The labour market is starting to catch up with the rest of the economy, as yesterday’s jobs data from the Australian Bureau of Statistics shows.

The June employment numbers were still good — there’s decent jobs growth and participation is continuing at very strong levels. But it’s clear that strong growth of the kind that got unemployment briefly down to 4.9% last year has vanished and we may be set for period of rising unemployment — vindicating the Reserve Bank’s decision to cut interest rates to new lows.

For several years now, the ABS has been urging us to keep an eye on the trend numbers for jobs, not the more volatile seasonally adjusted numbers. However, the two are now starting to line up. In seasonally adjusted terms, just 500 new jobs were created (against market forecasts of around 10,000) compared to trend figures of around 26,000 new jobs. But the ABS revised the trend jobless rate for May up to 5.2%, catching up with the seasonally adjusted jobless rate, which reached that level in April. The two are now aligned.

Indeed, June’s trend jobless rate of 5.2% was actually 5.24% and was rounded down; had it been 5.25%, it would have been rounded up and the headlines would have been very much different. Seasonally adjusted monthly hours worked in all jobs also fell by 100,000 hours in June; they rose on a trend basis, but at a slower rate than May.

The annual level of jobs growth is still above historical levels: 2.6% trend and 2.4% seasonally adjusted. But the rate of growth is softening, albeit off a high base, reflecting that the labour market is turning. AMP’s chief economist Shane Oliver reckons the jobless rate is headed back to around 5.5% by the end of 2019. He said yesterday:

We see a further slowdown in jobs growth over the next six months as the housing construction downturn flows through the economy. This is likely to see trend jobs growth fall well below the roughly 19,000 new jobs needed each month to absorb new entrants to the labour force … As a result unemployment is likely to drift up to 5.5% towards the end of the year.

That raises the question of where, exactly, jobs growth is going to come from, not merely to stop a rise in unemployment to 5.5% — which will hammer wages growth and push households into even lower spending levels — but to get it back to 5% and then push it down to the kind of levels the Reserve Bank believes will start to exert some pressure on inflation.

Construction employment has been falling over the last year and is set to contract further, as Oliver notes. Manufacturing is in deep trouble, having shed over 100,000 jobs over the last year. It’s only in services — health, education, professional services — that employment growth has been strong.

The government will be hoping the tax offsets and the RBA’s interest rate cuts will be enough stimulus. But the tax offset payments, if they’re not banked by worried households, will only help the retail sector (and foreign manufacturers of household goods). The only sector really going gangbusters — iron ore exports — hardly employs anyone. Construction, given the current state of the housing market in the major cities, will need more than the interest rate cuts to put a floor under falling employment.

That suggests infrastructure spending by all levels of government may still be needed to prevent unemployment from rising. Over to Frydenberg and Morrison, and their state counterparts, on that front.