Australian Bureau of Statistics
6527.0 - Household Expenditure Survey, Australia: User Guide, 1998-99
Latest ISSUE Released at 11:30 AM (CANBERRA TIME) 15/11/2001
|Page tools: Print Page Print All RSS Search this Product|
The household is adopted as the basic unit of analysis because it is assumed that sharing of the use of goods and services occurs at this level. If smaller units, say persons, are adopted, then it is difficult to know how to attribute to individual household members the use of shared items such as food, accommodation and household goods.
The HES produces estimates of average household expenditure on goods and services and selected other payments for the 1998-99 financial year.
Measurement of expenditure
Expenditure can be measured according to the following approaches:
For other items such as durable items and items purchased on credit which are not fully consumed or paid for during the recall or reporting period, the situation is different. Estimates for individual households will vary according to the approach adopted. For groups of households, however, the estimates will ‘average out’ so that the estimates for groups of households can be said to be indicative of payments and consumption as well as acquisitions.
For example, the 1998-99 HES collects expenditure on acquisitions of washing machines over three months. Say that we have a group of 1,000 households, and on average, 96% of them have washing machines. Of those who have washing machines, on average, over ten years, they fully consume their machine, acquire a new one and pay $700 for the machine in five equal instalments of $140.
Classification of expenditure
Expenditure is classified according to the Household Expenditure Classification (HEC) which is given in appendix 3.
The list shows the classification of goods and services, which is the primary focus of the HES. It also includes ‘selected other payments’ which comprise income tax, repayments on mortgage principal for the household’s place of residence, other housing costs of a capital nature such as internal renovations, and superannuation and life insurance.
Expenditure for private purposes
The HES provides estimates of expenditure on goods and services used for private purposes. It therefore excludes expenditure for business and other investment purposes. Operating expenses of unincorporated businesses are either not collected or are deducted from reported expenditure. If survey participants report business expenditure, it is picked up in questions in the household questionnaire or space provided in the diary, in which there is an opportunity to report amounts which ‘have been or will be charged to a business’. If amounts have been or are going to be charged to a business, then these are deducted from expenditure during processing.
Deduction of refunds and trade-ins
The HES measures net or ‘out of pocket’ private expenditure on durable goods, non-durable goods and services for private purposes. Estimates therefore do not refer to the full costs of goods and services used but only the costs payable by the household for goods and services used.
In the case of a refund which is received or expected, the amount of the refund is deducted from expenditure to produce a net figure. For expenditure on visits to general practitioners, for example, Medicare and private health insurance refunds are deducted.
In the case of trade-ins, these amounts are also deducted from expenditure to produce a net figure. For example, if the cost of a motor vehicle is partially financed by a trade-in of another, the amount of the trade-in is deducted from the cost for the acquired vehicle.
In the case of the sale of land, houses and motor vehicles, the sale price net of outstanding loans is deducted from expenditure and in the case of houses and motor vehicles, amounts of successful insurance claims are deducted from expenditure. Deductions are made even if there is no expenditure on that item by the household. Sales and claims made in the recall period for items which are not replaced during that period are included to compensate for sales and claims made outside the recall period for items replaced during the recall period.
Where trade-ins, sales and insurance claims exceed the costs of acquisitions of the same expenditure item, expenditure is recorded as negative. For example, if someone sells a luxury motor vehicle and buys a less costly model, the amount of expenditure recorded in the HES would be negative.
HES estimates of expenditure include the full retail value of employer-subsidised goods and services for food, alcohol, tobacco, clothing and footwear, and other items collected in the dairy (see table A2.1 to identify items collected in diary). Employer subsidies for other items, such as the use of vehicles, housing costs, electricity and telephone services, are not included because data collected on employer subsidies (or income in-kind) cannot be fully reconciled with data collected on business refunds.
Other in-kind expenditures, such as the consumption of vegetables grown by the household or provided by another household (not in return for labour) are excluded.
Timing of expenditure
The total period covered by expenditure estimates is a function of the recall or reporting period at the time of interview and the timing of interviewing. For the 1998-99 HES, interviewing was conducted throughout the 1998-99 financial year. For most types of expenditure, data were taken from diaries in which survey participants recorded their expenditure over a two week period, beginning the day after interview. Diary derived estimates therefore refer almost entirely to expenditure during the 1998-99 financial year.
