Unpaid work

23.172    In the 2008 SNA, the 'general production boundary' is defined in paragraph 6.24 as encompassing all activities 'carried out under the control and responsibility of (institutional units) that use inputs of labour, capital and goods and services to produce outputs of goods and services'. Most unpaid work fits within this definition. The production boundary used in the 2008 SNA is more restricted than the general production boundary. Production is defined in paragraph 6.27 as excluding the value of most unpaid work and as comprising:

1. the production of all goods and services that are supplied to units other than their producers, or intended to be so supplied, including the production of goods or services used up in the process of producing such goods or services;
2. the own-account production of all goods that are retained by their producers for their own final consumption or gross capital formation;
3. the own-account production of knowledge-capturing products that are retained by their producers for their own final consumption or gross capital formation but excluding (by convention) such products produced by households for their own use;
4. the own-account production of housing services by owner-occupiers; and
5. the production of domestic and personal services by employing paid domestic staff.

23.173    Therefore, the 2008 SNA excludes from production all own-account production of services (which are the equivalent of 'unpaid household work') within households other than services produced by employing domestic staff and housing services produced by owner-occupiers. The 2008 SNA also omits from production the value of volunteer and community work that is provided free by householders to non-profit institutions or other households. The value of this work is not included in the costs of production of the recipients of the services generated by the unpaid work.

23.174    A prerequisite for the measurement of total unpaid work is a satisfactory definition of what constitutes such work. The boundary between productive and non-productive activity is not clearly distinguishable in many cases. For example, the distinction between unpaid work and leisure is often very difficult to draw.

23.175    A widely accepted principle for determining the scope of total unpaid work is the 'third person' or 'market replacement' criterion originally stated by Reid in 1934, and re-quoted by many writers:

23.176    Household production consists of those unpaid activities which are carried on, by and for the members, which activities might be replaced by market goods or paid services, if circumstances such as income, market conditions and personal inclinations permit the service being delegated to someone outside the household group.¹¹⁵

23.177    Under this criterion a household activity would be considered as unpaid work if an economic unit other than the household itself could have supplied the latter with an equivalent service.

23.178    Arguments can be made for and against the inclusion of some of the activities that would qualify as unpaid work by applying Reid's 'third-party' criterion. Many household activities that meet the Reid criterion, for example cooking and shopping, could be considered leisure activities in a number of circumstances. In the Australian studies, travel to/from work has been excluded from unpaid work because it is not possible to hire someone to travel to work on one's behalf and it is also clearly associated with paid employment rather than household production. Work done from home in relation to paid employment (for example, telephoning clients) which is unpaid but frequently a necessary part of the job, has been classified with paid activities and excluded from estimates of unpaid work. The unpaid assistance provided by relatives and others in family businesses has also been excluded, as the value added by such activities is already included in production in the national accounts.

23.179    Caring for others, for example playing with children, from some perspectives is a debatable inclusion in unpaid household work, even though it satisfies the third person criterion. Individuals perceive the status of these activities differently. Some people would view the raising of children as unpaid work, while others would view it as something more akin to leisure. Some would argue that these caring activities should not be classified as work or leisure but something else—they are activities that satisfy biological and cultural codes of behaviour to ensure desirable outcomes for the whole of society. Despite these reservations, the Australian studies include caring in the scope of unpaid household work. The ABS continues to recognise that the distinctions between paid work, unpaid work and leisure are still subject to world-wide debate and refinement.

23.180    In summary, unpaid work is defined in the Australian studies as comprising unpaid household work and volunteer and community work. Unpaid household work consists of domestic work about the house, childcare and shopping and associated communication and travel. The following list gives a broad indication of the activities included under various categories of unpaid work.

