Focus on PDV
About the data
The primary data source for this section on post, digital and visual effects (PDV) businesses is the Service Industry Surveys (SIS) undertaken by the Australian Bureau of Statistics (ABS) in 1996/97, 1999/00, 2002/03 and 2006/07. These surveys were published as Film and Video Production and Distribution in 1996/97 and 1999/00, Television, Film and Video Production in 2002/03 and Television, Film and Video Production and Post-production Services in 2006/07 (cat. no. 8679.0). It is not known if or when another survey will be conducted.
Initially, the PDV sector was surveyed as part of the broader production and post-production industry, and data prior to 2006/07 was derived by Screen Australia (then the Australian Film Commission) isolating and analysing those businesses that earned the majority of income from the provision of PDV services. In 2006/07, the ABS used similar methodology to categorise businesses either to a new Australian and New Zealand Standard Industry Classification (ANZSIC) category specifically for post-production services and other motion picture and video activity (code 5514) or to one for motion picture and video production services (code 5511), enabling it to present the two sets of data separately for the first time: see Focus on PDV: Overview and definition.
The use of this data has some limitations.
Before 2006/07, the data generated from the survey was taken from all employing units on the ABS Business Register that had been classified to film and video production (1993 ANZSIC code 9111) and was based on a sample of all film and television businesses, including PDV businesses. In 2006/07, the survey data was also based on a sample but this was taken from all employing and significant non-employing units classified to 2006 ANZSIC code 5514. Significant non-employing units were defined as those with an annual turnover of $70,000 or more. They accounted for 29.6 per cent of the total number of post-production businesses in 2006/07, contributing 4.8 per cent of estimated total income, 3.6 per cent of estimated total expenditure and 5.6 per cent to estimated employment (the latter being working proprietors and partners as, by definition, non-employing units do not have employees).
All figures reported are estimates only and are subject to both sampling and non-sampling errors. The size of these errors, or the degree of confidence that can be placed in the figures, is stated where applicable.
Due to several factors – slightly different methodologies, sample stratification, the low sample size of contributing units, the snapshot objective of each survey, and the gaps between surveys – historical change is not an element inherent in the survey’s design, and the ABS urges caution when considering historical comparisons. However, there has been a high degree of consistency in methodology and questions asked, particularly for the 1999/00 and 2002/03 surveys, so the data can be used to indicate some direction of change and, to a lesser degree, the extent of change rather than to provide statistically valid rates of change. To help overcome this limitation, attempts have been made to qualify conclusions drawn from such comparisons via other sources where possible. Commenting on change in industry with the expanded scope found in the 2006/07 survey is more problematic because it is difficult to apportion how much of the change was significant and not attributable to non-employing units and statistical error.
The ABS also urges caution in comparing historical state estimates, as data for multi-state businesses was compiled differently. In 2006/07 and 2002/03, it was assigned according to the state of actual operations, while in 1999/00 it was assigned to the state of the main business address of the multi-state business, known as the state of the head office, rather than the state of actual operations. Some evidence of double counting appears as a result, and the counts of businesses do not always add up to the total for Australia, but these discrepancies are minor.
Additionally, it should be stressed that before 2006/07 the PDV data was drawn from a survey in which the primary goal was not to measure the PDV sector separately, but rather include it in a wider industry of film and video production.
The data is also based on what could be termed the ‘independent’ PDV sector, as it does not include PDV services supplied by broadcasters. Commercial broadcasters are surveyed separately using a different survey instrument, and public broadcasters are not surveyed at all due to the difficulty of isolating radio activities from those of television. However, PDV work outsourced by broadcasters would be included as income in this analysis (i.e. broadcasters as clients rather than providers of PDV services).