It’s time for more sensible thinking about productivity

When will we tire of all the bulldust that’s talked in the name of hastening productivity improvement? We never do anything about it, but we do listen politely while self-appointed worthies – business people and econocrats, in the main – read us yet another sermon on the subject.

Trouble is, when the sermons come from big business – accompanied by 200-page reports with snappy titles – they boil down business lobby groups doing what lobby groups do: asking the government for special favours – aka “rent-seeking”.

The main thing that improves the productivity of labour is employers giving their workers more and better machines to work with.Credit: AP

You want higher productivity? It’s obvious: cut the company tax on big business, and give us a free hand to change our workers’ pay and conditions as we see fit.

When the sermons come from econocrats, they’re more like professional propagandising: calls for “reform” – often of the tax system – that are usually theory-driven and lacking empirical evidence that they really would have much effect on productivity.

What we get in place of genuine empiricism is modelling results. Models are a mysterious combination of mathematised theory, sprinkled with ill-researched estimates of elasticity and such like.

We’ve become so inured to all this sermonising that we’ve ceased to notice something strange: although in a market economy it’s the behaviour of business that determines how much productivity improvement we do or don’t get, any lack of improvement is always attributed to the government’s negligence.

This is where the business rent-seekers and the econocrat propagandists are agreed. The econocrats willingness to point at the government comes from the biases in their neoclassical theory, which assumes, first, that businesses always respond rationally to the incentives they face and, second, that government intervention in markets is more likely to make things worse than better.

Big business is happy to use this ideology to hide its rent-seeking. (If you wonder why neoclassical economics has been dominant for a century or two despite surprisingly little evolution, it’s partly because it suits business interests so well.)

The other strange thing we’ve failed to notice is that the modern obsession with the tax system and regulation of the labour market has crowded out all the economists’ conventional wisdom about what drives productivity improvement over the medium term.

But before we get to that wisdom, a health warning: there’s a famous saying in economics that the sermonisers have stopped making sure you know. It’s that, for economists, productivity is “a measure of our ignorance”.

Just as economists can calculate the “non-accelerating-inflation rate of unemployment”, and kid themselves it’s next to infallible, when you ask them why it’s gone up, or down, all they can do is guess at the reasons, so it is with calculations of productivity. Economists can’t say with any certainty why it’s up or why it’s down. They don’t know.

It’s the behaviour of business that determines how much productivity improvement we do or don’t get.

Ross Gittins

Even so, in the present opportunistic sermonising, all that the profession thought it knew has been cast aside.

Such as? That productivity improvement is cyclical and hard to measure. Recent quarterly results from the national accounts will probably change as better data come to hand, and the accounts are revised.

It’s true that the measured productivity of labour actually has fallen over the three years to June this year, but it’s likely this is, to a great extent, a product of the wild swings of the pandemic and its lockdowns. As Reserve Bank economists have argued, these effects should “wash out”.

It’s well understood that the main thing that improves the productivity of labour is employers giving their workers more and better machines to work with. But Australia’s level of business investment as a share of gross domestic product is low relative to other rich countries.

Growth in non-mining business investment has declined from the mid-2000s and stagnated over the past decade. It’s grown strongly recently, but it’s not clear how much of this is just tradies taking advantage of lockdown tax concessions to buy a new HiLux ute.

Point is, why do the sermonisers rarely acknowledge that weak business investment spending does a lot to help explain our weak productivity improvement?

Another factor that should be obvious is our recent strong growth in employment, the highest in about 50 years, with many people who employers wouldn’t normally want to employ, getting jobs. This will lower the workforce’s average productivity – but it’s a good development, not a bad one.

Again, why do the sermons never mention this?

Yet another part of the conventional wisdom it’s no longer fashionable to mention is the belief that productivity improvement comes from strong spending – by public and private sectors – on research and development. Have we been doing well on this over the past decade or so? I doubt it.

And, of course, productivity improvement comes from giving a high priority to investment in “human capital” – education and training.

So, why no sermons about the way we’ve gone for a decade or more stuffing up TAFE and vocational education, or the way school funding has given “parental choice” for better-off families priority over the funding of good teaching in public schools?

Too many of those sermons also fail to mention the small fact that all the other developed economies are experiencing similar weakness – suggesting that much of our poor performance is explained by global factors, not the failure of our government.

Related to this, the preachers usually compare our present performance with a much higher 30- or 40-year average, implying our weak performance is something new, unusual and worrying.

Or, we’re told that, whereas productivity improved at an annual rate of 2.1 per cent, over the five years to 2004, it worsened to 0.9 per cent over the six years to 2010, and improved only marginally to 1.2 per cent over the nine years to 2019, before the pandemic.

This is all highly misleading. The fact is that periods of weak improvement are more common than periods of strong improvement, which are rare.

Our period of unusually strong improvement from the late 1990s to the early noughties is paralleled by America’s strong period from 1995 to 2004, which the Yanks usually attribute to rapid productivity improvement in the manufacturing of computers, electronics and semiconductors.

We usually attribute our rare period of strong improvement to the belated effects of the Hawke-Keating government’s program of microeconomic reform. Maybe, but computerisation and the information revolution are a more plausible guess.

Either way, contrary to the sermonisers’ implicit claim that the present period of weak improvement is unusual, it may be closer to the truth that weakness is the norm, interspersed by occasional bursts of huge improvement, caused by the eventual diffusion of some new “general-purpose technology” – the next one likely to be generative AI.

The Business Briefing newsletter delivers major stories, exclusive coverage and expert opinion. Sign up to get it every weekday morning.