4.5% Jobless Isn't Australia's Demand Signal

ANZ-Indeed Australian job advertisements rose 2.5 percent in August after a revised 1.9 percent in July. Seasonally adjusted, that is four consecutive monthly gains: May 2.3, June 0.3, July 1.9, August 2.5. The series sits 7.8 percent higher than a year earlier and 21.4 percent above its 2010s average. The Australian Bureau of Statistics will not publish August labour force until 24 September. Desks are already trading July.

The dominant read of July is cooling. Seasonally adjusted unemployment ticked to 4.5 percent. Employment fell 15,800 to 14,807,200. Hours worked dropped 12 million, or 0.6 percent. Participation slipped to 66.9 percent. After the 20 August release, several economists said another cash-rate hike had become unlikely. The Reserve Bank’s August Statement on Monetary Policy had already lifted the year-end unemployment forecast to 4.5 percent. That is the wire.

Ads can rise while the jobless rate does

If 4.5 percent were a sufficient signal that employers were pulling demand, advertised vacancies should not have climbed for a fourth month. They did. Indeed’s Callam Pickering located the August lift in Victoria, Western Australia, and Queensland, with New South Wales slightly down. Christmas hiring showed up in retail and food service. Education and cleaning gained. Software, data, and other tech roles kept easing. That is not a uniform national cooling. It is a seasonal and geographic mix that a single unemployment rate cannot see.

July’s own composition already argued against reading the headline as a demand collapse. Full-time employment rose 16,300. Part-time employment fell 32,200. Hours fell faster than headcount. The unemployment rate can drift higher when participation holds up better than jobs, as Westpac noted after the print, without firms cancelling vacancies. Eurostat’s flat jobless rate hid the same composition problem. Australia’s July mix is the local version: fewer part-time posts, more full-time, fewer hours.

The ABS itself warned of larger-than-usual uncertainty around the July estimates while it modernises the Labour Force Survey. Industry and occupation tables are suspended through the August reference period. The 24 September release can test employment, hours, and participation. It cannot yet show whether August hiring landed in retail, mining states, or tech. Payroll withholding would be the cleaner contemporaneous check. Until then, ANZ-Indeed is the demand series that actually moved in August.

Jasmine Zheng at ANZ still called July labour-market data soft and still expects ads to trend lower as higher rates bite with a lag. That forecast can be right later and still leave August’s 2.5 percent as a fact now. Seasonal retail ads typically peak in early October. If September ads keep rising for the same Christmas reason, the “cooling demand” story will have been a calendar, not a cycle.

A 4.5 percent unemployment rate is a slack ratio. It is not a vacancy count.

What rates still mis-price is the July print as a veto on 29 September. After the jobs release, hike odds were talked down. By 17 September, LSEG data put an 87 percent probability on a lift from 4.35 percent. Governor Michele Bullock’s August minutes still called conditions “a little tight” and said leading indicators pointed to only limited near-term easing. The Fed has already hiked into a different labour map. U.S. retail sales printed a demand signal the same week. Australia’s imported fuel bill is a separate inflation channel; Hormuz already repriced ANZ diesel buffers. None of that is resolved by July’s 4.5.

The test is dated. If the 24 September ABS print shows employment, hours, and unemployment all easing together, the job-ad rise was a seasonal lead that did not convert — and the cooling read survives. If hours and full-time employment hold or rise while ads stay bid, 4.5 percent was a composition and sampling print, not a demand signal. A material downward revision to ANZ-Indeed that concentrates the 2.5 percent in duplicate or Christmas-only postings would kill the contradiction from the other side. Until one of those arrives, markets are using last month’s slack ratio to price next week’s cash rate, and skipping the vacancy series that kept climbing.