23.127    The 2008 SNA recommends inclusion of part of households' non-market production within the production boundary and the use of a satellite account for recording the other part. The 2008 SNA production boundary includes subsistence production in agriculture, other goods produced by households for their own consumption, the own-account construction of dwellings and housing services provided by owner-occupied dwellings, and paid services of domestic servants in the household sector. Excluded are services generated from unpaid work, including services for the producing household, services for other households and volunteer and community work.

23.128    The 2008 SNA suggests that, in practice, goods produced in households for own use are to be included within the production boundary if the production is believed to be quantitatively important in relation to the total supply of those goods in the country concerned. The ASNA includes an imputation for the market value (less the input cost) of the more common types of such production in Australia (fruit, vegetables, eggs, beer, wine and meat) for inclusion in estimates of household final consumption expenditure. An estimate for such 'backyard production' is also included on the income side of the accounts, as part of gross mixed income.

23.129    A number of commentators, including Ironmonger¹¹³, have expressed concern that the production boundary records only a partial picture of the production of household goods and services and the accompanying use of capital and labour. For example, household members can obtain goods and services by buying them from the market. This activity is fully captured in the national accounts. Households can also produce goods and services entirely themselves, using their own labour and capital. While such production of goods will be captured if it is significant, the production of services (other than housing services provided by owner-occupied dwellings) is not measured in the national accounts. The use of market inputs would be measured in the national accounts to the extent that the non-market production of services involves the use of market inputs.

23.130    The exclusion of most forms of household non-market production of services from the national accounts is due, in part, to the difficulties in measuring non-market output. In particular, non-market activities, by their very nature, must be valued using imputations and it is not always clear what these imputations should be. Also, it is more difficult to define non-market production than to determine the scope of market activity. Because of these concerns, national accountants generally hold the view that broadening the accounts to include a wide range of non-market activity would produce a less useful tool for analysing overall economic activity.

23.131    Nonetheless, as economic activity crosses over from non-market to market, or vice versa, this can lead to distortions in the accounts. A classic example is the marriage of a housekeeper to his or her employer. Prior to the marriage, the housekeeper's output (presuming that housekeeper was being paid a wage) was included in GDP. After the marriage, the same output is excluded if the new spouse is not paid a wage; however, there has been no change in underlying economic activity. Only the institutional arrangements underlying the activity have changed.

23.132    In order to provide a more comprehensive picture of economic activity, the 2008 SNA suggests that satellite accounts be used. Household satellite accounts are where the concepts, in particular the production boundary, underlying the core accounts are altered, but they do this in such a way that there are clear linkages with the core accounts. These can be compiled in both monetary and non-monetary terms. Thus, it would be possible, for example, to make non-monetised comparisons based on time spent in formal and informal economic activity as well as to monetise unpaid work, if so desired. Therefore, a household satellite account can provide comprehensive information on household economic activity within a framework that is consistent with the core national accounts, without subjecting the core accounts to the vagaries associated with defining and measuring household non-market output.

23.133    It is possible to widen the scope of household activity to look at frameworks that encompass not only household production but also describe consumption, saving and accumulation of wealth activities in households. This could be done at either the macro or micro level; that is, at the level of the household sector as a whole or disaggregated by types of household. A macro framework has been developed by Eurostat and a provisional micro framework has been developed by the ABS. Statistics Netherlands has developed a framework that seeks to show macro-micro linkages. Each of these frameworks is discussed below.

23.134    As mentioned above, one of the main issues in measuring non-market household production is to determine an appropriate method for valuing the production. Three approaches have been suggested:

• the unpaid work approach;
• the input approach; and
• the output approach.

23.135    The most common method used to date has been the unpaid work approach, which takes account only of (unpaid) working time and its imputed value. ABS studies to date have used this approach.

23.136    The input approach values household production as the sum of the values of all its inputs: time use, intermediate consumption, and capital costs.

23.137    The output approach values household production at its imputed output value, in the same way that in-scope household non-market production is valued in the core national accounts.

23.138    The essence of this approach is to multiply hours of unpaid work, usually obtained from a time-use survey (TUS) by an appropriate wage rate. The first Australian unpaid work study, published in 1990, used data from a 1987 pilot TUS. Three basic methods of valuation were used:

• the opportunity cost method;
• the individual function replacement cost method; and
• the housekeeper replacement cost method.

