Australia's #1 industrial directory for equipment & suppliers

Statistically speaking: Construction figures are conflicting

By: Peter Gosnell
21 October, 2010

Feature of the week: It's been said that statistics will confess to anything, if you torture them enough, and there are few industry sectors where conflicting admissions are extracted as regularly as in construction.

This month the Australian Bureau of Statistics (ABS) released its Building Activity Australia report for the three months to June 30, 2010.

According to the ABS report, the value of all construction rose 5.7% seasonally adjusted in the three months to June 30 after a 7.3% rise in the March quarter.

In dollar terms, the rise saw the value of all construction activity in the June quarter lift to $21.571 billion.

So on seasonally adjusted terms across all sectors of construction activity – new residential, renovation, non-residential, etc. – the figures convey an expectation of unambiguous growth. But then there are the trend estimates.

"The trend estimates should be interpreted with caution," the ABS said in the data notes attached to the June quarter report.

"… the underlying behaviour of building activity may be affected by initiatives within the Government stimulus package, including the 'Building the Education Revolution' (BER) program and Social Housing Initiatives, as well as other developments associated with global economic conditions.

"Though the impact of the BER program was excluded from the trend estimates in the June 2009 through to the March 2010 releases of this data, this will no longer be the case from the June quarter 2010."

The decision to include the BER programs was taken by the methodologists in the ABS' time series analysis unit.

"We were expecting a building peak to last two or three quarters, and we just didn't see that," an ABS spokesperson told

"Because that hasn't happened, there's no point in leaving the numbers out, because people are assuming it's a massive change, and it's not.

"At the moment, BER is contributing about half the value of all education-related construction work, but that doesn't mean that education work wouldn't have proceeded without the stimulus package," the spokesperson said.

Not exactly contradictory. Nor incidentally does it necessarily correlate with claims by the Federal Government about the effectiveness of the BER. But in terms of the tales told by statistics, the potential emerges for varying interpretations of the numbers on construction activity, which nevertheless support a general picture of robustness – whether the stimulus programs are included or not.

Contrast this with the dim view being advanced by the Housing Industry Association (HIA).
The same day the ABS released its June Quarter report, the HIA issued a press release titled: 'Land Sales Ease, Confirming Weaker Housing Outlook'. Here were different statistics, equally compelling, conveying a less sanguine outlook.

"A sustained weakening in residential land sales volumes over the nine months to June 2010, and a raft of other leading indicators, point to a renewed downturn in new home starts next year," HIA chief economist Dr Harley Dale said.

His colleague, HIA economist Matthew King said he expected the slump in land sales volumes to translate into weakening residential construction activity by the middle of 2011.

"There is a significant lag between the sale of land all the way through to dwelling completion," King said.

While reluctant to quantify the duration of this lag, King said it was something of the order of nine to twelve months.

"Given that lag, a slowdown in June quarter land sales really won't show up in a slow down in housing starts until at the very minimum the beginning of 2011," he said.

"Our report certainly suggests a slow down in residential building activity in the June quarter next year."

Statistics like these give the construction industry the ammunition to lobby for a fresh injection of stimulus before Winter returns in 2011. If of course, the statistics can be believed.

Have your say...

We welcome thoughtful comments from readers
Reload characters
Type the characters you see in this box. This helps us prevent automated programs from sending spam.