Estimates for infrequently purchased or more expensive items are derived from the household questionnaire (see explanation in Data Collection section in chapter 3) which collects expenditure information for goods and services on a recall basis. These less frequently occurring items are collected over periods longer than the two week diary reporting period so that sufficient numbers of households report expenditure to enable the calculation of reliable expenditure estimates. For example, in 1998-99, survey participants were asked to recall how much they spent on motor vehicle registration over the last 12 months. Recall periods differ between items, ranging from the household’s last payment (which may be as short as the last week) for rent payments to two years for house purchases.
Table A3.1 (in appendix 3) indicates the items collected in the household questionnaire and their associated recall periods. In general, longer periods are used for items which are expensive, are acquired infrequently or are acquired at irregular intervals. Shorter periods are used for items which are purchased more frequently or are less significant and therefore not well remembered.
The use of different recall periods means that estimates for different expenditure items, in some cases, refer to different periods. The estimates of average expenditure on motor vehicle registration, for example, cover the 12 months prior to the beginning of interviewing to the end of interviewing (i.e. July 1997 to June 1999). For house purchases, the period is two years prior to the beginning of interviewing to the end of interviewing (i.e. July 1996 to June 1999). Household questionnaire derived estimates therefore refer to varying periods prior to the 1998-99 financial year as well as during the 1998-99 financial year.
Studies which use HES data tend to assume that all expenditure estimates refer only to the common reference period of July 1998 to June 1999. This is generally true for diary derived estimates but is a valid assumption for estimates derived from the household questionnaire only if expenditure prior to the 1998-99 financial year was the same as during the 1998-99 financial year.
For household questionnaire estimates, if the volumes or prices of purchases were lower during the period prior to the 1998-99 financial year, then average expenditure over the preceding period plus the 1998-99 financial year will be less than average expenditure over the 1998-99 financial year only. Similarly, if prices or volumes were higher during the preceding period, the HES estimate will over-estimate average expenditure in the 1998-99 financial year. The longer the preceding period (which is equal to the length of the recall period), the greater the likelihood of discrepancy. In cases where expenditure is expected to have changed, researchers may wish to acknowledge or adjust for these differences.
Weekly household expenditure
Estimates of weekly expenditure do not refer to any given week but are weekly equivalents. They are derived by dividing reported expenditure for all members of the household by the number of weeks in the relevant recall or reporting period. For household questionnaire items, recall periods vary from the last two years to the last three months, and for some items the last payment is reported (see appendix 3 for details). For diary items, the reporting period is two weeks.
Although the HES is primarily a survey of household expenditure, information is also collected on household income because:
The income data collected in the HES relates to usual cash income, that is gross receipts of recurring and usually regular cash flows. The resulting income estimates are a reasonable proxy for weekly cash income and can be used in their own right in income distribution studies.
Usual cash income
Usual cash income refers to income which is most frequently received over a given period rather than the income which is actually received. This is a better explanatory variable for average expenditure because it excludes variations in income which are unlikely to result in variations in expenditure. Week to week variations in actual or average income are unlikely to affect average expenditure because the financial obligations which drive expenditure are fairly stable.
Receipts which are excluded
Receipts which are not recurring and usually regular or are not cash flows are excluded from the HES. Examples include:
Sources of income
Income is collected according to source. Main sources of income include employee income, own business income, government pensions and allowances and other income (including property income such as rent, interest and dividends and other transfer income such as regular recurring receipts from superannuation and child support). A detailed list of the types of income for which HES estimates are available is given in appendix 2.
Employee income was collected in the 1998-99 HES from each person aged 15 years and over who worked for an employer or in his/her own limited liability business. Publication estimates of employee income are the sum of usual weekly pay, average weekly receipts from leave loading and regular bonuses, and the average weekly value of selected in-kind income from employers.
Usual weekly pay covers wages and salaries, tips, commissions, piecework payments, penalty payments and shift allowances, remuneration for time not worked (e.g. sick and holiday pay) and workers’ compensation paid through the payroll.
To obtain usual pay, survey participants are asked to report the amount of their most recent pay and what period the pay covers. They are then asked if that pay is usual, and if not, they are asked to supply a usual amount and the period covered. Estimates are based on the last (actual) pay if that pay is usual, otherwise on the reported usual pay. Pays are divided by the number of weeks they cover to produce estimates of usual weekly income.
To obtain information on leave loading and regular bonuses, survey participants are asked if they received any leave loading or regular bonuses in the last 12 months. If they do, they are asked to report the amounts received. The amounts are divided by 52 weeks to obtain equivalent average weekly income which, due to the length of the recall period, is considered to be the same as usual income.
With the exception of subsidies for goods and services which cannot be distinguished from refunds, the difference between the full retail value of a good or service provided by an employer and the amount paid by the household member is added to the income of employees.
