Productivity measurements confuse experts

If you don't quite get all the talk about productivity, I'll let you in on a secret - the Productivity Commission doesn't quite get it either.

A few days back it published the first of what will be its regular productivity updates. The good news is it shows Australia getting (slightly) more productive. The bad news is it can't be sure.

Productivity ought to be easy to calculate. It is a measure of how many units a nation or a firm gets out for each unit it puts in. Assessing the productivity of something such as car assembly is straightforward.

If Holden or Toyota were to produce twice as many cars using no more labour and other inputs, they would have doubled their productivity.

But what about hairdressing? If the workers in a salon cut twice as many heads of hair per day would they really have lifted their productivity? That would depend in part on the quality of the haircuts. But it would also depend on something else. Some of what they produce is the attention they give to each customer. It is an output as well as an input.

The phenomenon is especially important in aged care. The Bureau of Statistics recognises this by not calculating productivity for the "healthcare and social assistance" sector even though it's Australia's biggest employer.

Nor does it calculate productivity for "public administration and safety" workers or for "education and training" workers. Combined, those three industries account for 26 per cent of all employment. Two decades ago they accounted for 20 per cent. The hard-to-measure sectors are becoming more important.

An Australian Treasury economist, Gerry Antioch, this month published his estimate of the size of the so-called "persuasive industries" in the United States.

He has included lawyers, public relations specialists, advertising writers and ministers of religion. He has also part-counted workers such as police who spend only part of their time persuading.

Previously regarded as making up 26 per cent of US employment, persuasive industry jobs now account for about 30 per cent of all employment. It is almost impossible to measure their productivity. What is there to count? Conversions? Persuasions? Success in convincing buyers to switch from one brand to another?

It's not only the persuasive and caring industries whose productivity is hard to measure. The Productivity Commission says the measured productivity of the electricity industry is unreliable in part because of the extra money it is spending putting cables underground. The extra spending has benefits - "visual amenity and safety" - but they don't show up as increased output. It's the same for sewage. It is now better disposed of, but that isn't counted as output either.

And some important inputs aren't counted at all, leading to bizarre and otherwise unfathomable fluctuations in measured productivity.

Astoundingly, rainfall isn't counted as an input to the water supply industry. Of course it should be - without rainfall there would be no water supply. But the omission means that during times of drought the measured productivity of the industry dives, even if its efficiency hasn't changed.

Rainfall is also an input for farms. Without it they would produce little, but it also isn't counted when calculating their productivity. As a result, according to a commission research note, recorded productivity growth in the farm sector has at times been negative, "not because farmers became less technically efficient, but because it didn't rain as much".

In mining, the key unmeasured inputs are the minerals being mined.

Yet as the research note puts it: "No amount of conventionally measured inputs - labour, capital, materials etc - can produce a tonne of coal or a barrel of oil without a coal seam or an oil deposit from which to extract it."

The more resources that are extracted, the harder it is to find new easily accessible resources, and so the less productive the industry seems, even if it is operating just as efficiently as before.

The Productivity Commission research note is entitled: "On productivity - the influence of natural resource inputs".

The authors put forward calculations that suggest the mining industry has been getting more efficient even though resource depletion means its measured productivity has been standing still.

Sober economists don't take seriously movements in the quarterly productivity numbers. They are weighed down by errors and likely to be revised.

Only lobbyists and politicians habitually quote them. The Prime Minister did so on Monday.

There's no denying that productivity matters. The Nobel prize-winning economist Paul Krugman famously said: "Productivity isn't everything, but in the long run it is almost everything. A country's ability to improve its standard of living over time depends almost entirely on its ability to raise its output per worker."

But he went on to say that many

of the people who pontificate about productivity "don't know what they are talking about".

Cutting the wages of workers at Holden will help Holden survive, but it won't make them more productive. Productivity is output per unit of input. Labour productivity is output per hour worked.

There are times when it would help us if measured productivity shrank, such as when employers hang on to their workers during downturns, when resource companies spend billions opening new mines and gas fields that won't produce for years, or when the government spends billions on education reforms whose payoffs will be decades away and probably unmeasurable. We need to grope towards higher productivity.

But we need to accept that at times we will go backwards and that much of our groping will be in the dark.

Twitter: @1petermartin

Ross Gittins is on leave.