If you've seen the pieces I've been writing on the labour force framework recently, I want to show you something.

It has to do with the way politicians talk about the "unemployment rate".

They often talk about the unemployment rate like it's a competition, and they sow confusion by doing so.

We saw an example last week.

But first, I want to talk about a statistic that politicians should compete over, because it's far more meaningful for job seekers.

It shows how long it's taking to find a job in Australia at the moment, in different areas of the country.

More than 100 labour markets

You'll often hear people talking about the "labour market" as though Australia has a single pool of employed and unemployed people.

But the world's more complicated than that.

In fact, the Bureau of Statistics divides Australia into 108 labour markets.

The geographical boundary of each one is designed to capture the labour supply (where people live) and demand (where people work) of unique markets around the country.

ABS officials consulted experts on labour market geography and analysed the Census of Population and Housing's travel to work data to identify the different areas.

The map below shows where the boundaries are.

I've made a table (see below) that shows the length of time it's been taking an officially "unemployed" person to find a job in all 108 markets.

I've shown the data from March 2021, because that's when Australia's economy was recovering strongly from last year's lockdowns and the participation rate was at an all-time high (before the Delta outbreak in June).

I've included March data from previous years so you can see how they compare across time.

Bear in mind we've been living through extraordinary times in the past 18 months.

In March this year, our international borders were still closed, net overseas migration had turned negative for the first time since 1946, job vacancies were near record highs, and billions of dollars of stimulus were circulating in the economy.

They weren't normal times.

Anyway, you can toggle through the pages by pressing the arrow buttons near the top right-hand corner of the table.

I've put the headline numbers for each state and territory in different colours.

ABS officials say the "duration of job search" statistic can be used to identify the number of long-term unemployed people in every labour market in the country.

And they collect detailed data from all 108 markets to track the fortunes of job seekers who've been looking for work for these lengths of time:

• Under three months
• Between three and 12 months
• Between 12 and 24 months
• Over 24 months  

Of particular concern are job seekers who haven't found work after a year of looking.

If your duration of job search is 52 weeks (12 months) or more, you're considered long-term unemployed.

ABS officials say there are consequences to being out of work for such long periods, including financial hardship and loss of skills.

The longer someone is unemployed the harder it becomes for them to find employment again.

As you can see below, the long-term unemployed have been finding it increasingly difficult to find employment over the past decade.

That fits with the next graph, which aggregates every job seeker's experience into a single statistic.

It shows how long it's typically taking officially unemployed people to find a job in Australia.

Notice how the duration of the job search trended strongly downwards after the 1990s recession, but then started trending back up after the global financial crisis in 2008.

It's become harder for job seekers to find employment in the last 13 years.

Isn't that the type of statistic politicians should be competing over?

A lower unemployment rate?

Which brings us to the last section.

I mentioned earlier how politicians often talk about the "unemployment rate" like it's a competition and how they sow confusion by doing so.

We saw an example of that last week from our federal Treasurer.

You'll see politicians from all parties doing it, but this is the most recent example.

Note the last point Josh Frydenberg made, about the unemployment rate specifically.

Let's consider two things.

First, the unemployment rate does not tell you what you think it's telling you.

As I've explained in Part 1, and Part 2, and Part 3 of this series on the labour force framework, the unemployment rate is not telling you what percentage of Australians are without work each month.

It's far more technical and counterintuitive.

It's telling employers and economists what percentage of people in the "labour force" are immediately available to fill a vacancy.

The unemployment rate ignores millions of other people who aren't employed and who'd like to be working, but who can't get work at the moment.

But when the unemployment rate is framed in competitive terms, it makes you think that the number is telling you what percentage of Australians are without work.

It's very misleading.

Second, in the economics profession at the moment there's a really important debate about the concept of "full employment".

According to the current orthodox theory, an economy reaches "full employment" when the unemployment rate falls to the point where inflation and wages pick up noticeably.

This may sound strange, but it says an economy can be in "full employment" when the unemployment rate is 3 per cent, or 6 per cent, or 8 per cent, or 10 per cent.

It all depends on the structure of the economy at any given moment.

In April, Treasury officials published a paper saying they thought the full employment level of unemployment had changed in Australia in recent years.

They said they used to assume the economy was in full employment when the unemployment rate was roughly 5 per cent, but now they think the economy could have been handling an unemployment rate of 4.5 to 4.75 per cent without causing a wages break-out.

That meant they'd been miscalculating the economy's capacity to employ more people, for the past five years or so.

Could officials have been trying harder to drive the unemployment rate lower for all that time?

It put one in mind of Mr Frydenberg's Budget from 2019-20, which based its strategy around not pushing the unemployment lower than 5 per cent.

See below.

At any rate, the Reserve Bank thinks the unemployment rate could probably fall even lower than that now.

That's part of the reason it's comfortable forecasting an unemployment rate of 4 per cent by the end of 2023.

It thinks the economy can sustain it.

But the question is: Why can the economy sustain it?

What's happened in the economy that means the unemployment rate might have to drop to its lowest level in decades for wages to pick up?

Does it have to do with the creeping casualisation of the workforce? The declining power of unions? The globalisation of our labour supply?

As you can see from the Wage Price Index, wage growth has been declining steadily since the global financial crisis.

It hit record lows last year.

Policymakers have struggled to understand why, because it's a global phenomenon.

But that's the context in which the RBA's forecasts for a 4 per cent unemployment rate need to be understood.

The RBA thinks the unemployment rate can be driven lower in coming years — due, in part, to billions of dollars of stimulus in the economy — and we'll need it to be that low for wages to pick up.

So, see why it can be so misleading to frame the unemployment rate in competitive terms?

A lower unemployment rate can mean a lot of things.