6503.0 - Household Expenditure Survey and Survey of Income and Housing, User Guide, Australia, 2015-16  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 06/07/2018   
   Page tools: Print Print Page Print all pages in this productPrint All



Household Expenditure Survey (HES) estimates are used to analyse the expenditure patterns of households across the population and to compare levels of expenditure between various population subgroups. These analyses support the development, implementation and evaluation of social and economic policies, particularly for potentially disadvantaged groups such as pensioners, one-parent families and the unemployed. The HES estimates of expenditure are also used in determining the ‘basket of goods’ that is used to compile the Consumer Price Index (CPI) as well as the relative importance of each expenditure class in calculating the index.

The Australian System of National Accounts (ASNA) also provides estimates of expenditure, which is comprised of the household sector and Non-profit Institutions Serving Households (NPISH). The ASNA estimates of expenditure are compiled from many sources, most of which do not provide information for different population subgroups within the household sector. The ASNA is designed to provide a systematic summary of economic activity of the Australian economy and to present a statistical picture of the structure of the economy and the detailed processes that make up domestic production and its distribution. Within the national accounting framework, the data show how the household sector relates to the corporate and government sectors in Australia and to the rest of the world.

As the HES and ASNA estimates of household expenditure have been developed on a different scope basis for different purposes, there are a number of differences in the resulting estimates of expenditure. This comparison presents data on the consumption expenditure of households from the two data sets and describes and quantifies some of the main scope, definitional and methodological differences between them.


The HES and the ASNA estimates of expenditure on goods and services are presented for the years 2003-04, 2009-10 and 2015-16 (Tables 1 and 2 of data cube 'HES-ASNA expenditure comparison').

The HES is conducted six-yearly and enumerated over a 12 month period. For most types of expenditure, data are taken from diaries in which survey respondents record their expenditure over a two week period, either by attaching their shopper dockets or by writing down their expenditures. Diary derived estimates therefore refer almost entirely to expenditure during the reference year. Estimates for infrequently purchased or more expensive items, or for services which are billed periodically, are derived from the household questionnaire which collects expenditure information for goods and services on a recall basis (often supported by billing documentation). These less frequently occurring expenditures are collected for recall periods longer than the two week diary reporting period so that sufficient numbers of households report expenditure to enable the calculation of expenditure estimates with relatively low sample error. For example, in 2015-16, survey respondents were asked to recall how much they spent on furniture and appliances over the last three months, and on motor vehicle registration over the last 12 months. For other items, such as insurance, rent and utilities bills, survey respondents are asked for the value of their most recent bill and the length of time to which the payment relates. Estimates derived from the household questionnaire therefore refer to some time periods prior to the reference year as well as during the reference year.

The estimates of expenditure are initially coded to a detailed household expenditure classification and the results in the Household Expenditure Survey, Australia: Summary of Results, 2015-16 (cat. no. 6530.0) product are generally presented according to that classification. However in this comparison (and Data Cube 12 of the summary of results product), the HES expenditures have been concorded to the United Nations' Classification of Individual Consumption by Purpose (COICOP), which broadly aligns with the classification used in the ASNA. All estimates have been converted from weekly to annual figures.

The ASNA estimates of expenditure in this comparison are from Table 42 'Household Final Consumption Expenditure’ (HFCE) in Australian System of National Accounts, 2016-17 (cat. no. 5204.0). The ASNA estimates consist of expenditure by households and by private non-profit institutions serving households (NPISHs) on goods and services. The ASNA is compiled from many data sources, most of which derive from statistical surveys or as a by-product of government administrative processes. Details are available in Australian National Accounts: Concepts, Sources and Methods (cat. no. 5216.0).


There are a number of scope and measurement differences that can be broadly quantified between the HES and the ASNA household expenditure estimates (Table 3 of data cube 'HES-ASNA expenditure comparison').

The HES estimates of household expenditure include the imputed rent for subsidised rentals which are not included in the expenditure attributed to the household sector in the ASNA ($8 billion in 2015-16). For comparison purposes, this expenditure is deducted from the HES estimates in Table 3.

