Monthly Household Spending Indicator methodology

This is not the latest release View the latest release
Reference period
June 2022
Released
9/08/2022

Introduction

The experimental Monthly Household Spending Indicator is derived using aggregated, de-identified banks transactions data from some of Australia’s banking and financial institutions.

The ABS transforms the banks transactions data in order to derive the Monthly Household Spending Indicator. As this data is not designed for statistical purposes, its scope varies from Australian National Accounts concept of household final consumption expenditure (HFCE) and the Retail Trade turnover estimates for retail businesses.

The indicator should be considered experimental at this stage, as further enhancement to the transformation processes and methodology are expected in the future.

Concept

Household spending is the goods and services paid for by Australian resident households within the Australian domestic territory or by Australian resident households abroad at the point of transaction. 

Household spending captures the point of expenditure; it is where payment is made for a good or service, not necessarily at the point where ownership changes, or when a service has been delivered, or where the goods and services are consumed. The expenditure of Australian resident households for business purposes is not included within the scope of household spending. Household spending is not a traditional economic statistic included in existing statistical frameworks, nor is it defined in the System of National Accounts (SNA) 2008.

Source of the data

The data is provided to the ABS by participating banks for statistical purposes. The data is provided to the ABS following the end of each calendar month. Once all data is received it is validated and collated.

The data is received in different formats and classifications from the participating banks. Some banks provide data at a low level of categorisation e.g., at a bank allocated merchant category level, and others provide categorised data at the Australia and New Zealand Standard Industry Classification (ANZSIC) or retail or non-retail level.

Data is received at different frequencies, with some banks providing daily data and others weekly. All participating banks provide data at an identifiable state and territory level.

The aggregated data does not contain information about individuals or households. Banks transactions data is only accessed by ABS staff required to produce relevant statistics, including the indicator. The ABS is committed to upholding the privacy, confidentiality and security of the information it collects.

Scope, assumptions, and coverage

The indicator is derived from aggregated, de-identified card and banks transactions by resident households. The data is provided to the ABS from participating banks. The source data does not cover the whole population and does not include all payment types. While cash payments, direct transfers outside the banks, cryptocurrency and BPay transactions are all forms of household spending, these are not currently included due to lack of available data sources or ability to identify such transactions in the current data set.

The following assumptions have been used in producing the indicator:

  1. All transactions represent final consumption expenditure. The data includes household transactions that may feed into the production process (e.g., a household purchase of seeds to grow vegetables at home). As it is not possible to identify or adjust for these transactions, it is assumed any ‘intermediate consumption’ transactions in the data have an immaterial impact on the final estimates.
  2. Non-resident transactions (net expenditure overseas) captured within the bank transaction data have an immaterial impact on the estimates. Lacking any means to identify and adjust non-resident transactions, it is assumed that expenditure patterns between resident and non-residents exhibit similar trends on consumption goods and services.

Estimation methods are used to adjust for under coverage of the population, under coverage within each spending category and over coverage of institutional sectors. For more information see Estimation.

Outputs are produced for nine out of the thirteen major Classification of Individual Consumption According to Purpose (COICOP) Divisions. Outputs are produced at the Australia, and state and territory level. For more information see Outputs.

Transformation of the data

The aggregated banks data is not designed for statistical purposes. In addition to normal micro and macro editing procedures, a number of data transformation steps and adjustments are made to derive the Monthly Household Spending Indicator. These include:

  • Classification and concordance mapping
  • Calendarisation data and derivation of trading day ratios
  • Alcoholic beverages and tobacco adjustment
  • Residual category adjustment
  • Estimation
  • Seasonal adjustment and Calendar adjusted estimates

Classification Concordance and Mapping

The bank data has varying classifications. To produce aggregate COICOP outputs, a concordance structure was developed to map the transactions data from each participating bank. This maps bank data, by merchant category or ANZSIC class, to the COICOP division classification and creates a consistent data set for further transformation. 

For example, bank data spending initially classified as “grocery stores and supermarket” expenditure would be allocated directly to the Food COICOP division. This mapping is unique to the household spending indicator, and while this process does increase comparability with HFCE estimates, HFCE is produced at a much more detailed product level which enables more accurate categorisation.

Calendarisation and derivation of trading day ratios

Following the end of each month, daily and weekly data is received for that month. The daily data is aggregated up for the whole calendar month.