23.139    Each of these methods used wage rates that were on a 'before-tax', or gross, basis.

23.140    The second study, completed in 1994, used data from the first national TUS of 1992, and retained these three methods. It refined the housekeeper replacement cost method, and also distinguished between a gross opportunity cost method and a more appropriate net opportunity cost method, based on after-tax wage rates. The individual function and housekeeper replacement cost methods remained on a gross basis.

23.141    The third Australian study, completed in 2000 and based on the results of the 1997 national TUS, used the same methods as the second study. The study also introduced a hybrid of the individual function and housekeeper replacement cost methods.

23.142    A more detailed discussion of this approach is outlined in the Unpaid work section below.

23.143    Under this approach, the household is regarded as a production unit in which commodities and services are produced by combining work, intermediate consumption and household durables. This approach allows for better integration of household production into the system of national accounts. The formula used is as follows:

23.144    This formula is similar to that used in the national accounts to value the non-market output of the general government and non-profit institutions serving households sectors. The input-based approach was used by the German Federal Statistical Office in its estimates of the value of German household production in 1992.

23.145    The input-based method is used to measure the non-market output of households, such that the value of labour component relates to unpaid labour. Accordingly, the observations in the preceding section on the unpaid work method also pertain to this component.

23.146    The taxes less subsidies on production component refers to transfer payments made by households to governments and vice-versa that are recorded as secondary income transactions in the core national accounts but would be considered to relate to non-market household production. These transfer payments would then be reclassified in the household satellite account.

23.147    The consumption of fixed capital component relates to the depreciation of household durables used in the household production process. In the core accounts, purchases of durables (e.g. motor vehicles, refrigerators, washing machines) by households are recorded as final consumption expenditures and not as capital formation. In the satellite accounts, household expenditure on consumer durables would need to be reclassified from final consumption to gross fixed capital formation.

23.148    The more difficult aspect of measuring the consumption of fixed capital component would be in actually determining the appropriate amounts of depreciation in each period. The 'perpetual inventory method' (PIM) would require information about the decline in the efficiency of assets as they age, asset lives, the distribution of these lives about the average life, and changes in the price of assets. The ABS currently provides estimates of the stock of household durables as a memorandum item in the national accounts balance sheets. These data could be used as a starting point for deriving estimates of consumption of fixed capital.

23.149    The intermediate consumption element would consist of goods and services acquired by households that are used up in household production. To the extent that this production fell outside the production boundary, measuring the associated intermediate consumption would require identifying and reclassifying expenditures treated as final consumption in the core accounts. For some goods or services, it would be reasonable to assume that all expenditure on them should be classified to intermediate consumption. For example, meat purchases would all be classified to intermediate consumption because meat products generally have to be prepared or cooked before they are ready for a meal. Other goods or services could be used in production or as final consumption. For example, ice-cream can be eaten as such or used as an ingredient in desserts. As it is usually eaten directly, it would probably be allocated to final consumption. On the other hand, fruit, even though it is eaten mostly fresh, might have to be allocated to intermediate consumption as most fruits that are eaten fresh need to be rinsed, peeled, stored and distributed. The alternative to allocating expenditure on a particular product to either intermediate consumption or final consumption would be to split expenditure based on studies of the use of the product.

23.150    In deciding which expenditures should be classified as capital and intermediate in the household satellite account, the ABS would consider work already undertaken internationally in this area.

23.151    Estimates of household production developed using the input-based method could be presented in their own right or used to develop alternative estimates to those shown in the core accounts.

23.152    In the output-based valuation method, the gross output from household non-market production is valued by multiplying the volume of household output for different activities by market-equivalent prices for each activity. The rationale for this approach is that market goods and services could replace those generated in the household; therefore, the most appropriate way of valuing household non-market production is to use the prices of similar market production. Under the output-based method, the gross value added in household production is equal to the value of gross output less the value of intermediate inputs (where intermediate inputs are as described in the preceding section).

23.153    This method is considered to be the best for comparisons with national accounting aggregates, which are generally based on the use of market prices for valuing output. Valuing output in this way ensures that outputs are valued independently of their inputs, and avoids problems arising because of productivity differences between market and non-market producers.