The ASNA household sector estimates of expenditure also include some components which are not included in the HES measure. The components which can be identified, and where a deduction from the ASNA aggregate can be estimated, totalled approximately $103 billion in 2015-16, and include:

    • imputed service charges for life insurance and superannuation funds, imputed financial intermediation services indirectly measured (FISIM), and an imputed service charge for workers' compensation insurance (together estimated at $67 billion);

    • current expenditure of NPISHs such as churches and charities on goods and services (estimated at $36 billion

The net result of adjusting for the quantifiable differences shown in Table 3, is to bring the aggregate results closer together. In 2015-16, the adjusted HES expenditure was $710 billion compared with adjusted ASNA expenditure of $865 billion. However, there are three remaining scope differences that cannot be easily quantified. The HES excludes both people living in non-private dwellings and people living in very remote regions of Australia. It is estimated that this excludes approximately 3% of people living in Australia. The ASNA household sector estimates also include expenditure by NPISHs across all consumption categories. While NPISH expenditures made for education, health and 'other services' are estimated to be about $36 billion in 2015-16, the remaining expenditure by NPISHs is not quantified and therefore not allocated by commodity.


The following section compares expenditure on selected HES items with expenditure items published in the ASNA, or with the source data available for those ASNA expenditure items.


In the HES, expenditure on most goods is collected from diaries in which survey respondents, aged 15 years and over, record their expenditure over a two week period. The range of goods includes, for example, food, clothing, household consumables and personal effects. Goods that are purchased less frequently and that are of higher value, such as household appliances, furniture and floor coverings, and audio visual and photographic equipment, are collected from a household spokesperson using a three month recall period. The estimates are compiled net of trade-ins where applicable.

ASNA expenditure estimates for goods are benchmarked primarily to the Retail and Wholesale Industry Survey which is conducted about every seven years (last conducted in 2012-13). The monthly Retail Trade Survey is then used to extrapolate movements in expenditure between benchmark years. A range of other sources are used for expenditure for which retail surveys are not able to provide an accurate measure. Other sources include information obtained about sales by manufacturing and wholesale establishments and sales by organisations selling directly to the public.

The ASNA includes an estimate for the value of goods produced at home for own consumption, such as home grown fruit and vegetables. This consumption is not captured in the HES. This component is not separately identifiable in the ASNA estimates and has not been deducted in the comparisons.

In 2015-16, the adjusted HES estimate for expenditure on the selected goods listed in Table 4 (of data cube 'HES-ASNA expenditure comparison') was $179 billion, compared to $245 billion for the same goods in the ASNA. Part of the lower estimate for HES can be accounted for by scope differences, with the exclusion of expenditure by people living in non-private dwellings and very remote regions, as well as expenditure by NPISHs which cannot be quantified at the component level. However, it is also possible that respondents in the HES omitted to report some expenditure, particularly small value purchases for which no shopping receipt is received.

Alcoholic beverages and tobacco

HES estimates of expenditure on alcoholic beverages and tobacco are reported by respondents in their two week expenditure diary.

ASNA benchmarks for alcoholic beverages are compiled using data from the Retail and Wholesale Industries Survey and other ABS surveys of businesses in the hospitality industry. These benchmarks are extrapolated forward using monthly retail sales data. Tobacco expenditure is estimated using information from the Australian Customs Service from documents lodged for excise purposes as well as for imports and exports.

In 2015-16, the HES estimates for expenditure on alcoholic beverages ($10 billion) and tobacco ($6 billion) were much lower than the ASNA estimates ($19 billion and $17 billion, respectively). It is generally acknowledged that expenditures on alcohol and tobacco are underreported in household expenditure surveys.

Housing rent

Actual rent

Information on household expenditure on rent for housing is collected in the biennial Survey of Income and Housing (SIH) as well as in the HES which is a subsample of SIH respondents. In the HES and SIH, the actual gross rent paid by renters is reported during the interview with the household spokesperson.

The COICOP item 'Actual rentals for housing' includes not only rent payments for the selected dwelling, but also includes rent payments for other properties and non-holiday accommodation to better align with ASNA estimates. Rent payments for other properties and non-holiday accommodation are not collected in SIH only years.

