This information paper presents methodological changes for the use of transactions data to compile the Australian Consumer Price Index (CPI). The methods recommended in this paper are planned to be implemented into the CPI from December quarter 2017.
The implementation of these new methods would represent a significant enhancement to the Australian CPI. This includes using all the products available in the datasets (rather than a sample of products) and weighting products by their economic importance (rather than using unweighted price indexes). The use of multilateral methods will require fewer resources in the medium term.
Extensive consultation with international experts and key stakeholders has been undertaken to secure support for these enhancements; and to manage statistical risk. Implementation of these enhancements will occur in the December quarter 2017. Consultation included external expert reviews by Professor Kevin Fox and Professor Jan de Haan; and a stakeholder engagement program including a call for public submissions. The ABS also conducted numerous bilateral and multilateral consultations with key stakeholders, including: the Reserve Bank of Australia; the Treasury; Department of Social Services; Department of Finance; and State Treasuries.
Transactions (scanner) data refers to point-of-sale purchases from retailers and is currently used to compile approximately 25 per cent of the CPI weight. The ABS currently uses a 'direct replacement' of observed point-in-time prices with a unit value calculated from the transactions data. The international price statistics community has reached a consensus that new methods, called multilateral methods, are the most effective way to exploit the full amount of information provided in transactions datasets. The forthcoming update of the International Labour Organization (ILO) Consumer Price Index Manual will recommend the use of multilateral methods for temporal aggregation when using transactions data.
As part of the ABS research program into Enhancing the CPI, a number of multilateral methods have been considered for implementation into the CPI. This research assessed a selection of multilateral index methods against a framework comparable to the ABS Data Quality Framework (DQF); and provided empirical results. In light of additional research and external consultation and review(footnote 1) , including with international experts, the ABS is now in a position to make an informed decision regarding the implementation of multilateral methods. This information paper provides the plans for implementation and the methodological details of these changes.
The methods discussed in this paper will be used to compile 28 Expenditure Classes (ECs) in the Australian CPI. The historical analysis presented in this paper indicates this methodological change would have had a minor downward impact on the overall CPI. This may not translate into a similar result in future periods.
In summary, the specific methods to be implemented by the ABS to compile the CPI from transactions data are:
- GEKS-Törnqvist as preferred multilateral method compiled using respondent classes as EAs.
- Weighted Time Product Dummy (TPD) method to be run concurrently (for comparison purposes).
- Aggregate EAs together using Törnqvist index formula to form respondent indexes.
- Aggregate respondent indexes together using Lowe Index formula.
- Mean splice extension method with a rolling window of 9 quarters
The following is a timetable for implementation:
- ABS to continue to monitor these methods during June quarter 2017 and September quarter 2017; and make refinements as required.
- ABS to implement new methods in December quarter 2017. This is to align the implementation of multilateral methods with the introduction of updated household expenditure data for weighting purposes(footnote 2) .
1 See appendices 1 and 2. <back
2 See Information Paper: An Implementation Plan to Annually Re-weight the Australian CPI (cat. no. 6401.0.60.005) <back