The RBA minutes only refer to one set of house price index data, explicitly highlighting the "up-tick" in prices recorded by RP Data-Rismark's daily index prior to the RBA's last board meeting on 3 July:
"In contrast, indicators suggested that the housing market remained subdued. Dwelling activity was likely to have fallen further in recent months and indicators generally suggested that activity would remain relatively weak in the near term. Notwithstanding an apparent tick-up in June, dwelling prices were around 6 per cent lower than in early 2011, with the largest falls in Melbourne and Brisbane. Household credit continued to grow at around the pace of the past year (broadly in line with incomes). Interest rates on housing loans were around 50 basis points below their 15-year average."
It is a subtlety, but what is most profound in the paragraph above is that Australia's central bank is describing movements in house prices over the month of June only 3 days after the month has ended! This is a transformational world away from the the house price technology that existed several years ago whereby quarterly, as opposed to daily, data was reported with a multi-month lag. Even today APM and the ABS report their quarterly data with a one month lag.
To be clear, while the RP Data-Rismark index is the only price data highlighted in the RBA's minutes, the central bank prudently reviews information from both APM and RP Data-Rismark, and still includes the ABS index data in the Statement on Monetary Policy.
Rismark and RP Data have a great deal of respect for Dr Andrew Wilson and the Fairfax-owned APM, which report a useful "stratified median price index" based on the pioneering Prasad and Richards methodology. Rismark and RP Data also compute this stratified median index internally, in addition to our old adjacent-period monthly hedonic, and simple medians.
It should also be said that it was in part the RBA that inspired Rismark and RP Data to invest many millions of dollars into the research and development of complex regression-based monthly and daily indices that seek to eliminate the compositional biases that arise in housing transactions due to the twin problems attributable to the heterogeneity of the housing asset itself (no two homes are identical), and the "non-synchronicity", or infrequent trading, associated with the overall stock (note that in absolute terms transaction volumes are very high at north of 400k per annum).