In producing the ASNA imputations of gross actual rent, Census data are used to provide a benchmark for rental costs and the stock of dwellings. The rent component of the CPI and other factors are used to extrapolate the rental benchmarks forward, while the benchmark for the stock of dwellings is updated using ABS building completions data and an estimate for demolitions and other stock changes.

In 2015-16, the HES and ASNA estimates of actual rent were $50 billion and $45 billion, respectively.

Imputed rent for owner-occupiers

In the HES, the services for owner-occupied housing are estimated as the market value of the rental equivalent. The market value is estimated using the relationship between dwelling price (available from Valuers General data) and the market rent that a dwelling would receive for the smallest available geographic area (available from Census data). Rental yields based on average dwelling prices and reported rents have been calculated for each rental dwelling in the Census. These rental yields are then aggregated by strata to derive a final rental yield which is applied to the HES reported value of each owner-occupied dwelling. Imputed rent is estimated for the primary residence of owner-occupiers.

The COICOP item 'Imputed rent for rentals' includes not only imputed rent for owner-occupiers, but also the subsidies received by renters. For the purpose of this confrontation, these subsidies have been deducted from the COICOP estimates, as shown in Table 6 of data cube 'HES-ASNA expenditure comparison'.

Imputed rent in the ASNA is estimated for owner-occupiers not only for their primary residence, but also from any secondary residences such as holiday homes.

In 2015-16, the HES and ASNA estimates of imputed rent for owner-occupiers were $144 billion and $142 billion, respectively.

Purchase and operation of own vehicles

HES estimates of expenditure include new and second hand vehicles purchased in the last 12 months, net of trade-ins or vehicles sold in the same period. Fees and charges associated with the purchase, such as stamp duty, are excluded. From 2009-10 HES, more detailed commodities information was collected for salary sacrificed motor vehicles to improve the separate estimates of expenditure on motor vehicles and any associated running costs such as fuel, insurance, registration, servicing and tyres. Also from 2009-10 HES, motor vehicles provided free or at a reduced cost by employers to their employees for their own private use were included in the expenditure estimates.

The ASNA estimates of household sector expenditure include the total value of new and second hand vehicles purchased, excluding stamp duty and other fees. For second hand vehicles traded through dealers, the purchased trade-in values are netted so that only the dealers' margins are included, consistent with the HES treatment. Similarly, the ASNA aggregate estimates for the household sector do not include private sales of used vehicles between households as this expenditure is netted out in the aggregate accounts, consistent at the total sector level with the HES treatment. ASNA expenditure on the operation of vehicles is estimated using data from the Survey of Motor Vehicle Use (cat. no. 9208.0), Retail and Wholesale Industries (cat. no. 8622.0), Australian Industry (cat. no. 8155.0), State government car registrations and the HES. Salary sacrificed vehicles are included, as are the benefits the household sector derives from the personal use of vehicles provided by employers.

In 2015-16, the HES and ASNA estimates for the purchase of vehicles were $24 billion and $23 billion, respectively.

Education services

The HES collects education related expenditure during the last 12 months, including, for example, tuition, boarding fees and school sports fees. Repayments in the last 12 months against student loans, such as the HECS and HELP schemes, are also collected.

In addition to the direct expenditures reported in HES, the ASNA also includes current education related expenditures by NPISHs, such as students in the Catholic education system and most other private schools. These expenditures are estimated to contribute approximately 40% of the ASNA estimate (or $18 billion) in 2015-16.

After deducting expenditure by NPISHs, the ASNA education services expenditure estimate was $27 billion in 2015-16, compared with the HES estimate of $21 billion. Note also that the ASNA estimates include accrued expenditures against student loan liabilities at the time the debt accrues, rather than when the student repays the loan.


The HES collects information from the household spokesperson on household spending on medical and other health services for the previous three months, net of any reimbursements such as refunds from private health funds. Spending on health related goods, such as prescriptions and first aid supplies, are reported in the two week expenditure diary completed by all survey respondents aged 15 years and over.

There are two important differences between the ASNA and HES estimates of expenditure on health. Firstly, the ASNA includes expenditure by NPISHs on behalf of the household. Many private hospitals and nursing homes are run by NPISHs, and all expenditure by these institutions is allocated to health. Secondly, the expenditures on health by residents of non-private dwellings, such as nursing homes, and by people living in very remote regions of Australia are excluded from the scope of the HES.

