Australian retail sales plunged in August, recording a decline of 0.6%, the largest monthly decline since March 2013.
Following a 0.2% decline in July, it saw annual growth in sales slow to 2.1%, the weakest outcome in over four years.
Following warnings on the outlook for household spending from the RBA and ABS earlier this week, it’s created renewed concern on the health of household budgets.
Has years of high and increasing levels of household indebtedness, coupled with weak household incomes growth, finally come home to roost? That’s the question many are asking after today’s data.
While undeniably weak, Tapas Strickland, economist at the National Australia Bank (NAB), says that the decline in headline retail sales may have been accentuated by sampling effects in the ABS survey.
“There is some evidence that sampling effects did weigh on today’s numbers,” he said in a note released following the retail sales report.
“The non-fully enumerated part of the sample declined a sharp 1.1% in the month, while the fully enumerated sample declined by less, down 0.3% month-on-month.”
Strickland says the fully enumerated figure measures sales at larger firms that are included in the survey every month, usually totaling around 500.
The non-fully enumerated figure looks at sales at other retailers which, in most instances, are smaller than those in the enumerated sample.
“Consequently, trends in the fully enumerated tend to be less volatile,” says Strickland, adding that the non-fully enumerated sample tends to be very volatile month to month.
The difference between the two sample groups is captured perfectly in the chart below from the NAB.
According to the NAB’s estimates, the sampling change in August likely detracted 0.3 percentage points from the headline sales figure, meaning that without it sales would have fallen by around 0.3%.
Still ugly, but not as ugly.
And, if previous trends are maintained, Strickland says that some payback to August’s weak number may arrive when the September report is released.
“We note previous sharp falls in the non-fully enumerated sample have tended to reverse in the following month, so a bounce back next month is plausible,” he says.
However, even with those nuances in the data, he says that it’s still hard to explain the broad-based weakness seen in August.
“Given the broad-based declines, it is unclear what caused the weakness in August,” he says. “There was a ramp up in geopolitical tensions, house price growth has slowed, while utility prices have risen sharply, and there is more talk of rising interest rates.”
More broadly, Strickland says retail sales are likely to remain subdued until wages begin to strengthen given debt levels and a hesitant consumer.
Strickland was one of the few analysts to predict a decline in retail sales in August, forecasting a drop of 0.3%.
That call was centred around the the release of the NAB’s cashless retail sales index — a measure of retail non-cash spending by NAB’s customer base — which foreshadowed the ugly decline in the ABS retail sales report in August.
Given its recent track record, there’s sure to be a few eyes on the NAB’s September report when it is released in late October.