SECTION 4: STATISTICAL MODEL AND DATA SOURCES

CULTURAL AND CREATIVE INDUSTRY SUPPLY CHAINS

To maintain consistency with the national accounts, it is proposed that activity in the cultural and creative industry supply chains would be estimated for Australia using a top-down approach from the aggregates published annually in the ABS’ input output tables. The input-output tables provide information about the supply and use of products in the Australian economy, and the structure of, and inter relationships between, Australian industries. The most recent input-output tables currently available are for 2008-09 and were prepared in accordance with the international standards contained in the System of National Accounts 2008¹⁴.

The input-output tables present information for 111 Input-Output Industry Groups (IOIGs), which are aggregations of ANZSIC classes¹⁵. The IOIGs are not granular enough for the purpose of cultural and creative activity satellite accounts because many of the IOIGs contain ANZSIC classes that are wholly out-of-scope and/or ANZSIC classes that contain significant amounts of out-of-scope activity. The aggregate data for IOIGs would therefore need to be split.

The primary method of splitting the IOIG data to smaller industry levels would be to use the product details tables published as part of the input-output suite. The product details tables contain the value of Australian production for the 1,284 product categories in the ABS’ Input–Output Product Classification (IOPC). Each product category represents the goods and services characteristic of an industry and can be mapped to the ANZSIC classes in which they are produced.

The production value of products mapped to a cultural or creative ANZSIC class can be divided by the total Australian production of the broader IOIG to form a benchmark share. The benchmark share can then be applied to the aggregate data of the IOIG, such as Gross Value Added (GVA), to derive estimates for the cultural and creative ANZSIC class. ANZSIC classes in the wholesale and retail trade industries would be handled slightly differently since their output is represented in input-output tables by the trade margins on the goods they purchase for resale. The benchmark shares for wholesale and retail would therefore be the trade margins on products mapped to cultural and creative ANZSIC classes, divided by total retail and wholesale trade margins.

Using benchmarks from input-output tables in this way requires a default assumption that all the products produced by an IOIG have an identical production function. This assumption is used in the ABS’ tourism satellite accounts and is likely to be more valid for some IOIGs than others. Errors resulting from the use of this assumption at the individual ANZSIC class level are expected to tend to offset when estimates are aggregated to higher industry levels in a cultural and creative satellite account¹⁶.

The primary method of splitting IOIG data (explained above) would enable estimates to be constructed for the vast majority of ANZSIC classes in the cultural and creative industry supply chains. The remaining ANZSIC classes contain significant amounts of out-of-scope activity that would require further splitting using supplementary methods, or alternatively, activity in these ANZSIC classes could be estimated within the occupations component of the satellite account (component 2). The ANZSIC classes that contain significant amounts of out-of-scope activity are listed in the table below.

Code	ANZSIC class name	Comments
1320	Leather Tanning, Fur Dressing and Leather Product Manufacturing	Jewellery and Crafts activities in these classes are in-scope.
2029	Other Ceramic Product Manufacturing
2599	Other Manufacturing n.e.c.	Musical instrument manufacturing is in-scope.
3020	Non-Residential Building Construction	Cultural facilities construction activity is in-scope.
3109	Other Heavy and Civil Engineering Construction
4520	Pubs, Taverns and Bars	Patron spending during live entertainment events is in-scope. There are several existing studies that provide data on this topic but the data is not sufficient for a satellite account.
4530	Clubs (Hospitality)
6639	Other Goods and Equipment Rental and Hiring n.e.c.	Art work, video recorder and sound equipment hiring are in-scope.
6962	Management Advice and Related Consulting Services	Casting agent services and other specialised services for cultural and creative activity are in-scope.
7211	Employment Placement and Recruitment Services
7212	Labour Supply Services
7299	Other Administrative Services n.e.c.	Cultural event management, arts promotion and theatre and concert booking services are in-scope.
7510	Central Government Administration	Administration activity for cultural or creative policies and programs is in-scope.
7520	State Government Administration
7530	Local Government Administration
8101	Technical and Vocational Education and Training	Education and training in cultural and creative fields is in-scope.
8102	Higher Education
9499	Other Repair and Maintenance n.e.c.	Jewellery and musical instrument repair are in-scope.
9551	Business and Professional Association Services	Associations and groups which predominantly service the interests of cultural and creative industries and professions are in-scope.
9552	Labour Association Services
9559	Other Interest Group Services n.e.c.

