Monday, June 6, 2011

This time it's a recession we don't have to have

Though nothing is certain in the unpredictable world of the national accounts, it's highly unlikely we'll see a second, successive quarter of ''negative growth'' when the figures for this quarter are released in early September. Which, in a way, will be a pity.

Why? Because even if the hiccup caused by our natural disasters had spread itself over two quarters - after the 1.2 per cent contraction in the March quarter - not even the most ignorant journalist or most excitable market trader would have believed the consequent ''technical recession'' was a real recession.

The reason the rule about two successive quarters of falling real gross domestic product amounting to a ''technical'' recession (whatever that is) won't lie down and die is its handiness: it's simple to judge, objective and involves minimum waiting.

The reason it should bite the dust is that it's a completely arbitrary rule of thumb containing no science, which is perfectly capable of telling us we're in recession when we're not, or failing to tell us we're in recession when we are.

It's also unreliable in another sense because it's based on the Bureau of Statistics' first estimate of the quarterly change in GDP, which is subject to heavy revision in subsequent months and years. (Keep reading.)

It was because all the contraction after the global financial crisis was crammed into a single quarter (the fall of 0.9 per cent in the December quarter of 2008) that the Rudd government got away with the claim that we avoided recession even though, in truth, we had a mild recession involving a rise in unemployment of almost 2 percentage points.

Kevin Rudd, Ken Henry & Co timed their stimulus packages with the clear (if unacknowledged) objective of ensuring there weren't two quarters of contraction in a row. They did this in the belief that, whatever the realities of the situation, the fuss the media would make about ''technical recession'' would be certain to further damage business and consumer confidence and make the talk of recession self-fulfilling and the downturn deeper.

They were right and they deserve a medal. They understood - as few macro-economists do - the central role the management of our animal spirits plays in determining the severity of downturns.

They also understood that the effectiveness of cash splashes and other give-aways is determined as much by their effect on how people feel about the future as by the size of the increase in spending they immediately bring about.

Once the second successive quarter had been avoided, the government happily trumpeted the (false) news that we'd avoided recession - no doubt believing it was merely reinforcing the more confident outlook.

But no good deed goes unpunished. The opposition was happy to accept the no-recession line but, in a novel twist on the post-hoc-ergo-propter-hoc fallacy, turned it back on the government, arguing that all the money Labor spent trying to moderate a recession was obviously wasted. (The fallacy says, since A preceded B, therefore A caused B. The opposition's version was, since there was no recession, therefore there was no need for all the spending to stop it happening.)

Not many people remember it was the old two successive quarters rule that lured Paul Keating into making his hugely resented remark about the recession we had to have. Though, in 1990, Keating and Treasury had been assuring us we were in for no worse than a ''soft landing'', by the time the national accounts for the June quarter showed a contraction of 0.9 per cent, most people needed no convincing we were in deep trouble.

Even so, Keating persisted with his denial. But by late November, on receiving the September quarter accounts showing a whopping contraction of 1.6 per cent, he realised the game was up and (to his eternal regret) decided to brazen it out, preferring to be seen as a conniving knave rather than the miscalculating fool he really was.

At his news conference to respond to the accounts, his opening words were: ''The first thing to say is, the accounts do show that Australia is in a recession. The most important thing about that is that this is a recession that Australia had to have.''

But here's the joke. Remember what I said about initial estimates being subject to heavy revision? By now, the first of his successive contractions has been revised from minus 0.9 per cent to plus 0.4 per cent. The second has been revised from minus 1.6 per cent to minus 0.4 per cent.

So applying the two-quarter rule to the bureau's by-now reasonably accurate estimates, Keating confessed to a recession that didn't exist. Except, of course, that at the time everyone knew from the evidence that it did - and they were right.

Since, according to the latest estimates, the economy grew in the following, December quarter, it wasn't until early September the next year - nine months later - that the two-quarter rule was signalling the arrival of recession.

A rule of thumb that throws up so many false negatives - in 1990-91 and 2008-09 - is a rule that should be ditched. The economist Saul Eslake has road-tested a different rule, showing it has produced no false signals. It defines recession as ''any period during which the rate of unemployment rises by more than 1.5 percentage points in 12 months or less''.

But I prefer the definition offered by David Gruen, of Treasury: ''A sustained period of either weak growth or falling real GDP, accompanied by a significant rise in the unemployment rate''.