Monday, September 18, 2023

Productivity debate descends into damned lies and statistics mode

Last week we got a big hint that the economics profession is in the early stages of its own little civil war, as some decide their conventional wisdom about how the economy works no longer fits the facts, while others fly to the defence of orthodoxy. Warning: if so, they could be at it for a decade before it’s resolved.

Economists want outsiders to believe they’re involved in an objective, scientific search for the truth and are, in fact, very close to possessing it. In reality, they’ve long been divided by ideology – views about how the world works, and should work – which is usually aligned with partisan interests: capital versus labour.

You see this more clearly in America, where big-name “saltwater” (coastal) academic economists only ever work for Democrat administrations, while “freshwater” (inland) academics only work for the Republicans.

In the 1970s, the world’s economists argued over the causes and cures for “stagflation” – high inflation and high unemployment at the same time. Then, in the 1980s, we had a smaller, Australian debate over how worried we should be about huge current account deficits and mounting foreign debt, won convincingly by the academics, who told the econocrats to forget it – which they did.

Now, the debate is over the causes of the latest global surge in inflation. At a time when organised labour has lost its bargaining power, while growing industry “concentration” (more industries dominated by an ever-smaller number of big companies) has reduced the pressure from competition and increased the pricing power of big firms, is a lot of the recent rise in prices explained by businesses using the chance to increase their profit margins?

A related question is whether it remains true that – as business leaders, politicians and econocrats assure us almost every day – all improvement in the productivity of labour (output per hour worked) is automatically reflected in higher real wages.

And that’s the clue we got last week. The Productivity Commission issued a study, Productivity growth and wages – a forensic look, that concluded that “over the long term, for most workers, productivity growth and real wages have grown together in Australia”.

So, all the worrying that silly people (such as me) have been doing – that the workers are no longer getting their cut of what little productivity improvement we’ve seen in recent years – has been proved to be a “myth”.

For the national masthead that prides itself on being read by the nation’s chief executives, this was a page one screamer. Apparently, even though real wages are 4 per cent lower than they were 11 years ago, workers are getting “their fair share of pie”.

When workers’ real wages rise by less than the improvement in labour productivity, the study calls this “wage decoupling”. It says “it is important to get the facts right on wage decoupling. Unfortunately, debates about the extent of wage decoupling, its sources and its implications are often dogged by differences in the methods and data”.

“This is because analysts can pick and choose among a wide range of measures of real wage growth, and their choices can lead to different, sometimes misleading conclusions.”

This is very, very true. Trouble is, sauce for the goose is sauce for the gander. The clear inference is that “the commission’s preferred measure” is the single correct way of measuring it, whereas all those who get different results to us are just picking the methodology that gives them the results they were hoping for.

Get it? I speak the objective truth; you are just fudging up figures to defend your preconceived beliefs about how the world works. Yeah, sure.

I hate to disillusion you, gentle reader, but this is what always happens in economics whenever some group says, “I think we’re getting it wrong.” They produce calculations to support their case, but some don’t like the idea, so they produce different calculations intended to refute it.

Because economics is factionalised, most debates degenerate into arguments about why my methodology is better than yours. That’s why a change in the profession’s conventional wisdom can take up to a decade to resolve. But intellectual fashions do change.

The study finds that the mining and agriculture industries – which account for only 5 per cent of workers – have experienced major wage decoupling over the past 27 years, but for the remaining 95 per cent of workers, in 17 other industries, the difference between productivity growth and real wage growth has been “relatively low”.

Sorry, but that’s my first objection. It’s not relevant to compare productivity growth by industry with real wage growth by industry. Some industries have high productivity, some have low productivity and, in much of the public sector, productivity can’t be measured.

Despite the things it suits the employer groups to claim, the reward held out to workers for at least the past 50 years has never been that their real wages should rise in line with their own industry’s productivity.

For reasons that ought to be obvious to anyone who understands how markets work, it’s never been promised that, say, carpenters who work in mining or farming should have rates of pay hugely higher than those who work in the building industry, while the real wages of carpenters working in general government should never have changed over the decades because their (measured) productivity has never changed.

It’s an absurd notion that could work only if we could enforce a rule that no one could ever change jobs in search of a pay rise.

No, as someone somewhere in the Productivity Commission should know, the promise held out to the nation’s employees has always been that economy-wide average real wages should and will rise in line with the trend economy-wide average improvement in the productivity of labour.

When you exclude the two industries that contribute most to the nation’s productivity improvement, it’s hardly surprising that what’s left is so small you can claim it wasn’t much bigger than the growth in most workers’ real wages.

Then you tell the punters that, over 27 years, they are less than 1 percentage point behind – a mere $3000 – where they were assured they would be.

The report finds – but plays down – that the national average real wage fell behind the national average rate of productivity improvement by an average of 0.6 percentage points a year – for 27 years.

That’s if you measure wages from the boss’s point of view (which is economic orthodoxy) rather than the wage-earner’s point of view. But I can’t remember hearing that fine print explained in the thousands of times I’ve heard heavies telling people that productivity improvement automatically flows through to real wages.

View wages from the consumer’s perspective, however, and the national average shortfall increases to 0.8 percentage points a year. And nor did anyone ever tell the punters that it may take up to 27 years for their money to arrive.

You guys have got to be kidding.