The contribution of tourism to cultural and creative activity would be implicitly captured within the industry supply chain estimates. Within the ABS’ tourism satellite account this contribution is already drawn out in estimates of ‘cultural services’ output and GVA, which are based on visitor spending allocated to heritage and the creative and performing arts. These estimates are thought sufficient to meet stakeholder needs.

State and territory splits of the industry supply chains component could potentially be developed using a method similar to how the ABS’ state accounts are constructed¹⁷. Input-output tables would allow GVA for each Australian industry in the cultural and creative satellite accounts to be decomposed into their primary inputs:

• Compensation of Employees (COE), which is the total remuneration payable by enterprises to employees in cash or in kind;
• Gross Operating Surplus (GOS) and Gross Mixed Income (GMI), which are the incomes from production received by the owners of corporate enterprises and unincorporated enterprises; and
• Other taxes less subsidies on production, which includes employers’ payroll tax, motor vehicle taxes, land taxes, municipal and improvement rates, and a number of other items.

These primary inputs of each industry could potentially then be split by state and territory based on their shares of national indicators.

• The indicators used to split COE, GOS and GMI in the satellite accounts would be different from the ones used for the ABS’ state accounts, since these do not produce state data with the required level of industry granularity. Instead, the state and territory splits for a cultural and creative satellite account would be based on a combination of data directly collected by the ABS’ annual Economic Activity Survey (EAS) and the Business Activity Statements (BAS) collected by the Australian Taxation Office (ATO). These sources together provide estimates of employment by state, wages and sales income for each individual entity on the ABS’ Business Register. The combined data sets are used by the ABS to produce state and detailed industry estimates in Australian Industry (cat. no. 8155.0)¹⁸.
• Other taxes less subsidies on production would be split with an extension of the state matrices that are used for the ABS’ state accounts. The matrices contain a modelled allocation of taxes and subsidies from Government Finance Statistics (cat. no. 5519.0.55.001).

This possible method of producing state and territory splits would need to be assessed further once Australian estimates are developed since there may be unforseen issues with the quality of the underlying data. Also, while this method may be suitable for constructing Australian cultural and creative activity satellite accounts, it should be noted this is not necessarily suitable for broader application, particularly for industries that have many large multi state businesses.


CULTURAL AND CREATIVE OCCUPATIONS IN OTHER INDUSTRIES

The cultural and creative activity undertaken outside of the cultural and creative industry supply chains is proposed to be valued as a share of the COE aggregates in the ABS’ input-output tables. COE is the total remuneration payable by enterprises to employees, in cash or in kind, comprised of wages and salaries, and employers’ social contributions (the latter includes contributions towards retirement benefits such as superannuation).

An argument can be made that the value of this activity should also include shares of GOS and GMI, which represent returns to the capital used by workers. It is not proposed to do this in the Australian satellite accounts as there are considerable conceptual and practical difficulties with identifying how the use of capital is spread within an industry across different types of workers.

COE for each industry is proposed to be apportioned to cultural and creative activity by estimating the share of total employee remuneration payable to people employed in cultural and creative occupations. Multiple jobholding is prevalent among cultural and creative occupations and consequently this estimate would factor in main jobs and secondary jobs.

• For main jobs, total employment would be based on the employment aggregates contained in input output tables. Employment in cultural and creative occupations would be estimated by applying to the input-output employment aggregates the workforce occupation structure from Census data, adjusted for intercensal movements using Labour Force, Australia, Detailed, Quarterly (6291.0.55.003). The resulting employment estimates (for total and cultural and creative) would be applied to average earnings data from the ABS’ Survey of Employee Earnings and Hours (SEEH)¹⁹.
• For secondary jobs, employee earnings would be estimated using data from the ABS’ 2007 Survey of Employment Arrangements, Retirement and Superannuation (SEARS)²⁰. Until more recent data is collected, SEARS would provide multiple job holding rates for selected occupations and industries that can be applied to the workforce estimates described above. Average employee earnings in secondary jobs are also available in SEARS and this data could be adjusted over time for wage inflation using the ABS’ Wage Price Index, Australia (cat. no. 6345.0).