The weekly values are divided by 7 and the derived value (1/7) allocated to each day of the relevant week. To capture trading day effects, a trading day ratio is calculated and applied to ensure the weekly data appropriately reflects household spending patterns across the days of the week.

Alcoholic beverages and tobacco adjustments

Identification of spending on alcoholic beverages and tobacco COICOP varies across the bank data. Where not identifiable in the data a select adjustment is made. This creates expenditure shares based on the identifiable alcoholic beverages and tobacco data, which is then applied to the data where alcoholic beverages and tobacco spending is not identifiable. 

Residual category adjustments

The level of detail for most categorised bank transaction data directly maps to COICOP. However, there are instances where the categorisation is less detailed and represents multiple COICOP categories. In these instances, a residual category adjustment is applied so all transactions are categorised to a specific COICOP.

The residual category adjustment borrows information from other bank data where more detail is provided in order to make required adjustments to allocate the more aggregated data.

Estimation

There are potential areas of both under and over coverage. For example, not all Australian resident households are represented within the supplied bank data resulting in under coverage of the population. Instances of over or under coverage can introduce bias and variance in the estimates. 

To account for coverage issues, a proportional modified Denton benchmarking method is applied. This is a standard ABS method used for temporal benchmarking, for example for quarterly GDP. In the absence of quarterly household spending, quarterly HFCE is used for the benchmarks. This balances timeliness and is conceptually more aligned with the target concept than other data sources or key aggregates available. Benchmarking to HFCE is applied at the state by COICOP division level. Total spending is then calculated as the aggregate of the state by division series.

Benchmarking to Quarterly HFCE:

  • adjusts for coverage error in the bank transactions data from differences in its relative coverage of the bank transactions data by product and state.
  • adjusts for changes in coverage of the bank transactions data over time, which are not representative of changes to total spending.
  • adjusts for coverage error arising from differences in quarterly growth rates between the sampled and unsampled populations.

When the benchmark becomes available for the extrapolated periods, the growth rates are adjusted to meet the benchmark series. This results in revisions. Revisions from the initial to the benchmarked growth rates can be used to provide an indication of the accuracy of the initial estimates of the spending indicator in terms of size (variance) and direction (bias).

Seasonal Adjustment and Calendar Adjusted estimates

The indicator exhibits systematic seasonal and calendar related effects. Producing a seasonally adjusted series would improve the quality of the output in these ways:

  • Interpretability – predictable seasonal effects will be removed to better identify short-term and long-term behaviour
  • Relevance – increased suite of statistics available for users
  • Coherence – comparability with existing outputs including retail trade and HFCE

A minimum of 3 years of data is usually required to seasonally adjust data. While we have 3 years of data (January 2019 to December 2021), this 3-year period is far from usual because of the pandemic and so seasonally adjusted estimates are not currently produced.

In the absence of seasonal adjustment, adjustments have been applied to allow for length of month and trading day effects present in the month-to-month movements for many series. These are referred to as calendar adjusted estimates.

The calendar adjusted series uses trading day adjusted bank transactions data, which differs from the input data used in the original series and includes a length of month adjustment prior to benchmarking.

The application of these methods to produce a calendar adjusted series generates a series adjusted for some of the seasonal influences present in the data.

Outputs

Monthly indices are produced for total household spending and nine of the major COICOP divisions:

  • Food
  • Alcoholic beverages and tobacco
  • Clothing and footwear
  • Furnishing and household equipment
  • Health
  • Transport
  • Recreation and culture
  • Hotels, cafes and restaurants
  • Miscellaneous goods and services

As levels of spending are not available from the indicator, index values are produced to enable users to calculate movements between any two time periods. The index values are set to 100.0 at the beginning of the time series, January 2019. The index values are available in both original and calendar adjusted terms for Australia, and each state and territory.

Privacy and confidentiality

Legislative requirements to ensure privacy and secrecy of this data have been adhered to. In accordance with the Census and Statistics Act 1905, results have been confidentialised to ensure that they are not likely to enable identification or a particular person or organisation.

All personal information is handled in accordance with the Australia Privacy Principles contained in the Privacy Act 1988. For more information, see ABS privacy.

