Saturday, July 25, 2015

We're not doing too badly on productivity

Rummaging through the media's rubbish bins this week, I happened upon some good news. According to the Productivity Commission's annual update, the productivity of labour improved by 1.4 per cent in 2013-14.

And get this: in the 12-industry "market sector" of the economy, it improved by 2.5 per cent in that year and by 3.7 per cent the year before.

To give you an idea, the 40-year average rate of market-sector productivity improvement is 2.3 per cent. So, despite all the worrying we've been doing in recent years about our poor productivity performance, it seems we're now doing quite well.

In which case, how come no one wanted to tell us? I can think of three reasons. First is the media's assumption that good news is of little interest to their customers.

Second is that the Productivity Commission's preference is for brushing aside the labour productivity figures and getting us to focus on the figures for "multi-factor productivity", which show an improvement of just 0.4 per cent in 2013-14 and 0.4 per cent the year before. This compares with the 40-year average of 0.8 per cent a year.

Third is that the nation's economists are engaged in a campaign to persuade us we need a lot more micro-economic reform so as to raise our rate of productivity improvement and, hence, the rate at which our material standard of living is rising.

They'd make the same argument whether our productivity performance was good, bad or indifferent, but it helps the selling job if they leave us with the impression our recent performance is poor.

Anyway, let's take a closer look at the commission's new figures. Productivity, which compares the growth in the output of goods and services with the growth in inputs of labour and capital, is a measure of the efficiency of our production. When outputs grow faster than inputs, the economy – the economic machine, so to speak – has become more efficient.

The simplest (and probably least inaccurate) way of measuring productivity is to take the increase in the quantity of goods and services produced during the year and divide it by the increase in the total number of hours worked to produce the stuff.

The main way to increase the productivity of workers is to give them more machines to work with. But the commission believes a more revealing measure is multi-factor productivity. You calculate this by dividing the increase in output by the increase in labour inputs plus the increase in capital inputs (use of machines and other equipment).

The main thing causing an increase in multi-factor productivity is technological advance – the invention of better machines plus improved ways of running businesses. But also improvements in "human capital" – the rising education and skill of the workforce.

That's all fine in theory, but it gets pretty ropey in practice. For a start, we have no way of measuring the productivity of the public sector (healthcare, education and public administration) because, for the most part, it doesn't sell its output in the market.

The market sector covers the financial services, mining, construction, manufacturing, transport, retail trade, wholesale trade, information media and telecommunications, electricity gas and water, agriculture, accommodation and food services, arts and recreation services, rental hiring and real estate services, professional scientific services, administrative support services, and "other" services industries.

That's 16 industries – though, for reasons it doesn't explain, the commission's 12-industry measure of market sector productivity doesn't include the last four industries on that list. Even so, the 12 industries account for 65 per cent of gross domestic product.

A much more serious problem is that the measurement of multi-factor productivity is quite dodgy. It's measured as a residual, meaning that any error in measuring the three other items in the sum will (and does) make the measurement of multi-factor productivity wrong.

More particularly, economists have no way of accurately measuring capital inputs. Just one of their problems is that they can't distinguish between more machines and better machines, meaning their so-called measure of multi-factor productivity excludes much of the technological advance it purports to measure.

The besetting sin of economists is the way they confidently quote their figures to a trusting public, without breathing a word about the data deficiencies and dubious assumptions that lie behind their calculations. When they fail to issue a product warning, it's the duty of the conscientious economic journalist to call them out.

In such circumstances, the commission's results need to be treated with scepticism – particular when, as was true in the noughties, they were so unprecedentedly low as to be implausible.

But let's look at the commission's breakdown of the latest year's supposedly weak result of 0.4 per cent. Half of the 12 industries – all of them in the services sector – achieved remarkably strong improvements, ranging between 1.1 per cent and 5.4 per cent.

Three industries – mining, construction, agriculture – had growing production but marginally declining multi-factor productivity. We know the problems in mining and construction are temporary. Agriculture's poor performance came mainly from drought.

The last three industries – utilities, manufacturing and transport – suffered declining production but lesser declines in inputs, meaning their multi-factor productivity deteriorated quite significantly.

We know the utilities, particularly electricity and gas, are coping with major structural changes, not helped by the earlier misregulation of poles and wires. We know manufacturing is still recovering from the high exchange rate caused by the resources boom. Whatever transport's problem is – we're not told – it will get over it.

That's the trouble with the supposedly worrying figures for multi-factor productivity. Apart from the ropiness of their calculation, when you investigate the stories for the particular industries involved you can't find anything major to worry about.

I'd say that, despite all the barrow-pushers wanting to use poor productivity figures to bolster whatever cause they're promoting, we're not doing too badly on productivity.