The key take-away from this dialogue was that all the folks claiming that the Q1 and Q2 core CPI outcomes should be ignored because of measurement bias don't know what they are on about. The RBA certainly does not buy into it. Stevens/Lowe/Battellino were crystal-clear in stating that they were surprised by the broad-based strength of core inflation in the first half of the year, and were repeatedly dismissive of attempts by politicians to explain it away via anomalous events/factors...
Dr Leigh: On the measurement of inflation, there has been an argument recently that the ABS's fairly infrequent rebasing of the CPI has led to an upward bias in the CPI, maybe in the order of half a percentage point. Are you concerned about that? Are you tending to use, say, the household implicit deflator rather than the CPI because of that?
Mr Stevens: We are not doing the latter. Perhaps I will get Phil to talk to the technical aspects of the frequency of updating.
Dr Lowe: At the moment, the ABS does a survey once every six years to work out exactly what we are spending, what goods and services we are buying. So the weights get changed once every six years. I think the analysis that various people have done suggests that the fact that they only do that weighting once every six years does lead to a slight upward bias in the CPI, particularly towards the end of that six-year period, because what happens over time is that we all substitute towards the goods that are slightly cheaper and away from the goods that are fairly expensive, so the true costs that we are actually paying for goods and services are not rising quite as quickly as the CPI. But the effect is fairly small. In most of the estimates I have seen, it is around 0.1 of one per cent...None of that really undermines the credibility of the CPI, though. This effect is there, but it is fairly small in the overall scheme of things.