Data limitations and revisions

The data provided by the participating banks is not designed for statistical purposes. While a rich data source, there are still limitations in using this data for the purposes of a Monthly Household Spending Indicator.

Coverage

The data provided by the participating banks does not include all domestic resident households nor does it capture all household spending across the COICOP divisions. Card and bank transactions are an appropriate and preferred mode of payment for most of the published COICOP divisions, however, it does not represent all spending due to the absence of cash-based spending and other payment modes.

Household spending as a measure of economic activity

Due to the varying payment modes, and high concentration of non-card or bank transaction payments for some COICOP divisions, a household spending indicator could not be produced for all major COICOP divisions.

The appropriateness of using aggregated banks data as a measure of household spending across each COICOP division was assessed with selected Divisions excluded from the Monthly Household Spending Indicator.

The COICOP divisions that are currently excluded from the indicator are:

  • Rent and other dwelling services
  • Electricity, gas and other fuels
  • Communication Services
  • Education Services

Revisions

Revisions are a change in the value of a published value and may arise due to a variety of reasons.

Revisions may be applied to the Monthly Household Spending Indicator due to:

  • Revisions to the source data
  • Refinements to decisions made around the treatment of data anomalies in the series
  • Implementation of methodological and process improvements
  • Estimation following the application of the quarterly benchmark. 

Methodological enhancements

The methods and data sources used will be subject to ongoing review to improve outputs and maintain the relevance of this indicator.

Analysis of Household Spending Changes

Index numbers

Movements in indexes from one period to another can be expressed either as changes in index points or as percentage changes. These indexes measure change over time; they do not measure differences between states/territories. The following example illustrates the method of calculating changes in index points and percentage changes between any two periods: 

Table 1. Monthly Household Spending, National, current price, original, index, total = Index numbers: 

June 2021 = 104.7
Less June 2020 = 93.2
Change in index points = 11.5
Percentage change = 11.5/93.2 x 100 = 12.3%

Percentage changes calculated can be used to illustrate two different kinds of movements in index numbers:

  • movements between corresponding months of consecutive years
  • movements between consecutive months.

Interpreting monthly changes

Indicator estimates are produced in current price original and current price calendar adjusted terms. Calendar adjusted estimates account for trading day impacts and length of month. As the time series lengthen, seasonally adjusted estimates will become available. Until such time, through the year movements will be referenced in the analysis of the household spending indicator.

Rounding

Published index numbers are rounded to one decimal place. Published percentage changes are calculated on the underlying data and rounded to decimal place. Any discrepancies between percentage changes published and percentage changes derived from published index numbers are due to rounding.

Differences to other ABS estimates

Household spending draws its basis from HFCE. Household spending captures the point of expenditure, not necessarily where ownership changes or when a service has been delivered, or where goods are consumed. HFCE has greater coverage and detail compared to household spending. Conceptually HFCE is recorded at the point where a household incurs a liability to a seller. That is, at the time ownership changes for a good or where a service is delivered, and this is not always at the point where payment is made. The table below shows some of the key differences between the Household Spending Indicator, HFCE and Retail Trade.

 HFCERetail TradeHousehold Spending indicator
Data source(s)• Retail Trade
• Administrative data
• Government Finance Statistics
• Business Indicators
• Consumer Price Index
• Scanner data
• Building Activity
• Survey of Income and Housing
• Australian Petroleum Statistics
• Household Expenditure Survey
• Other ABS surveys
Retail Business SurveyAggregated, de-identified banks transactions data
FrequencyQuarterly, AnnualMonthly, QuarterlyMonthly
Classification structure/ Lowest level of compilationCOICOP/Sub class Australian and New Zealand Standard Industrial Classification (ANZSIC)/ClassCOICOP/Division
Classification CoverageAll COICOP• Retail Industry
• Selected subdivisions from Accommodation and Food Services Industry
Selected COICOP
Conceptual coverageHFCERetail Trade turnover (Survey)/30% of HFCE(a)Household purchases/68% of HFCE(b)

a) This coverage, or contribution, varies each period. Retail Trade makes up approximately 30% of HFCE, but in some quarters is as high as 35%.

b) Household spending coverage is based on the published outputs planned for the indicator in terms of COICOP Divisions. These Divisions have been assessed to have reasonable coverage, alignment and coherence.

Back to top of the page