A Desk is a Dangerous Place
D. Thomas
March 2001
A Desk is a Dangerous Place From Which to Watch the World
John le Carre
And so is an ivory tower in Canberra or a yacht on Sydney harbour
December 2000 quarter GDP figures for Australia were released today by the ABS and show a contraction of 0.6% relative to the September quarter in seasonally adjusted terms, the first contraction in ten years. Growth compared to the December 1999 quarter was 2.1% - much less than the 4% plus figures we have become accustomed to over the past three years.
And yet in November, in the middle of this severe slowdown, Treasury revised upwards its growth forecast for 2000/01 from 3.75% to 4.0%! And, in August, just before the slowdown, the Reserve Bank of Australia enacted the fifth interest rate hike in less than 12 months.
Much the same was happening in the USA, with aggressive rate rises until mid-2000 and panic setting in early in 2001 as rates were cut twice by a total of 1% in January.
How did the government economic boffins get it so wrong?
Household consumption spending accounts for 60% of GDP and it takes five quarters before rate increases force consumers to cut spending. That is, an interest rate slowdown would have hit in the first quarter of 2001. It came a quarter early simply because consumers went over the top in the December 1999 quarter, so December 2000 was going to be pale by comparison anyway. Consumer spending growth will be slow now until late 2002.
But to someone unaware of the five quarter lag, were there any early signs that may have prevented the most recent errors? Yes, there were several. First, retail sales growth slowed significantly early in 2000 from the hectic pace of 1999. Of course, 2000 was a volatile year due to distortions in spending patterns relating to Y2K, GST, and the Olympics. But the slowing was unmistakable by the end of the unspectacular June quarter, when growth was boosted by pre-GST stockpiling.
In March (Business Review Weekly, March 17 2000, page
31) Charlie Nelson warned that:
- Retail spending growth in 2000 would slow to 5% from 7.2% in 1999 (in fact, it slowed to 4% the difference attributable to the rapid increase in motor fuel prices);
- There were signs of a slowdown in propensity to borrow for consumption spending;
- Demand for housing finance, followed by furniture, was about to slump;
- Y2K special factors increased the risk of inappropriate monetary policy settings.
In addition, consumer marginal propensity to spend fell in the June quarter and stayed low thereafter. That is, consumers reduced the amount they consumed per discretionary dollar, in favour of saving and loan repayment (Chart 1). Even the Westpac Consumer Confidence Index (not as reliable an indicator in our view) was weakening from the March quarter (Chart 2).
Chart 1
Chart 2
A further factor was petrol prices. The rapid rise of fuel prices from late 1999 ate into discretionary funds and this factor alone caused retail spending to drop by 1%. By mid-2000, it was clear that petrol prices were not going to drop in a hurry.
The last time that petrol prices escalated rapidly was in 1990 – just before the last recession. At that time too, interest rates had been rising just as they did in 2000.
The 6% fall in capital formation was no surprise –
it takes three quarters for rate rises to cut capital formation.
In addition, substantial housing investment was brought forward to
beat the GST. This, plus the
pre-Olympics construction, has been known about since 1999 and the
subsequent contraction should have surprised no-one.
The housing slowdown has been evident in surveys of consumer
intentions since early 2000 (Chart 3).
Chart 3
There were two components of GDP that I did find surprising. Exports, typically 20% of GDP, have been rising strongly (by more than 10% in annual volume terms) throughout 2000, stimulated by the plunge in the value of the Australian dollar. In the December quarter, growth slumped to just over 5%. This is a dreadful result in the circumstances and needs to be examined closely to determine whether some remedial action, such as marketing, is needed. It is probably more than a post-Olympics effect.
In a counter-Keynesian oddity, government consumption spending growth slumped after very strong growth throughout 1999 and early 2000. This slump may be reversed in the lead-up to the 2001 federal election.
Overall, it seems that our government forecasters and policy makers sometimes get things very wrong, with severe consequences for business and consumers.
It is clear that the econometric models used by Treasury to predict economic growth are very bad at predicting downturns and recoveries. This was the case globally in the early 1990’s recession and no improvements are evident. An urgent, open review of these forecasting models is needed.
It is also clear that there is insufficient diversity of opinion listened to by the “official family” of government and market economists. It is rare that a dissenting view is published in the Australian Financial Review, for instance. And when it is, that view is often ridiculed – witness Treasurer Peter Costello’s recent attack on Duncan Ironmonger who had the cheek to suggest some weeks ago that Australia may be in recession.
It is time for the politicians and government economists to stop playing god and spend some time in the real world.
What lessons are there in all this for business?
When you hear the cry “this time it’s different” - because
it’s a new economy, or we’ve fireproofed the economy, or ... – then
it’s time to head for the hills.
And if you, as a manager, are spending too long at your desk then you are
in danger.