public sector other residential buildings data for all states and territories.
From and including the March quarter 2003
The March quarter 2003 issue of the publication Building Activity, Australia (ABS cat. no. 8752.0) was revised and included extensive changes to the format and data content of the tables. Changes included more state and territory data and fewer private/public sector dissections. Generally however, data no longer included in the publication continued to be available on AusStats.
The March quarter 2003 issue of 8752.0 also included, for the first time, data based on the new 2000 Functional Classification of Buildings (FCB). The new FCB affected only the non-residential sector and data on the new basis was released commencing with the September quarter 2001.
From and including the June quarter 2003
New 'concurrent' seasonal adjustment methodology, replacing forward factor methodology, was adopted, commencing with the June quarter 2003 issue of cat. no. 8755.0
The final issues of the state and territory publications were issued. Many state/territory series were included in 8752.0 and all series previously in the state/territory publications were included in AusStats. A new product, referred to as a 'data cube', containing a time series of most of the variables in the discontinued publications, was introduced.
From and including the December quarter 2003
A review of the sample design resulted in changes to the target sample numbers for private sector new selections for 'new houses' and for 'alterations and additions to houses' mainly, but also some changes for 'new other residential'. The sample re-allocation resulted in an increase in the target number of new selections for sampled strata from about 4,000 to about 4,400 jobs or about 10%. It ensured the continued accuracy of data, particularly for the number of new houses commenced and the value of alterations and additions to houses, by state and territory.
From and including the June quarter 2005
The sample design this quarter was refined to reduce the sample size. This was done by a) using the same value boundaries across different strata; b) raising the value of modelled A&A jobs from $40k to $50k; and c) collapsing non-residential and non-house residential strata into a single quarterly strata as opposed to having three separate strata representing the three months in the quarter.
Complete enumeration of the Northern Territory was discontinued and sampling introduced.
Modelling of small (<$50,000) residential jobs was introduced. Such jobs were no longer selected in sample. This change was not expected to have a noticeable impact on estimates (negligible estimated bias and small impact on variances).
From and including the September quarter 2005
Modelling of small (<$250,000) non-residential jobs was introduced. Such jobs were no longer selected in sample. This change is not expected to have a noticeable impact on estimates (negligible estimated bias and small impact on variances).
From and including the June quarter 2006
Complete enumeration of the public sector jobs was discontinued and sampling introduced.
Data availability comments
Data is available in the form of publications, time series on the ABS Website and in a data cubes. Information is also released in Australian Economic Indicators (and corresponding State versions)
DATE OF LAST UPDATE FOR THIS DOCUMENT
17/04/2012 10:08 AM