The big ugly unemployment number from the first phase of the Covid crisis lands today when Statistics New Zealand publishes its labour force data for the June quarter.
But beware of politicians pushing that number to promote stories of economic doom or surprising economic strength.
Economists say today's number, more so than ever, needs to be read in the context of all the available labour market data.
Despite a huge number of job losses during the quarter, there's every chance the unemployment figure will come in looking much better than many had expected.
Economists have picked a wide range in latest previews - up from last quarter's 4.2 per cent to between 5 per cent and 6.4 per cent.
Usually the variance in predictions is just a small fraction of a per cent in either direction.
ASB senior economist Mike Jones says things like the labour market participation rate, underemployment and hours worked will be the key to understanding why the unemployment figure lands where it does, after one of the weirdest quarters in New Zealand's economic history.
"Given the various moving parts, the unemployment rate could conceivably print anywhere in a 5-8 per cent range without surprising us greatly," he says.
"We expect that employment contracted 2.8 per cent over the quarter, amounting to a loss of around 74,000 jobs."
Stats NZ labour market and household statistics senior manager Sean Broughton agrees the unprecedented nature of Covid-19 and the subsequent lockdown will make a deeper look at the data crucial.
"All of the action will be under the hood," he says.
Today's labour market data includes the official unemployment number - as distinct from more immediate but less comprehensive data like the weekly number of people on a Jobseeker Benefit.
There is a correlation between the two figures, but Jobseeker Benefit numbers don't catch large numbers of people who are unemployed but not on a benefit.
To further confuse things, Jobseeker Benefits do include people who may have some hours of work - but who wouldn't count as unemployed in the official stats.
To get a complete picture of the labour market, Broughton and his team run the Household Labour force Survey, which gives us that official figure.
Polling 15,000 households (about 30,000 people) across 13 weeks they generate an average for the quarter - rather than a set point in time.
The survey has a quite specific definition of who is unemployed.
"They have to be not working," Broughton says, "They have to have been actively seeking work in past four weeks or due to start in four weeks. They have to be actually making an effort, contacting employers and applying."
But any work at all, no matter how little, means you don't count as unemployed.
So even "if you are just working one hour and you're paid the minimum wage for that hour we'd pick that as part-time," Broughton says.
"We'd then ask you if you want to work more hours. And if you said yes, then you are under-employed. That puts you into those more nuanced categories within underutilisation."
For all its strict parameters, the Household Labour Force Survey is a useful tool to measure the relative state of the economy.
It dates back to 1986 and isn't subject to the political definitions of unemployment used to determined eligibility for benefits.
It means, for example, we can see how bad things are relative to the record 11 per cent unemployment rate we hit in the recession of 1991.
"I see the Q2 labour stats as a key waypoint, in what would be quite a marked deterioration in the labour market," says ASB's Jones. "The really big one is the participation rate. So if you look at what's happened in the US and Australia, it basically meant that the unemployment rate didn't go up nearly as much as people expected it too."
That's partly a reflection of lockdown, he says.
"If you lose your job in lockdown, you're not going to be actively seeking employment – as you have to do to be classified as unemployed. People just waited it out because they knew that no firms were hiring."
Under-employment will also be important, he says.
That's the number of people who are employed but say they'd like to work more.
"Another one will be hours worked. That's going get hammered due to lockdown," Jones says.
"Looking at those numbers will give you an indication of the real slack in the labour market – rather than the unemployment number which might stay in the 5 or 6 [per cent range]."
For all that uncertainty, where the official unemployment number lands will still matter relative to some key forecasts.
In particular the forecast of the Reserve Bank, which it is likely to beat.
The RBNZ has 7 per cent pencilled in for the June quarter, with 9 per cent as its peak by the end of the year.
"For the RBNZ, the labour data comes exactly a week before the release of the 12 August Monetary Policy Statement, likely too late to be factored into the Statement's economic forecasts," Jones says.
Regardless, employment will remain below the Bank's Maximum Sustainable Employment target for some time.
Because the bank is forward-looking, it will remain in a mood to keep "grinding monetary policy easier" for some time yet, he says.