In 2015-16, estimates for expenditure on health in the HES and ASNA (after removing NPISH expenditure), were $20 billion and $55 billion, respectively. However, the different treatment of refunds for health services, means that some of the ASNA expenditure is captured as insurance services expenditure in the HES.

Insurance and other financial services

The HES uses the group questionnaire to collect the most recent payments made for non-life insurance, for example, health, car and household contents insurance. Expenditure on building insurance is not collected for rental investment properties since expenses for rental properties are deducted from income when survey respondents report their profit or loss from this investment. Financial services in the HES includes only expenditure on direct fees paid to financial institutions such as account keeping fees and BPay fees. Interest on loans is not included. Estimates are compiled from fees reported on the most recent bank statement.

The ASNA estimates for insurance attributed to the household sector include only an imputed service charge for insurance services, which is calculated as the difference between premiums paid plus investment earnings on premiums (called premium supplements), less expected claims. The ASNA insurance estimates also include life insurance and superannuation funds, for which the imputed service charge is equal to the administrative expenses of operating the funds. The ASNA estimates also include any profits of non-mutual funds, and workers' compensation insurance, neither of which are measured in the HES.

Direct charges paid to financial institutions by households are included in ASNA, as they are in HES. However, other financial services expenditures attributed to the household sector in the ASNA include an imputed service charge paid by households for financial services associated with loans and deposits. These service charges are calculated as the difference between interest rates financial institutions pay on money borrowed from households (deposits) and money lent to households (loans). This service charge is called financial intermediation services indirectly measured (FISIM).

In 2015-16, HES estimates of expenditure on insurance and other financial services were $38 billion compared with the ASNA estimate of $17 billion (after excluding the imputed service charges). It is not possible to quantify all of the differences between the two estimates such as the exclusion of building insurance in the HES estimate and the differences in the measurement of insurance services between the two sources.

Other services

HES expenditure on services that are paid for at the time of use are generally collected in the two week expenditure diary. This includes expenditure on transport services (such as bus, rail and taxi fares), takeaway food and meals in restaurants, recreational activities (such as cinema and sporting tickets) and expenditure on travel for less than four days. For less frequent or more complex items, expenditure is collected on the group questionnaire, including the most recent bill for utilities and communication services, such as gas, electricity, internet service providers (ISPs) and telephone services. Expenditure on personal travel for four nights or more is also collected on the group questionnaire. Where possible, the cost of expenditure on individual travel items is collected and attributed to the relevant commodity. The exception is travel purchased as part of a package.

In the HES, child care expenditure is collected on the individual questionnaire from parents or guardians with children 12 years and under. The usual weekly cost of child care in the last four weeks is collected, with expenditure calculated net of child care benefits and the child care rebate.

In addition to HES data, the ASNA benchmark estimates of household sector expenditure on services are obtained from a range of sources, including:
There are differences between HES and the ASNA in the categorisation of some services. In the ASNA, expenditure on package holidays is separated out so that expenditure on individual services is attributed to the relevant item e.g. transport services, accommodation services, recreational and cultural services.

In total, expenditure in 2015-16 on the selected services, was $164 billion in the HES. About 30% of the ASNA expenditure on 'Other services' is estimated to be miscellaneous expenditure by NPISHs. After deducting this NPISH expenditure, the ASNA estimate was $212 billion in the same year. Other expenditure by NPISHs is not able to be quantified and is therefore excluded. The ASNA household sector expenditure estimates also cover a broader scope of accommodation services than the HES, encompassing the costs of hostel accommodation for the aged or disabled (excluding those providing medical supervision).

Further information

Comparisons between the ASNA sector level estimates and the household survey estimates for income and wealth are also presented in this product. Data cubes for each comparison are also provided in the 'Downloads' tab of this product.

A concordance between the HEC and COICOP classifications used to classify expenditure data from the HES is also available from the 'Downloads' tab of this product.


ABS (Australian Bureau of Statistics) 2017, Australian System of National Accounts, 2016–17, cat. no. 5204.0

ABS 2016, Australian System of National Accounts: Concepts, Sources and Methods, cat. no. 5216.0