Own business income
Own business income was collected from all persons aged 15 years and over who were working as owners or partners in unincorporated enterprises. Own business income is the share of profit/loss of the enterprise accrued to the person. Profit/loss consists of the value of the gross output of the enterprise after the deduction of operating expenses and an allowance for depreciation of assets used in producing the output. Losses occur when operating expenses and depreciation are greater than gross receipts and are treated as negative incomes.
The HES collects own business income in the last financial year because records of own business income are rarely available for more current periods. Sometimes, particularly during the early stages of interviewing, households cannot provide information on the last financial year and instead, provide information on the financial year prior to the last financial year. In cases where the preceding financial year’s profit/loss is collected, it is assumed that this is representative of current income and is not indexed or updated in any other way. During processing, the amounts are divided by the number of weeks over which the business was operational during the financial year to obtain equivalent average weekly income which, due to the length of the recall period, is considered to be the same as usual income.
Property income was collected from all persons aged 15 years and over who report net receipts accrued in the recall period as a result of ownership of assets. It comprises returns from financial assets (interest, dividends), from non-financial assets (rent) and from royalties. Amounts of property income are collected for the last financial year. The amounts are divided by 52 weeks to obtain equivalent average weekly income which, due to the length of the recall period, is considered to be the same as usual income.
Interest is collected from deposits (including term deposits) with banks, building societies, credit unions and other financial institutions.
Rent comprises receipts from properties other than owner-occupied dwellings. It includes receipts from lodgers and others who were sub-letting part of the dwelling, but excludes receipts from boarders who were counted as members of the household. Analogous with own business income, rent is net of operating expenses such as repairs and maintenance and interest payments. It is also net of depreciation. Losses occur when operating expenses and depreciation are greater than gross receipts and are included in income estimates as negative incomes.
Dividends comprise income households or persons receive from investments in corporate equities, such as ownership of shares. Income includes imputation credits.
Royalties include receipts in return for the use of patented and copyright materials.
Cash transfer income
Cash transfer income was collected from all persons aged 15 years and over who reported they were currently receiving regular and recurring receipts other than those obtained from employee, own business or property income. It consists of government pensions and allowances, other pension and life assurance annuity benefits and other current cash transfers.
Government pensions and allowances are receipts paid by government to persons under social security and related government programs. They include pensions paid to aged persons, benefits paid to veterans and their survivors and study allowances for students.
Other pension and life assurance annuity benefits include regular superannuation, life insurance and annuity receipts.
Other current cash transfers include private scholarship or study allowances, workers’ compensation not paid through the payroll and child support payments (non-government).
The HES collects current transfer information by asking recipients what their last payment is and the period it covers. Assuming that transfer payments are fairly uniform, the last actual receipt is considered a good proxy for usual income. The receipt is divided by the period it covers to produce an estimate of average weekly income.
Income of children aged less than 15 years was collected from the first parent or guardian interviewed. Only values of income which are readily accessible to the child or the parent or guardian are collected.
Timing of income
The total period covered by income estimates is a function of the recall period at the time of interview and the timing of interviews. Table 1 shows the length of the recall periods for different income items and, given that interviews were conducted over the 1998-99 financial year, shows the total period covered by the income estimates.1 RECALL PERIODS AND TOTAL PERIODS COVERED BY 1998-99 INCOME ITEMS
Studies which use HES data tend to assume that all income estimates refer only to the common reference period of July 1998 to June 1999. This is at least approximately true for employee and cash transfer income. For own business and property income, it is a valid assumption only if income levels are constant between the last financial year and the 1998-99 financial year. In cases where income levels are expected to have changed, researchers may wish to acknowledge or adjust for these differences.
The timing problem is likely to be greatest for households for which the major source of income is unincorporated business activity. Recorded income will relate to the previous financial year, while expenditure will mostly relate to a period within the current financial year. If business profitability is significantly different between the two years, then there may be a significant discrepancy between the recorded income and expenditure components which do not reflect the saving pattern of the household. While such differences will disappear to a certain extent through summing across households, there may still be an impact on aggregate estimates if, for example, all farmers had a bad season in one year and a good season in the following year. More importantly, there will be a definite impact on the quintile analysis of HES data.
HES income and expenditure estimates therefore do not balance for individual households or for groups of households and the difference between income and expenditure cannot be considered to be a measure of saving.
This page last updated 20 January 2006
Unless otherwise noted, content on this website is licensed under a Creative Commons Attribution 2.5 Australia Licence together with any terms, conditions and exclusions as set out in the website Copyright notice. For permission to do anything beyond the scope of this licence and copyright terms contact us.