The Australian estimates for this component of the satellite accounts could potentially be split by state and territory using the Census employment profiles of each jurisdiction, coupled with adjustments for the differences in average wages by state using Average Weekly Earnings, Australia (cat. no. 6302.0). This method has not been previously trialled but is expected to be feasible.


VOLUNTEER SERVICES TO CULTURAL AND CREATIVE INSTITUTIONS

The value of volunteer services in Australian cultural and creative satellite accounts is proposed to be estimated using the ‘replacement cost’ method that is used by the ABS’ non-profit institutions Satellite Account, based on the United Nations’ Handbook. In this method, each hour of a volunteer’s time is valued at what it would cost to replace with paid labour²¹.

Data on volunteers are collected periodically in the ABS’ General Social Survey (GSS)²². In this survey people are asked about the type of organisations they volunteered for, the number of hours they worked unpaid and the types of work activities they performed. The number of hours volunteered with the ‘arts/heritage’ organisation type would be multiplied by replacement wage rate assumptions based on Employee Earnings and Hours Survey data. The replacement wage rate assumptions would vary with the work activities the volunteers performed and the occupation categories to which these activities match.

The volunteer hours data required for this estimate was last collected in 2006, but may be collected again in the survey scheduled for 2014. For the reference periods in between, volunteer hours could be estimated by applying the volunteering rates from the 2006 survey to the changed Australian population levels.

Australian estimates of volunteer hours in ‘arts/heritage’ could potentially be split by state and territory using the GSS’ state level estimates. Some of the relative standard errors on these estimates were high in the 2006 GSS and consequently it may be more appropriate to estimate volunteer hours by applying the Australian volunteering rates to each jurisdiction’s population. These methods have not been trialled previously but are expected to be feasible. The replacement wage rate assumptions applied to the state estimates would be adjusted for state differences in average wages based on Average Weekly Earnings, Australia (cat. no. 6302.0).


NON-MARKET OUTPUT OF MARKET PRODUCERS

As explained in section 3, the non-market output of market producers has previously been estimated in the ABS’ non profit institutions satellite account and the same method is proposed for use in Australian cultural and creative activity satellite accounts.

Non-market output of market producers would be valued as the difference between the output of market units when calculated by cost summation (the standard national accounts valuation method for non-market units), and output as calculated by sales value (the standard national accounts method for market units). Where output on a cost valuation basis exceeds output on a sales valuation basis, the difference is taken to be the non-market output of market producers. Where output on a sales basis exceeds output on a cost basis, non-market output of market producers is assumed to equal zero²³. This is represented in figure 7 below.

The estimate of this component would be compiled for cultural and creative industries using BAS data for individual entities, like in the Australian non-profit institutions satellite account. The estimate would be restricted to market producers that are non-profit institutions.

State and territory splits of the Australian estimate could potentially be constructed using the BAS data and EAS data. EAS directly collects state details for significant businesses and this could enable the non-market output of multi-state businesses to be assigned. However, the data sources have not previously been used for this purpose and there could be unforseen issues.


COMPOSITION DETAILS

Some of the more detailed data sought by stakeholders about the composition of Australian cultural and creative activity can be developed from existing statistical collections, while others would require new collections to be undertaken, as explained below.

Data on the types of incomes received and expenses incurred by entities are collected as part of the ABS’ rolling EAS program and in selected cases data about their assets and liabilities are also collected. The range of data items in the standard EAS questionnaire are not as detailed as desired by stakeholders, however with funding, additional data items could be added to the questionnaire. The EAS produces Australian estimates for ANZSIC subdivisions, which are two levels of aggregation higher than ANZSIC classes. This delivers detail for some of the ANZSIC class aggregations in the cultural and creative industry supply chains, but not a majority. Finer industry breakdowns can become available for years when the EAS rolling program specifically targets parts of the culture and creative industries supply chain (e.g. the survey of businesses in film, television and digital games run for 2011-12). State estimates can become available in those years at ANZSIC subdivisions or lower (they are available annually for divisions with the items limited to wages and salaries, employment and sales income). In the future it may be possible for the ABS to make available details of income, expenses, assets and liabilities at finer industry levels using ATO business tax data.