• domestic work has been classified into broad groups as follows:
  • food preparation and clean-up: includes the cooking and serving of meals, and washing dishes;
  • laundry and clothes care: includes washing, ironing, mending and making clothes;
  • general housework: includes cleaning the bathroom/toilet, vacuuming, dusting and tidying;
  • grounds and animal care: includes gardening, pool care and feeding and tending to animals;
  • home maintenance: includes repairs or improvements to the home, equipment, and motor vehicles and chopping wood; and
  • household management: includes paperwork, bills, budgeting, organising, packing, selling household assets and disposing of rubbish.
• childcare – includes the physical, emotional and educational care of children and general interaction with, and supervision of, children;
• shopping – includes the purchasing of a wide range of goods and services for people in the household - purchasing durables and consumer goods and purchasing repair services, administration services, childcare, domestic and gardening services etc.;
• volunteer and community work – includes the physical care of adults, doing favours for others and active involvement in various forms of unpaid voluntary work; and
• all communication and travel associated with unpaid work is also included within the scope of unpaid work.

23.181    Two basic approaches to measuring unpaid work are identified: the 'direct' or 'output' method; and the 'indirect' or 'input' method. The first method involves the measurement of output by direct observation of prices and requires data on the quantities of services produced. This method is considered to be conceptually superior because it adopts the same approach as that as used to value market production and is therefore appropriate for comparisons with national accounting aggregates. In general, data to apply the output method are not available and the ABS, like most statistical agencies, has used ‘indirect' or 'input' methods to measure the value of unpaid work.

23.182    'Indirect' or 'input' methods involve valuing output in terms of the cost of inputs and require information about the time spent on household work which, in Australia, is provided by TUSs. It is similar but not identical to the approach adopted in the 1993 SNA for valuing other non-market output, for example, non-market services produced by government. However, non-market output is valued using all relevant costs of production. In valuing unpaid work, the ABS and most other practitioners use only labour inputs. There are two broad approaches to this application of the input method:

1. The market replacement cost approach: that is, what it would cost households in wages to hire others to do the household work for them. Three variants of this approach are:
   • individual function replacement cost approach;
   • housekeeper replacement cost approach; and
   • replacement cost hybrid approach.
2. The opportunity cost approach: that is, what household members would have earned in wages had they spent the same amount of time on paid work as actually spent on unpaid work. Two variants of this approach are:
   • gross opportunity cost approach; and
   • net opportunity cost approach.

23.183    The ABS recommends the replacement cost approaches in preference to the opportunity cost approaches. However, estimates based on opportunity cost have continued to be derived to provide data that can be compared with opportunity cost estimates produced in the past or by other countries.

23.184    The individual function replacement cost approach assigns values to the time spent on unpaid work by household members according to the cost of hiring a market replacement for each individual function. Thus, for example, time spent on cleaning is valued using a rate of pay for commercial cleaners, and time spent on child minding is valued according to the rate of pay for childcare workers. Use of this method is based on the key assumption that household members and market replacements are equally productive in their work activities.

23.185    It is not easy to determine an appropriate market rate of pay for household activities because commercial rates may embody a level of skill, responsibility or capital not required or reflected in household work.

23.186    The estimates derived using the replacement cost approaches will underestimate or overestimate the contribution of unpaid work to GDP depending on the relationship between the productivity of households and the market sector. If households and market producers are equally productive, that is, if they have the same average output per hour, the replacement cost approaches undervalue unpaid work by ignoring the contribution of non-labour inputs (e.g. capital). If households are more productive, the replacement cost approaches further understate the value of unpaid work because a household will do more work in a given time than a replacement would. If, on the other hand, households are less productive, (if they have, say, access to less capital or technology), the value estimates will be too high because they will be derived by multiplying market wages by the longer time that will be taken by households to do the same amount of work.

23.187    Two questions concerning the choice of an appropriate average wage concept are:

• Whether gross or net wages are most appropriate?
• Whether actual or paid working time should be used?

23.188    In the Australian compilations for 1992 and 1997, a gross wages concept, weekly ordinary time earnings, was preferred.

23.189    It could be further argued that total labour costs should be used, including employers contributions to superannuation, fringe benefits and workers compensation schemes. However, such data are not available in Australia by occupation on a per hour per employee basis. For this reason estimates of unpaid work including such additions to gross wages have not been calculated.

23.190    Regarding use of actual or paid working time, the latter is determined by law or collective agreements, and includes paid holidays and paid sick leave. Actual working time refers to the time spent actually working and includes paid and unpaid overtime but excludes public holidays and weekends (except in cases where a worker does work at those times).

