Monday, August 8, 2011

What economists don't know about productivity

Our top econocrats are out banging the drum for more micro-economic reform as the obvious answer to our flagging productivity performance, and civic-minded economics writers are taking up the call. Sorry, but, as the Scots say, I hae me doots.

Economists are trained to believe in the need for ''more micro reform''. They'd want it even if our productivity performance was fine. You get the feeling the physician is prescribing his favourite medicine without bothering to think much about the patent's symptoms.

Assuming you accept their premise that maximising economic growth is the sole object of the human exercise, the economists are right in their incessant quoting of Paul Krugman's line that ''productivity isn't everything, but in the long run it's almost everything''.

For the past 200 years, since the early days of the Industrial Revolution, the material living standards of people in the West have been rising almost continuously, thanks to continuing improvement in the productivity of labour and capital.

Have we had 200 years of continuing micro-economic reform to bring that improvement about? Of course not. So there has to be something seriously wrong with an economy that can't achieve a satisfactory rate of productivity improvement without regular injections of reform.

Like most other things in the economy, micro reform is subject to diminishing marginal returns. And when we run out of things to reform, what do we do then? A counsel of perfection isn't a lot of use.

We need to remind ourselves that governments don't actually run the economy, business people do. So if businesses aren't generating much productivity improvement, the obvious place to look is at the behaviour of business people.

But economists aren't trained to think that way. Thanks to conventional economics' foundation assumption that economic actors are always and everywhere rational - an assumption many economists claim not to really believe but which undergirds far more of what they do believe than they realise - their ideology holds that, as a general rule to which there are only limited exceptions, markets get it right.

It follows that, if the market isn't delivering satisfactory outcomes, it could be a case of ''market failure'', but it's much more likely to be a case of ''government failure''. It must be something the government's doing that's stuffing things up. Thus does every problem in the private sector become the government's fault.

This is almost the complete rationale for micro-economic reform. If you eliminate, reduce or reform government intervention in markets, thereby increasing the intensity of competition in those markets, everything will work much better. (The only major exception to the rule is tighter regulation of competition so as to counter business's natural tendency to gather and exploit monopoly pricing power.)

Though few economists seem to have noticed, our experience with micro reform in the 1980s and early '90s was surprisingly disappointing.

Under the Hawke-Keating government (and the Greiner and Kennett governments) we threw everything at it: floating the dollar, deregulating the financial system, phasing out border protection, reforming taxation, privatising a host of government-owned businesses, deregulating more individual industries than one person could remember and decentralising wage-fixing.

We gathered every bit of low-hanging fruit we could grab, and what did we achieve? A vastly improved productivity performance in the second half of the '90s - which lasted a measly five years before disappearing without trace.

We achieved a big lift in our level of productivity, but we failed to achieve what we should have achieved if micro reform was the true answer to the productivity problem: lift-off. That is, a lasting increase in our rate of productivity improvement.

In the time since then we've backslidden on little of that reform (here I don't class deciding not to denude individual workers of most of their bargaining power as a sin). Nor have we introduced much in the way of anti-competitive measures since then.

So micro reform has proved of limited potency in the search for productivity and, in any case, we've already used most of our ammunition. It's not by chance that we've had so little further reform since Paul Keating's day.

If you examine the long-term record on productivity improvement, you find most of it comes from technological advance (and much of that advance works its magic by pursuing further economies of scale - a factor that works to reduce competition, not increase it).

But if economists are so committed to productivity improvement, how much do they know about the conditions that foster technological advance? Surprisingly little. How much research effort have they put into furthering their understanding of it? Amazingly little.

Their workhorse model merely assumes tech advance is ''exogenous'' - it comes from outside the economic system, dropping from heaven like manna. That assumption is both wrong and unhelpful. Yet their attempt to understand how tech change is produced within the system - ''new growth theory'' - has been allowed to wither on the vine.

The best advice economists have to offer policymakers on promoting tech advance is a version of ''build it and they will come'': get your monetary incentives right (that is, cut the top tax rate) and sit back and wait.

Other disciplines are busy trying to unravel the mysteries of ''innovation'' and what conditions foster it. They could benefit from collaboration with rigorous-thinking economists, but it's all too touchy-feely for most economists.

Judged by their ''revealed preference'' - what they do, not what they say - economists don't have much genuine interest in productivity. They have a pet solution that does no lasting good, and if you're not prepared to swallow their nasty medicine - which no prime minister or premier since Keating's era has been - that's a sign of your lamentable moral weakness.

If our materialism-promoting politicians had any sense, they'd look elsewhere for answers.