Details on Research and Development (R&D) expenditure in Australia are collected annually by the ABS from businesses, government, higher education and private non-profit organisations²⁴. These collections aim to be near-complete enumerations of R&D expenditure and provide a high level of industry detail, including estimates for ANZSIC classes where confidentiality standards allow. It could be possible to use this data to compile R&D expenditure for the aggregations of ANZSIC classes which make up the cultural and creative industry supply chains, both for Australia and a split by state and territory. However, the compilation process would be fairly resource intensive.

Details on the number and size of entities in each industry can be compiled from the ABS’ Business Register. Entity counts are possible for each state and territory according to each entity’s main location of operation. This would be compiled consistently with the data for businesses (a subset of all entities) which the ABS currently publishes annually in Counts of Australian Businesses, including Entries and Exits (cat. no. 8165.0).


INDUSTRY LINKAGES

Following the estimation of cultural and creative activity by ANZSIC class, it would be possible to extend these estimates further to construct an Australian supply-use table describing the linkages between cultural and creative industries and the rest of the economy. An example of the structure of a supply-use table is given by table 14 in Appendix 1.

In the supply-use framework, product groups and primary inputs are shown in the table rows, and industries and final use categories in the columns. Figures in rows show the total supply of products, whether locally produced or imported, and how these products are used by industries as intermediate inputs to production, or consumed as final demand. At the bottom of a supply-use table, the rows show the primary inputs purchased by industries, and by final demand. Reading down the columns shows the inputs (intermediate and primary) into each industry, and the composition of each final demand category. Therefore all flows of goods and services in the economy are covered.

Construction of this table for cultural and creative activity in Australia would begin with the ABS’ published supply-use table for Australia. GVA and output for IOIGs would be split and redefined according to the estimates of activity for cultural and creative industry supply chains. Intermediate uses would be reallocated across IOIGs in the same proportions as output.

Stakeholders are invited to comment on: the estimation methods proposed for the satellite accounts; and whether their concept of cultural/creative tourism is sufficiently captured by ‘cultural services’ in the ABS’ tourism satellite accounts.

---

¹⁴ European Commission, International Monetary Fund, Organisation for Economic Co-operation and Development, United Nations and World Bank (2009), System of National Accounts, 2008, <http://unstats.un.org/unsd/nationalaccount/docs/SNA2008.pdf>.
¹⁵ An IOIG to ANZSIC concordance is contained in Australian National Accounts: Input-Output Tables (Product Details) (cat. no. 5215.0.55.001), Table 2. IOPG to IOPC Concordance.
¹⁶ See Australian National Accounts: Tourism Satellite Account (cat. no. 5249.0), explanatory note 78.
¹⁷ See chapter 21 of Australian System of National Accounts: Concepts, Sources and Methods (cat. no. 5216.0).
¹⁸ See technical notes 1 and 3 in the Explanatory Notes of Australian Industry (cat. no. 8155.0).
¹⁹ Published in Employee Earnings and Hours, Australia (cat. no. 6306.0).
²⁰ Published in Employment Arrangements, Retirement and Superannuation, Australia (cat. no. 6361.0).
²¹ Australian National Accounts: Non-Profit Institutions Satellite Account (cat. no. 5256.0), p35-36.
²² Published in General Social Survey: Summary Results, Australia (cat. no. 4159.0) and Voluntary Work, Australia (cat. no. 4441.0)
²³ Australian National Accounts: Non-Profit Institutions Satellite Account (cat. no. 5256.0), p35.
²⁴ Published in Research and Experimental Development, All Sector Summary, Australia (cat. no. 8112.0); Research and Experimental Development, Businesses, Australia (cat. no. 8104.0); Research and Experimental Development, Higher Education Organisations, Australia (cat. no. 8111.0); and Research and Experimental Development, Government and Private Non-Profit Organisations, Australia (8109.0).