BCEC Monthly Labour Market Update – September 2025

• Anaemic jobs growth in September sees unemployment rate rise to 4.5 per cent
• Participation remains high despite weakening labour market conditions
• Upcoming September quarter CPI figures key to RBA interest rate decision on Melbourne Cup Tuesday

Don’t panic ( just yet)

There’s now no disputing that the national labour market is softening. The question is how sharp the downturn will be and how long it will last. September saw just 14,900 net additional jobs created across the country. By comparison, around 25,000 additional jobs would have been required to keep the unemployment rate steady, had there been no change in participation.

That’s not a particularly large gap and could easily be reversed fully next month. In fact, based on the ABS published sampling errors, there’s a 38 per cent chance the actual change in employment was greater than 25,000 jobs.

The 95 per cent confidence interval for the estimate of September’s change in employment ranges from a fall in employment of 50,700 jobs to a rise of 80,500 jobs.
However, four out of the last five months have produced negative or very weak employment growth. Plus, the participation rate did not stay the same.

We’ve previously suggested record high rates of participation meant participation would likely fall in the event of a downturn in labour demand, which has the effect of limiting the rise in the unemployment rate. Australians didn’t follow that script in September.

Despite the weakening labour market, the participation rate increased 0.1ppt to 67 per cent, back near the record high 67.2 per cent set in January of this year.

Combined with weak jobs growth, that increase in participation saw the unemployment rate rise from 4.3 per cent to 4.5 per cent, the highest in nearly 4 years. Those with good memories may recall the unemployment rate was 4.2 per cent when the August figures were released but, as often happens, that was revised up to 4.3 per cent.

It’s being suggested the unemployment rate hitting 4.5 per cent will force the RBA’s hand to announce a further rate cut when the Monetary Policy Board meet on 2-4 November. Frankly, we’re not buying into that. The September quarter CPI figure, to be released on 29 October, is almost all that will matter in shaping the RBA’s announcement on Melbourne Cup day.

The 2.1 per cent annual increase in the CPI recorded to the June quarter suggested inflation had been reined back to well within the 2-3 per cent target range. The two monthly figures since then – 2.8 per cent in July and 3.0 per cent in August – cast doubt on that assessment.

The RBA will not be cutting rates if the September quarter CPI has jumped back toward the 3 per cent upper bound. It’s true that one of the RBA’s legislated objectives is to maintain full employment. However, it is also clear that they view low and stable inflation as a necessary precursor to sustaining that.

Sum of the parts greater than the whole – really

Another note of caution regarding the poor jobs numbers for September is the variation in outcomes across the states. QLD, SA, and WA actually recorded very strong jobs growth for the month: 0.5 per cent, 0.5 per cent, and 0.4 per cent, respectively.

They represent the highest rates of growth in a single month in those states in four months, six months, and four months, respectively. For our two most populous states, the 0.1 per cent decline in VIC equated to 3,200 fewer workers, more than fully offset by NSW’s +0.1 per cent growth (6,000 workers).

WA had the odd result of 0.4 per cent growth in employment alongside a 0.4ppt increase in its unemployment rate! The ACT recorded an even larger 1.1 per cent jump in employment (3,000 workers), although data for the territories tend to be volatile due to smaller sample sizes.

If you’re thinking that all sounds a bit inconsistent with the bleak outcome at the national level, you’d be right. When we add the individual state and territory estimates, the total shows an increase in employment across Australia of 30,800 jobs (0.2%) in September, more than double the published estimate of 14,900 jobs (0.1%).

We can see the source of the discrepancy lies in the seasonal adjustments, since the state and territory estimates do correctly sum to the national estimates in the original (non-seasonally adjusted) series. All the same, taking a state-by-state perspective, the labour market actually looked pretty healthy in September.