Who Keeps Stats on the Statisticians?

Remember Australia’s blockbuster August jobs report? The Australian Bureau of Statistics (ABS) initially reported that the country’s economy added a whopping 121,000 jobs that month, the highest number on record, blowing past the consensus forecast of 15,000 jobs.

To put this result in a context more readily understood by US investors, analysts with the Commonwealth Bank of Australia said the August number would be equivalent to a rise of about 1.3 million in US payrolls, when taking into account the different sizes of the two labor markets.

Well, if you thought it was too good to be true, then you were right.

But the ABS went way beyond its normal approach to revisions of initial estimates. This time around, the agency’s statisticians decided to throw out the usual seasonal adjustment for July, August and September, after deciding that typical seasonal patterns had not been evident during this period.

Instead, the revised data show that the Australian economy added just over 32,000 jobs in August. That’s still a strong result when compared to the country’s typical pace of job creation, but it’s a far, far cry from a gain of 121,000 jobs.

Even worse, September’s labor force survey was a huge disappointment. The number of jobs fell by nearly 30,000 last month, a stark contrast to the consensus forecast that the economy would add 15,500 jobs.

The part-time category drove these results, with job losses totaling just over 51,000. However, full-time employment rose by almost 22,000 positions. Though there’s no getting around that this is a poor jobs report, at the very least the economy experienced gains in the higher-quality category.

Even so, over a longer-term period, full-time job creation has fallen well short of part-time job creation. Over the past year, full-time employment has grown by just 18,700 jobs, while part-time jobs are up by 103,200.

The unemployment rate now stands at 6.1 percent, which is the high for this cycle. And the labor force participation rate is at 64.5 percent, which is the low for this cycle.

Total hours worked, which can augur future employment demand, fell by 0.9 percent month over month.

In explaining the agency’s rationale for implementing such a seemingly drastic change, acting Australian Statistician Jonathan Palmer said, “It is critical that the ABS produces the best set of estimates that it can, so that discussion is on what the estimates mean, and not the estimates themselves.”

To that end, the ABS plans to commission a review with an independent third party to develop an objective framework for deciding when it’s appropriate to apply seasonal adjustments when calculating future data.

The agency said it did not make this decision lightly and believes the new approach will result in a more meaningful set of seasonally adjusted data.

Since key data, especially employment figures, can be highly politicized, some observers take a skeptical view of how they’re calculated. And statisticians are hardly infallible, as we’ve experienced when monitoring employment data in our sister publication, Canadian Edge.

We certainly don’t like it when additional subjectivity appears to have been added to what should be an objective process. But for now at least, we’ll have to assume that this was a good-faith effort to accurately reflect what’s happening in the Australian economy.

While we’re disappointed by these revisions, economists with Westpac note that leading economic indicators suggest that the pace of employment growth should accelerate through the year’s end.

Stock Talk

Add New Comments

You must be logged in to post to Stock Talk OR create an account