Contents >>
Chapter 2 Nature and objectives of the classification
NATURE AND OBJECTIVES OF THE CLASSIFICATION
INTRODUCTION
2.1 To use statistical information about business units effectively, it is first necessary to organise that information into categories suitable for economic analysis. This can be done by using different classifications depending on the particular interests of users.
2.2 An industry classification is one way of organising data from a business unit perspective. It provides a standard framework under which units carrying out similar productive activities can be grouped together, with each resultant group being referred to as an industry. The term industry is used in its widest context, covering the full range of economic activities undertaken to produce both goods and services.
2.3 Each individual class is defined in terms of a specified range of activities. It is common for a business unit to engage in a range of activities wider than those designated as belonging to a particular class, and when this occurs we classify the unit based on its predominant activity.
2.4 Activities undertaken which belong to classes other than that to which the unit is classified, are described as its 'secondary activities'. The secondary activities of a unit play no part in assigning the class to which the unit is classified, but are useful for coverage and specialisation ratio analysis. Refer to paragraph 2.20 for the definition of coverage and specialisation ratios.
2.5 In an industrial classification, each unit has to be classified uniquely to one class so that only those units with the same predominant activities are brought together to form a class.
2.6 The ANZSIC has been designed primarily as a classification of Type of Activity Units (TAUs) in the Australian statistical system, and Kind of Activity Units (KAUs) in the New Zealand statistical system, although it can also be used for classifying some other unit types. The TAU and the KAU are defined in Appendices 1 and 2 respectively.
2.7 The ANZSIC is a hierarchical classification comprising four levels, namely Divisions (the highest level of the classification), Subdivisions, Groups and Classes (the lowest level of the classification).
PRINCIPLES FOR RECOGNISING ANZSIC 2006 CLASSES
Standard classification principles
2.8 All classifications, particularly those that are international or national statistical standards, should satisfy a number of fundamental principles. These include that the classification:
- is comprehensive in its coverage;
- has categories which are mutually exclusive;
- has categories which can be readily understood by users and data providers;
- is hierarchical to support its use for different statistical purposes;
- should remain stable over a period of time, or be designed so that it can easily be updated; and
- be based on a strong and consistently applied conceptual framework.
Conceptual framework
2.9 The conceptual framework adopted for the development of ANZSIC 2006 uses supply-side based industry definitions and groupings. Using this approach, units engaged in similar productive activities are grouped together. Units in an industry will therefore exhibit similar production functions (a term used to describe the transformation of intermediate inputs, through the application of labour and capital, to produce outputs).
2.10 In ANZSIC 2006, this framework has been applied more rigorously and consistently than before. Industry statistics will be more consistently compiled and analytically useful as a result. Refer to paragraphs 2.16, 2.17, 2.18 for more detail on the supply-side concept.
Relevance
2.11 Periodic reviews of major classifications such as the ANZSIC are required to ensure they remain current and relevant. They need to reflect new and emerging industries, and the new business methods and technologies applied in the production of goods and services. They also need to reflect new and emerging user requirements for industry classified statistics.
International comparability
2.12 In the interests of international statistical comparability, ANZSIC 2006, as far as practicable, aligns at the two digit subdivision level with the proposed ISIC. Rev. 4.
Real world organisation of units
2.13 A basic principle in the ANZSIC is that industry classes should reflect the way activities are actually organised in the real world. This ensures that industry classes reflect realistic and recognisable segments of Australian and New Zealand industry and that the businesses classified to them have the necessary data for statistical collections.
Economic significance
2.14 Generally, industry classes were formed only if they were economically significant in either Australia or New Zealand. In Australia, employment had to be at least 3,500, or annual turnover at least $AUS 250 million. In New Zealand, employment had to be at least 700, or annual turnover at least $NZ 50 million. Applying this test helps ensure that the number of classes formed is justified on cost/benefit grounds and that a reasonable balance is achieved across the whole of the classification.
2.15 The criteria were modified for those industries where the standard measures of output (e.g. value added, turnover) gave a misleading picture of economic significance e.g. wholesale trade, retail trade, financial and insurance services, public administration and safety. In these cases, additional classes were not formed simply because the economic significance criterion was exceeded. Other, more relevant criteria needed to be satisfied.
SUPPLY-SIDE CONCEPT
2.16 As noted above, the conceptual framework adopted for the development of ANZSIC 2006 uses supply-side based industry definitions and groupings. Using this approach, business units engaged in similar productive activities are grouped together.
