When the February unemployment figures are released on Thursday at 11.30am, there is a decent chance that, according to one measure, Australia will be in recession. But while it might not feel like the recessions of the 1980s or 1990s, it should serve as a warning to both the Reserve Bank and the government that the economy is extremely weak and does not need any more rate rises.
Whenever we have awful GDP growth like we saw in December, talk always turns to recessions, because of the old definition that a recession is two consecutive quarters of negative GDP growth.
It’s not a particularly good definition and most journalists will now preface it as a “technical” recession.
Personally, I don’t care much about GDP growth because in my view the only thing that matters with recessions is jobs.
Did anyone think that the 1990s recession only began in 1991 when the June quarter that year showed GDP had fallen 0.2% off the back of a fall of 1.3% in the March quarter? I doubt it. More likely everyone twigged rather earlier, given in 1990 the unemployment rate went from 6.1% to 8%.
We don’t need to wait for GDP growth to know we are in a recession, and doing so can mean we wait too long to do anything good about it.
It’s why I prefer to use the US economist Claudia Sahm’s rule of recessions.
Her method looks at changes in unemployment rates.
Specifically, if the three-month average rises more than 0.5 percentage points above the minimum of the previous 12 months (for example 4.2% compared to a minimum of 3.6%) then welcome to recession town, population you.
But if governments and central banks act as soon as the Sahm rule is triggered, you can avoid the recession becoming a deep one.
It’s clear that the measure works in Australia as well as the US. It’s also useful for comparing the size of recessions.
The 1990s recession, for example, was much worse here, while there’s a reason we call it the GFC and in the US it is the Great Recession.
Also it’s pretty clear measures such as the quasi job guarantee of jobkeeper was a lot better than whatever Donald Trump was doing in 2020:
If the graph does not display click here
This measure is important because of where we are right now.
In January, the Sahm measure hit 0.4 percentage points. If Thursday’s unemployment rate is 4.2% or higher, we will have gone over the 0.5 percentage point recession line.
Even if it doesn’t happen today, the RBA’s most recent estimates for unemployment suggest we go above 0.5 percentage points in the next couple of months:
If the graph does not display click here
The RBA knows about this rule because last year when the RBA examined the likelihood of a recession it argued that in Australia the Sahm rule should apply when the unemployment is 0.75 percentage points above the 12 months minimum not 0.5 percentage points.
This rather conveniently means they are not forecasting a recession.
The problem is that since 1978 every time the unemployment rate has risen in a year by the 0.6 percentage points the RBA is estimating it will rise, it has kept rising by more than the 0.75% the RBA says triggers a recession.
In effect the RBA is arguing we will not go into a recession despite forecasting rises in unemployment that have led to a recession.
Another concern is that youth employment often indicates where total unemployment is heading, and right now things are not great. The unemployment rate for 15- to 24-year-olds in the past 12 months has gone from 7.9% to 9.4%:
If the graph does not display click here
We can also adapt Sahm’s method for other figures – such as the change in the average hours worked per capita.
Even a casual look shows that the average number of hours worked falls pretty dramatically when a recession (or the GFC) hits:
If the graph does not display click here
If we use the same three months average but this time compare it to the maximum of the past year (because unlike unemployment, a higher employment number is better) we get the following graph that suggests we’re about to enter a recession:
If the graph does not display click here
So is a recession locked in stone?
Well, Claudia Sahm herself says her rule is “a historical pattern, not a law of nature”. And right now is a good time for the historical pattern to break because the economy remains very weird.
Unemployment and average hours worked might be falling, but oddly employment itself is holding up pretty well.
In the 12 months to January, employment grew a very healthy 2.6% – above the 30-year average of 2.1%. And the percentage of adults employed in the past year has fallen only slightly from 64.6% to 64.1%:
If the graph does not display click here
In recessions, unemployment rises, but employment also falls, and we are not seeing that now.
But the crucial point is that the Sahm rule of recessions is designed to ensure governments and central banks get in front of the problem rather than wait six months to find out how bad things are.
Alerts should be flashing in both the Treasury and the Reserve Bank. Now is not the time to do anything more to slow the economy – it’s already in danger of stalling.
• Greg Jericho is a Guardian columnist and policy director at the Centre for Future Work