23.191    In Australian studies the concept of paid working hours was used. Hourly wage rates were obtained by dividing weekly ordinary-time earnings by ordinary-time hours paid for. (Ordinary-time excludes overtime.) 'Ordinary-time hours paid for' refers to employees' standard or agreed hours of work that are paid at the ordinary-time rate. It includes stand-by or reporting-time hours that are part of standard hours of work, and any part of annual leave, paid sick leave or long service leave taken during the reference period.

23.192    The housekeeper replacement cost approach values the time spent on unpaid household work by household members according to the cost of hiring a housekeeper to undertake the relevant tasks.

23.193    The key assumption underlying this approach is that household members and housekeepers are equally productive in performing household work, which may or may not be true. For example, a housekeeper is likely to be more productive at cleaning than a household member who may also be looking after small children. Alternatively, a housekeeper may clean more quickly but less thoroughly than the household member. Use of this approach also assumes that there is a well-established labour market for persons who undertake all household tasks, which is not the case in Australia in the 1990s.

23.194    In both the 1992 and 1997 estimates, a female wage rate for domestic housekeepers was used as there was no male wage rate available. The tasks identified in the category included:

• preparing, cooking and serving meals and refreshments;
• purchasing food and household supplies;
• washing dishes, kitchen utensils and equipment, sweeping and washing floors and vacuuming carpets, curtains and upholstered furnishings;
• dusting and polishing furniture, and cleaning mirrors, bathrooms and light fixtures; and
• washing and ironing garments, linen and household articles.

23.195    In the 1992 and 1997 estimates, the housekeeper wage rate was used to value all household tasks including those that would not normally be undertaken by a housekeeper.

23.196    The housekeeper replacement cost approach is only applicable to the derivation of the value of unpaid household work and does not apply to the derivation of the value of volunteer and community work. Estimates of the value of total unpaid work under a housekeeper replacement cost heading are derived by adding estimates of unpaid household work derived using the housekeeper replacement cost approach to estimates of volunteer and community work derived using the individual function replacement cost approach, based on the persons wage rate.

23.197    Under this approach, the housekeeper wage rate was applied to those tasks normally carried out by a housekeeper (as described in the previous section). The value of tasks not normally undertaken by a housekeeper was estimated using the wage rates employed in the individual function replacement cost approach. The hybrid approach would appear to be appropriate given that Australians typically hire housekeepers to clean house interiors, manage laundry and occasionally prepare meals while they hire specialists to carry out childcare, household maintenance and gardening tasks. The hybrid approach was not used in the compilation of the published estimates for 1992.

23.198    The gross opportunity cost approach values unpaid work in terms of the earnings assumed to be foregone by householders when they devote time to unpaid work rather than paid employment. The approach assumes that the value of time spent doing unpaid work at home equals its 'opportunity cost' elsewhere, i.e. the valuation of the next best alternative use. The assumption is made that the worker has given up paid work to perform unpaid work and that its value per hour is equal to the individual's marginal hourly wage in the market. In other words, to do an extra hour of unpaid work, an hour of paid (market) work is given up. However, in practice, the total time spent on unpaid work is multiplied by the average wage applicable to relevant groups in the population.

23.199    There are many problems with this approach. Some reservations are outlined below:

• Labour market structures – the gross opportunity cost approach does not represent the way the choice between paid and unpaid work is made. Most workers have limited choice in the short run regarding the hours they have to work and few have the option to refuse overtime.
• Employment status – this method does not hold up well when patterns of labour force participation other than a rigid fixed-hours working week are considered.
• 'Psychic income' – a probably unquantifiable but theoretically precise valuation would take account of the worker's net psychic income from doing unpaid work and from doing paid work, i.e. where the net psychic benefit equals the psychic benefit from doing unpaid work minus the psychic benefit from paid employment.
• Relevance of market wage rates for individual workers – the question is how is the 'foregone wage' to be determined when an unpaid household worker has had no market employment and therefore an indeterminable potential wage?

