While there’ll be a lot of emphasis on GDP growth and “technical recessions” in coming months, the real, meaningful indicator of recession is unemployment.

For anyone who lived through the recession of the early 1990s, the memory of 11% unemployment, one million people out of work — at least briefly in early 1993 — and the sight of hundreds of people queueing for what few jobs were being advertised, is unforgettable.

That’s why the Rudd-Swan government’s efforts during the global financial crisis were so remarkable — unemployment peaked at 6% and 670,000, below the level of unemployment later reached under the Abbott government.

Labor managed to get on top of unemployment by the end of 2009, leading to a gentle peak and then a slow slide through the 2010 mining investment boom, driven by China’s stimulus package.

In the 1990s, it was instead a long march to a very high peak and widespread despair — a fate we avoided courtesy of a government prepared to do what it took to prevent recession.

If Scott Morrison and Josh Frydenberg can match Rudd and Swan’s achievement they’ll be doing very well. The Rudd government entered the financial crisis with unemployment at just 4.2%. Currently it is 5.3%.

The current participation rate is also a full half a percentage point higher than it was a decade ago. That remains a signal achievement of the Turnbull government, which oversaw a big rise in — mainly female — participation through 2017 and early 2018, but it will make lowering the unemployment rate harder.

We will very likely see a significant rise in underemployment from its current 8.5% rate. But the number to focus on is the overall unemployment rate — better to have people in jobs and wanting more hours than in no job at all.

The jobs market has also changed significantly since 2008. Back then, retail was the biggest employer, and health and social care had only just overtaken manufacturing as the second biggest employer.

Now health and social care is by far the biggest employer, and manufacturing has shrunk significantly. This will help cushion the employment impact of the crisis: health and social care will not merely not contract as a result of the virus shock, but expand, even as sectors like retail and manufacturing, which are in deep trouble, shed workers.

Indeed, this will in fact be a continuation of a trend: for several years now, health and social care, and to a lesser extent education, have driven a big rise in jobs growth — and participation — even as other sectors have stagnated, helping to keep unemployment to around 5%.

But another key shift will work against employment. While union claims about the spread of casualisation over the last 20 years have been exaggerated, the expansion of the gig economy has made employment more precarious. Around 7% of workers now use digital platforms to find work.

How that number has changed over the last decade is less clear — including the extent to which that simply means a switch from workers (such as students) who would have worked casual, and equally precarious, jobs a decade ago to Deliveroo riders now.

But many such workers may remain “employed” but facing a significant loss of income if they’re not in sectors of the economy, like food delivery, that will benefit from the crisis.

The gig economy is one aspect of a broader, continuing switch in the jobs market from production to services that is a decades-long phenomenon that has accelerated over the last decade.

Not just heath care but the hospitality and tourism sectors that have benefited from the switch in consumer spending away from products to “experiences”.

While some of these “experiences” can be delivered online, or by bike courier, most are, almost definitionally, face-to-face, and are therefore extremely vulnerable to a virus-driven clampdown.

But if the government is unable to keep unemployment down below the low 6% mark, we’ll be in territory we haven’t visited since the late 1990s, when we were still recovering from the recession we had to have.