Australia’s productivity growth has reverted to the same stagnant pattern as before the pandemic, according to the Productivity Commission’s .
Author
David Peetz
Laurie Carmichael Distinguished Research Fellow at the Centre for Future Work, and Professor Emeritus, Griffith Business School, Griffith University
Productivity is complex and often misunderstood in media and policy debates. So before we read too much into this latest data, here are six key things to understand about productivity.
1. It’s about quantities, not costs
“measures the rate at which output of goods and services are produced per unit of input”. So it’s about how many workers does it take to make how many widgets?
Most Australian workplace managers don’t know productivity correctly.
If someone says “higher wages mean lower productivity”, they don’t know what they’re talking about. Wages aren’t part of the productivity equation. People often cite “productivity” as a reason for a policy they like because they can’t say “we like higher profits”.
In fact, high wages can encourage firms to introduce new technology that . If labour becomes more expensive, it may be more profitable for firms to invest in labour-saving technology.
But lower productivity isn’t always a bad thing. Sometimes higher selling prices can lower productivity. It seems odd, but works like this: if prices for commodities such as iron ore or coal are high, it becomes profitable for mining companies to dig through more rock to get to it.
This takes more time. But it’s now worth extracting these small quantities, because they’re so valuable. For this reason, with high commodity prices, mining labour productivity between 2019-20 and 2022-23. Mining productivity had the largest on national productivity growth in 2022-23.
2. Productivity is directed by management, not workers
The biggest single factor that shapes productivity is technology. Who’s responsible for what technology a business introduces? Management. Workers often don’t have much of a say.
OECD research suggests new technology such as artificial intelligence (AI) meets lower resistance from employees over its introduction. That’s because new technology makes their firms more competitive and they want to keep their jobs.
Not surprisingly, there’s lots of research showing management that workers gets greater output.
Output will also be better with an . If people can do more things with their brains, they’ll be more productive.
3. Measuring productivity is dodgier the more complex it gets
Measuring labour productivity – output per unit of labour input – is fairly straightforward if you’ve got a single output that is sold in a free market, and you’re looking at a single input (labour). It’s not hard to measure, or describe, the number of cars produced per worker in a week.
It gets very tricky when you’re looking at (output per unit of, say, labour-and-capital input). Economists can’t even describe the denominator. (What even is a unit of “labour-and-capital”?) So they express what they measure as an index (giving it a value of 100 in some base year). All sorts of bold assumptions get made.
Estimates are highly creative. In its report, the Productivity Commission looked at revisions to quarterly growth figures and found productivity estimates are ““.
On of occasions, initial estimates are out by 0.5 percentage points or more. When your estimate is that productivity increased by 0.5% – the number for the year to this June quarter – the potential for error is huge.
Even more creative assumptions are made when you try to measure productivity in the public sector, when the market is not the aim.
Productivity is higher in classrooms when there are fewer teachers per student. At least, the bean-counters will tell you that, but the students will tell you the opposite.
So you should be very wary when someone says the “productivity challenge is […] greater and more pressing “, when the meaning is so contested.
4. It is best measured over long periods
Productivity growth is so erratic, that you can tell very little from one quarter’s figures. ““, as the PC report said.
Often the best thing to do, as the Australian Bureau of Statistics recognised is to average it over the whole of a “growth cycle”, that is, between one peak of growth and the next.
Trouble is, growth cycles vary in length, and the end point is not easy to pick when it happens, only later.
Growth averaged over a long period is a lot more meaningful than growth measured over a short period. At least the Productivity Commission showed alongside it’s latest quarterly estimates. But chances are your start date will be at a different stage in the growth cycle to your end date, so it’s not that good a measure.
5. Productivity is falling here and overseas
In Australia, productivity growth has been on a long-term decline since the 1960s, with a brief, unsustained upturn in the mid 1990s.
That pattern gives pause for thought: if big reforms to competition policy, industrial relations and wage fixing were , why was that unsustainable, and why did it then continue to decline? It pays to remember that a lot of reforms people advocate in the name of productivity growth have quite different aims and effects anyway.
Internationally, the picture is not much different.
Productivity growth across industrialised countries has unevenly but gradually . The world-wide adoption of what were often called reforms from the 1980s failed to improve productivity growth.
6. Productivity growth once drove living standards. Not any more
In theory, higher labour productivity . In practice, that is driven by the ability of workers to negotiate for higher wages.
It depends on how you measure it and what years you focus on, but from at least the early 2010s, productivity growth was much faster than hourly compensation per employee.
Again, it’s not just Australia. The OECD calls this the “” of wages and productivity.
Just because something can increase potential earnings growth, it does not follow that it will.