23.200    Opportunity cost measures tend to be higher than those for the housekeeper replacement cost approach because wages for professional housekeepers are lower than the economy-wide average wage which is used for the opportunity cost approach. In view of the above discussion, the opportunity cost approach will give useful results only if very strict and probably implausible assumptions apply:

• at the margin, time devoted to unpaid work precludes market work;
• the value of time at the margin is gross hourly wages; and
• the average potential hourly earnings of the not employed are equal to the average hourly earnings of the employed.

23.201    In the Australian studies relating to 1992 and 1997, estimates of unpaid work using the gross opportunity cost approach have been produced using average male and female wage rates, and also using the persons' average wage rate. The ABS regards the gross opportunity cost approach as the least appropriate of the estimation methodologies.

23.202    The decision to undertake paid work as an alternative use of time to unpaid work could reasonably depend on the remuneration (wages and salaries, superannuation, and fringe benefits) after tax and any work-related costs. The net opportunity cost approach recognises this and recommends valuation of unpaid work at the after-tax hourly wage rate less work-related expenses plus income by way of employer costs of superannuation and fringe benefits.

23.203    The rationale behind this approach is that the unpaid worker will be equating the value of doing unpaid work with the net benefit of working in paid work conferred by this 'adjusted' hourly wage rate. In the Australian context, estimates of the wage rate applicable to a net opportunity cost valuation are derived by subtracting from average annual ordinary time earnings, the relevant taxes and levies payable and work-related expenses, and adding to the result the imputed employer on-costs relating to superannuation and fringe benefits and then converting the final result to an hourly wage rate.

23.204    The calculation of the net opportunity cost wage rate is really an attempt to find the appropriate and most realistic net wage rate. Ideally, a number of other factors should also be taken into account, if it were possible to measure them. For example, there is the question of whether tax rebates and social security benefits should be considered. In the Australian context, these include family allowances, family income supplements, medical insurance rebates, and a variety of means-tested welfare programs. Indeed, the net opportunity cost of working in the paid work force could well be negative in some cases.

23.205    The preparation of the 1997 estimates of total unpaid work presented in this paper required the following sets of data:

• estimates of average time spent on household work obtained from the 1997 TUS;
• population estimates from the Census of Population and Housing;
• appropriate wage rates from the Survey of Employment, Earnings and Hours as well as the Labour Force Survey; and
• work related expenses from the Household Expenditure Survey.

23.206    The main limitations of the unpaid work approach are:

• the contribution to the value of household production that comes from other inputs (e.g. intermediate consumption and capital) is not captured;
• it cannot provide information on the labour productivity of household production;
• it cannot be used to analyse whether households are more efficient in their production than comparable market units;
• it does not take into account the joint production of services through simultaneous or parallel uses of time; and
• a choice is required among multiple wage rates and valuation methodologies, each of which have their limitations.

23.207    With regard to the last point, the pros and cons of the various approaches are well summarised in a 1999 Eurostat document entitled, Proposal for a satellite account of household production.¹¹⁶ (This proposal is discussed in more detail in the section 'The Eurostat Proposal'.) The document provides the broad consensus of national accountants in Europe about the preferred method of valuation of the labour input into household production. Of the possibilities, Eurostat recommends that the housekeeper replacement cost method is the most appropriate method to use to value household labour. The reasons cited include:

• the nature of the work performed by a housekeeper is rather similar to the nature of housework performed by a household member;
• housekeeper productivity is similar to that of the householder, as regards the performance of several household activities simultaneously, the quality of household equipment used and the amount of intermediate consumption involved; and
• the method of valuation is simple and straightforward.

23.208    A potential problem with the housekeeper replacement cost method is that a housekeeper does not perform all the tasks undertaken in households, such as household management, home maintenance, servicing vehicles and volunteer work. Using this approach then could possibly see an undervaluation of the labour input to household production unless those tasks not typically undertaken by a housekeeper were also included in the valuation on the basis of specialists' wages (in effect, utilising a hybrid of the housekeeper replacement cost and individual market replacement cost methods).

23.209    As to the issue of gross or net wages Eurostat recommends that gross wages, which include income tax and social security contributions paid by the employer and employee, be used, although it acknowledges that net wages do have certain advantages from a theoretical point of view. However, net wage statistics are generally not available on an occupational basis. (This is one of the reasons that Australia uses the gross wage concept in its valuation of unpaid work using the replacement cost methods).