The Real Estate Institute of Victoria recently released its summary of the 2012 Melbourne auction market and the headline is that things are getting ”better”.

After posting a lacklustre auction clearance rate of 59 per cent last year, renewed buyer interest has seen the sales level rise to 61 per cent in 2012.

Except this year’s clearance rate is actually 59 per cent.

The discrepancy comes courtesy of an extensive revision to REIV figures that have seen nearly a thousand ”sold” properties simply added to the end-of-year total. And these were results that were previously unreported or originally marked ”passed in” only to be reclassified as sold.

Stick with me because it’s only going to get more tangled from here.

This is how the REIV auction reporting system works:

Agents voluntarily report their auction results to the REIV on a Saturday or Sunday, with those numbers used to calculate the clearance rates that are published online and in the media. On any given weekend, about 9 to 15 per cent of auction results are not reported by these deadlines.

The industry group continues to collect those unreported results over the next few days, eventually publishing an ”updated” clearance rate the following Wednesday.

It’s fair to say the REIV system is pretty liberal in its reporting obligations. Depending on whether an agent has a Saturday or Sunday auction, they can have up to four days to report the result.

This is necessary, the REIV says, because some agents will close deals after the 5pm reporting deadline* on a Saturday and often do not work on a Sunday, hence the unreported and then late reported results.

Which brings us back to the whole issue of how the clearance rate can be both 61 per cent and 59 per cent simultaneously.

This chart compares the two data sets – the annual figures according to the reports issued on a Wednesday and the final end-of-year figures – both of which are sourced from the REIV:

Clearly, something doesn’t add up.

Despite the multiple and lengthy reporting deadlines, the REIV has continued to add ”unreported” results to the tally – to a grand total of 851 results over the course of this year.

But what is truly amazing is that an additional 962 results have now been classified as ”sold” between the two data sets.

In effect, it means that every one of those 851 missing results, in fact, turned out to be sold properties. Plus, another 111 results were revised to ”sold” that had been previously been recorded as ”passed in”.

The REIV notes that its ‘‘sales database is a live one and is subject to ongoing revision’’ and frequently cites the need for revision as necessary even for the Australian Bureau of Statistics.

Fair enough, up to a point.

But the REIV and its member agents don’t adhere to the same reporting and quality control standards as the ABS.

The REIV frequently refers to its weekly auction reports as a “snapshot in time” of the state of the market. But they can hardly be snapshots if data is continually added days, even weeks, after the event has occurred.

A property is put under the hammer on a specific date, at a specific time. Even with the REIV’s already generous reporting deadlines, it should be pretty clear what the outcome has been four days after the event.

It has either sold or it has not.

But rather than mark an “unreported” result as “passed in” even four days later, the REIV continues to allow agents to late report and, apparently, change previously reported outcomes.

The REIV also says that “these minor revisions do not change the state of the market”.

A move of two percentage points is no great shift, for sure.

But considering how much of the end-of-year commentary in 2012 has been focused on that two point gain, it’s clear that these “minor revisions” can certainly change the perception of the state of the market.

Source:  Chris Vedelago, The Sunday Age’s property reporter.