23.154    The output-based approach resolves the issue of the joint production of services through simultaneous or parallel uses of time. The value of the labour used simultaneously can be found by deducting intermediate inputs and capital costs from the market value of the joint outputs.

23.155    The data requirements for the output-based approach are extensive and not readily available, particularly data on the volume of household output for different activities and corresponding market-equivalent prices. For this reason, there have been very few output-based studies to date.

23.156    A satellite account for household production could be presented in the form of an input-output table. Such a presentation would provide breakdowns of the value added (into capital and labour components) and intermediate consumption (into the various types of products used up in the production of household output) for each type of household output. Non-household production would also be shown so that the relationships between the economic activity of households and that of the other sectors of the economy could be explored. Supplementary information on the volume of household outputs or the time spent in the production of the outputs could also be shown. The value of household outputs could be derived using either the input- or output-based methodologies.

23.157    Ironmonger and others have argued that the development of such an input-output table is essential for a proper analysis of household economic activity. Thoen lists the advantages of placing household production within an input-output framework:

    Household production can be linked to the SNA through the development of a satellite account with links through 'personal expenditures' which are common to both accounts: the complex interdependence between household and market activities in terms of the raw materials, intermediate goods and services, or labour inputs required to produce outputs can be analysed within a familiar accounting framework: the impact of macroeconomic policy on the 'household sector of the economy' can be analysed in terms of the substitutability of market supplied services for household production and the household capital/labour ratio and, consumer demand can be linked to the underlying household activities.¹¹⁴

23.158    Deriving a household satellite account in the form of an input-output table would be a more difficult exercise than deriving estimates of household production in aggregate because each of the components of production (labour, capital and intermediate consumption) would have to be allocated across the various types of household products. Ideally, this would be done based on studies of the various types of household activities. In the absence of pre-existing studies, it would be expensive to undertake such studies and it is highly unlikely that such expense could be justified. Alternatively, in cases where the allocation of a component is not clear-cut, indicators (such as the time spent on activities) could be used as a basis of allocation. This would reduce the usefulness of the input-output approach, as any analysis based on the relationship between inputs and outputs would be affected by (unknown) errors in the allocation process.

23.159    Eurostat commissioned Statistics Finland to develop a harmonised satellite system of household production. The Eurostat proposal is based on the European System of National and Regional Accounts (ESA 95), which is broadly consistent with the 1993 SNA. While Eurostat acknowledges that the output-based method has analytical advantages compared with the input-based method, it advocates the latter as the basis for measuring household production as there are currently insufficient data available to implement the former. The proposal however recognises that an output-based method could eventually be implemented. The focus of the system is the production account. The proposal has guidelines for adjusting the core income and capital accounts to provide comprehensive information on the consumption, income, saving and wealth of households. Such information would increase the analytical usefulness of the system as a whole.

23.160    If a satellite production account could be compiled that covered household production comprehensively, relatively little effort would be required to compile consistent income and capital accounts along the lines suggested in the Eurostat proposal.

23.161    The ABS has been at the forefront in the development of a conceptual framework for household income, consumption, saving and wealth (ICW). This framework was developed by the ABS in response to the process of revising the provisional 1977 United Nations (UN) Guidelines on Distribution of Income, Consumption and Accumulation of Households (known as M61). The UN guidelines were issued to assist countries to collect and disseminate income distribution statistics and to provide for international reporting and publication of comparable data. The provisional guidelines had a particular emphasis on linking income distribution statistics to current national accounting standards; they relate to the 1968 version of the System of National Accounts. There have been continuing demands for revisions to the 1977 UN guidelines to supplement the 1993 SNA. In particular, a need is seen to broaden the concept of income and develop analytical techniques to measure income inequality.

23.162    The ABS framework, published in A Provisional Framework for Household Income, Consumption, Saving and Wealth, describes how the range of flows and stocks of household economic resources can be brought together to provide a comprehensive measure of economic wellbeing for individual households. The framework also provides a conceptual link between these components of individual household economic wellbeing and those of the national economy as a whole. As such, the concepts and terminology used in the ICW framework are consistent with those used in the national accounts. Concepts, definitions and terminology have been modified where necessary because the focus of the ICW is on the individual household, rather than the household sector.

