The release this week of the April unemployment numbers reinforced just how quickly the economy has slowed, and also that despite the horror of the numbers, things still have a way to go before we reach the bottom.

This month, despite getting the big attention due to rising from 5.2% to 6.2% in April (a record one month rise), the unemployment rate was largely meaningless.

This isn’t because the unemployment rate is dodgy, or because of difficulties with Centrelink (you do not need to be on unemployment benefits to be classed as unemployed).

No, it was largely meaningless because the rate involves a number of moving parts all of which ensured it would be smaller than the “real” number.

For a start the unemployment rate is a measure of unemployed divided by the number of people who are in the labour force. In April 594,000 lost their job, but unemployment rose by only 104,000.

The rest just left the labour force – ie they were neither working nor unemployed because they were not actively looking for work.

The reasons are varied, but mostly it is due to a relaxing of some mutual obligation requirements, timing and the schools.

People receiving jobseeker may not have actually been looking for work – ie they are on unemployment benefits but not technically unemployed, and many have stopped working but are not looking for work because they are home caring for children who usually would be at school.

This later reason seems particularly so for women – only 8% of the 325,000 women who lost their job added to the unemployed. The rest left the labour force.

This is why the participation rate (the percentage of adults in the labour force) fell from 66% to 63.5%.

Had the participation rate stayed the same – ie all those people who left the labour force were just counted as unemployed – there would be about 1.3m unemployed rather than 823,000, and the unemployment rate would be 9.7% – the highest for 26 years rather than the highest for five years.

The unemployment rate for women would be 10.8% instead 5.8%.

But even these horror numbers hide something: those on jobkeeper.

The government quite sensibly has attempted to keep as many people as possible attached to their jobs. Those people are not actually working, but they are employed so again they do not count as unemployed.

They would, however, be classed as underemployed because they are working fewer hours than they would like.

That is why the number of unemployed rose “just” 15%, but the number of workers underemployed rose 50%.

When we combine the underemployment rate of 12.7% with the 9.7% unemployed it means 22% of the labour force in April was underutilised – well above the previous record of 18.2% set during the 1990s recession.

But things become really dramatic when we look at hours worked.

Hours worked cuts through everything – it doesn’t care whether you are on jobkeeper or in the labour force or not, it just counts how many actual hours of work were done.

And while total employment fell 4.6%, the number of hours worked fell by double that:9.2%.

So massive was the fall in hours worked that in April there was less work done per head of the population than ever recorded.

In March, 86 hours were worked on average by everyone over 15; in April this dropped to 78 hours – lower than even during the 1980s and 90s recession when fewer women worked and a smaller percentage of the population was over the retirement age.

It is a stunning fall.

What took 30 months in the 1990s recession took one month this time.

It is not something that can be switched around just as quickly. The RBA estimates that even if things go as hoped, in two years the unemployment rate will still be 6.5%.

That is not a sign of a need to worry about debt or deficits or winding back stimulus. It is the reverse.

And until the level of actual hours worked is back to where it was before the shutdown no one should be thinking about austerity.

• Greg Jericho writes on economics for Guardian Australia