2.17 Following this approach, business units in a particular class will use similar inputs and apply similar transformation processes to produce similar outputs. The study of production (e.g. industry performance, productivity and structural analysis) requires an industrial classification that follows this type of framework.
2.18 The supply-side concept can be strictly applied in developing only the lowest level of the classification (the class level). This is the level at which homogeneity of economic activity is greatest. For example, the application of this concept has led to the creation of a class for fossil fuel electricity generation and a separate class for hydro-electricity generation even though the final product, i.e. electricity, is the same.
INDUSTRY HOMOGENEITY
2.19 As indicated above, industry classes should be comprised of businesses that undertake similar economic activities i.e. they should be as homogeneous as possible. In the development of ANZSIC 2006, the ABS and Statistics NZ have used the information that is available to assess and increase homogeneity at the class level. This has helped ensure that the classification reflects the structure and organisation of Australian and New Zealand industry.
2.20 Homogeneity can be measured by the calculation of specialisation and coverage ratios. The specialisation ratio measures the extent to which units belonging to a particular class engage in the activities designated as primary to that class. The coverage ratio measures the extent to which the activities designated to a particular class are undertaken by units belonging to that class.
2.21 It is highly desirable that the specialisation and coverage ratios exceed 70 per cent for the formation of individual classes. This minimises the extent to which the output of each class includes output of activities which belong to other classes. As a consequence, users of industry statistics should note that classes do not contain all of the units which undertake the activities belonging to that class. Units engaging in these activities on a secondary basis will be classified to a different class according to their predominant activity.
APPLYING AND BALANCING CLASSIFICATION PRINCIPLES
2.22 The classification principles set out above for the creation of industry classes in ANZSIC 2006 often reinforce each other. For example, good industry homogeneity helps ensure the classification is on a strong conceptual basis and reflects the real world organisation of units. Economic significance helps ensure the classification reflects the real world organisation of units and that it reflects a contemporary view of Australian and New Zealand industry.
2.23 On the other hand, the review and development of the classification has to balance the often competing demands of relevance, continuity and comparability. The principles applied for the definition of classes are sometimes in conflict. For example, international comparability cannot be fully achieved if there is a different structure in Australian and New Zealand industry, or if some internationally recognised activities are not significant in the two countries. In these circumstances, judgement needs to be exercised to determine the most appropriate outcome.
FORMATION OF BROADER LEVEL CATEGORIES
2.24 As classes are grouped into higher levels of aggregation, the degree to which units exhibit similar production functions generally falls. At the higher levels of the classification, emphasis moves increasingly to the output side of the production function, so that at the top of the hierarchy, divisions are created and defined by looking more at what is produced, and less at the activities undertaken to produce that output.
2.25 This has the important consequence of causing units with very different inputs and transformation processes to be classified to the same division, due to similarities in the intent or purpose of the outputs produced. For example, hospitals and homoeopaths are both classified to the Health Care and Social Assistance Division, based on their common purpose to improve human health. This is typical of the Service Divisions.
2.26 At the division level, the main purpose is to provide a limited number of categories which provide a broad overall picture of the economy, and are suitable for summary tables in official statistical publications. Weight is also given to maintaining continuity with previous editions of the classification.
2.27 Between the highest (Division) and lowest (Class) levels of the classification, combination of groupings is driven by a number of factors including:
- improving the international comparability of the classification at the Sub-division level;
- identifying groups of economic activities with significantly different production functions; and
- promoting some of the more economically significant industries to this higher level of the classification.
2.28 For example, subdivisions created in ANZSIC 2006 specifically to improve international comparability include:
- Subdivision 12: Beverage and Tobacco Product Manufacturing;
- Subdivision 44: Accommodation;
- Subdivision 55: Motion Picture and Sound Recording Activities;
- Subdivision 58: Telecommunications Services;
- Subdivision 60: Library and Other Information Services; and
- Subdivision 70: Computer System Design and Related Services.
2.29 Examples of subdivisions created to better identify groups of economic activities with significantly different production functions include:
- Subdivision 26: Electricity Supply; and
- Subdivision 27: Gas Supply.
2.30 Finally, examples of subdivisions created to promote some of the more economically significant industries to this higher level of the classification include:
- Subdivision 11: Food Product Manufacturing; and
- Subdivision 77: Public Order, Safety and Regulatory Services.