23.163    More specifically the framework is designed to allow for the measurement of:

• a household's power or command over economic resources;
• the extent to which a household is able to both consume and accumulate wealth and to make choices between these options; and
• the changes that take place in a household's economic wellbeing over time.

23.164    Together, these measures constitute a model that reconciles the various elements of income, consumption and net worth at the individual household level. Such a reconciliation will enable derivation of measures of both household saving and total accumulation of wealth. The ICW presents a synthesis between economic and social statistics, particularly as they relate to the household economy. The framework, however, has a provisional status, and the ABS has not yet begun to make it operational.

23.165    It is worthwhile elaborating on the differences between the ICW and the Eurostat satellite accounting system for the household. Eurostat's system focuses on the macro-side of the economy, with the household sector being the main statistical unit. The production side of the household economy tends to be the central area of analytic interest. Production and generation of income accounts are seen as 'crucial' for the system of household satellite accounts. The input-output tables are a detailed elaboration of the household production account but represent only one part of a system of accounts for the household sector. The Eurostat framework does not provide as detailed an insight into the income flows and financing decisions of households as the ICW system does. It has guidelines for an extended system of accounts that describes the behaviour of the household sector in relation to consumption, disposable income and wealth. The ICW focuses on the micro-side of the economy, with four possible statistical units to measure the economic wellbeing of the population: persons, households, families and income units. The area of analytic interest is broader than the Eurostat central focus on production, with emphasis on how economic resources are mobilised within households and affect the different variables of household income, consumption, saving and wealth. Considerable work would be needed to further integrate the two systems. Integration would give better information on the dynamics of individual households, the household sector and the linkages to the market economy.

23.166    The Netherlands Central Bureau of Statistics started the development of satellite accounts of household production in 1991, using data from their national TUS conducted in 1987. Another TUS was conducted in 1998 and the data from that TUS are to be incorporated in the development of a System of Economic and Social Accounting Matrices and Extensions (SESAME), a form of a social accounting matrix (SAM). Traditionally, SAMs have been applied to specific types of analysis, focusing on causes and consequences of various aspects of inequality among household groups.

23.167    A SESAME is a 'core' SAM that has associated satellite tables and it provides a set of monetary and non-monetary macro-indicators potentially encompassing social, economic and environmental change. Such an integrated set of satellite tables can show:

• various stocks underlying the SAM flows, such as size and composition of the population by household group (including the potential labour force), production capacity by industry and the possession of assets (e.g. agricultural land, consumer durables and financial assets) and liabilities (e.g. external debts) by sub-sector;
• a decomposition of (changes in) values into (changes in) volumes and prices: this refers not only to products but also to various categories of labour services, and to fixed capital formation by industry;
• related non-monetary socioeconomic indicators, such as life expectancy, infant mortality, adult literacy, nutrient intake, access to (public) health and education facilities, and housing situation by household group; and
• some re-routings (e.g. final consumption by household group paid for by government and non-profit institutions serving households).

23.168    The Dutch system utilises an extra layer of operational principles to define the framework of its SAM for the household economy. In addition to defining productive activities by utilising the third person criterion (Chapter 3), it also defines productive activities in terms of formal and informal activities. The operational principle used is as follows:

23.169    Informal activities are productive activities which do not contribute to the national income as currently defined, and in which unpaid labour is involved.

23.170    When these principles are operationalised into a SAM there is a consistent representation of both the production processes and the income distribution and income spending processes. The system makes explicit the linkages between the formal and informal economies in terms of production and income generation and distribution and has entry points for data on fixed capital and consumption of fixed capital in the informal economy.

23.171    A key feature of this system is that goods and services in the informal sector are not always given a monetary valuation. The SAM is split up into two parts, making the module independent from the valuation of informal labour. The first part shows all transactions, expressed in monetary value, with the value of informal labour as zero. In the second part, all informal transactions are expressed in informal labour equivalents, such as working years, derived from the TUS. The Dutch see this framework as allowing formal and informal labour to be merged without disturbing the consistency in valuation. Various kinds of multiplier analyses can be applied where pricing is not a prerequisite. Future extensions will include informal fixed capital formation and a further disaggregation of labour by type, for example, by education level or position within household. Any number of disaggregations and micro-economic analyses could be made if the data were available, for example, data by household type. At this stage the framework is in place but very few of the cells have been developed.