6.1 The ABS is in an environment of transformation where new opportunities to access and interrogate ‘big data’ are becoming available to compile official statistics. One area the ABS has made significant progress is the use of transactions data to compile the CPI. Prices obtained from transactions datasets currently accounts for approximately 25 per cent of the CPI weight. The current methodology used for the implementation of transactions data replaced traditional (point-in-time) field collected prices with unit values in 2014.
6.2 While current methods used to compile the CPI represent a significant enhancement on traditional practices, opportunities exist to further enhance the methods used in the Australian CPI. To overcome the issues with traditional bilateral index formulae when using transactions data, academics and NSOs have reached consensus that multilateral methods are required to exploit the full amount of information provided in transactions datasets.
6.3 As part of the research program to Enhance the CPI, the ABS has conducted research on a selection of well-known matched-model multilateral methods for producing temporal indexes. Following further research and consultation with international experts, the ABS is now in a position to make a final decision on methods to be implemented into the Australian CPI. Section 5 of this publication details the specific methods the ABS plans to implement in December quarter 2017.
6.4 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 small sample of products) and weighting products by their economic importance (rather than using unweighted price indexes). In the medium term, the use of multilateral methods will require fewer resources to process and analyse price movements.
6.5 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. At the sub-group level, empirical evidence provided in Section 3 showed similar price trends overtime with some short-term departures.
6.6 User and stakeholder input is welcome on the details of this publication. For further information relating to the implementation of new methods for transactions data, users should write to:
Mr Andrew Tomadini
Consumer Price Index Section
Australian Bureau of Statistics
PO Box 10
Belconnen ACT 2617