The author has been described by News Ltd as an "iconoclast", "Svengali", a pollie's "economist muse", and "pungently accurate". Fairfax says he is a "Renaissance man" and "one of Australia’s most respected analysts." Stephen Koukoulas concludes that he is "85% right", and "would make a great Opposition leader." Terry McCrann claims the author thinks "‘nuance’ is a trendy village in the south of France", but can be "scintillating" when he thinks "clearly". The ACTU reckons he’s "an enigma wrapped in a Bloomberg terminal, wrapped in some apparently well-honed abs."

Thursday, November 5, 2009

Forecasting the future

Today I am presenting on housing affordability at the annual Melbourne Institute Economic and Social Outlook conference, which is the biggest economic conference of the year.

Other speakers include Treasurer Wayne Swan, Julia Gillard, Glenn Stevens, Warwick McKibbin, Malcolm Turnbull, Lindsay Tanner, Joe Hockey and most of the Federal Government and Shadow Cabinet. You can see the program here.

If there is one thing that we have learnt from this crisis, it is that public and private sector analysts, strategists, and economists are terrible at forecasting the future (notwithstanding some silly claims by certain individuals who rank themselves amongst a handful of other deities that foresaw the calamity).

This performance should really be no surprise given the inherently chaotic nature of financial systems. It is fair to say that while some such as the BIS highlighted the very vulnerabilities that ultimately exposed the system, and hypothesised about the specific crises that such risks could cause, nobody predicted its timing or gravity. Some see this inability to forecast the future as perverse vindication of the efficient markets hypothesis—I’d venture that these folks are clutching at straws while ignoring much evidence to the contrary.

I read a lot of research produced by investment banks and it is, frankly speaking, pretty sad watching all these economists mechanically revise their woefully inadequate projections in a lemming-like fashion to fit the facts after the event. And it happens like clockwork every month. Good men are made to look like fools.

These professional shortcomings were embarrassingly evident both leading into the crisis and during the rapid recovery since it has purportedly passed. Of course, chance dictates that if you throw enough analysts into a room a minority will accurately predict some states of nature at various points in time. And for a moment there, these guys get feted and walk the streets like giants among Lilliputians. But this euphoria tends to last only as long as their next forecast. For the life of me I cannot work out why they put themselves through this torment. I guess it is also a proven fact that there is deep demand for astrologers.

This is not to say that economists do not play a very important role in our society. I have not heard this analogy employed before (I presume it has somewhere), but for mine economists are best viewed as financial doctors. They are extremely good at collecting and analysing complex information. The most capable economists can provide profoundly important insights into exactly what transpired in the past—and what factors conspired to cause events that can change a nation’s socio-economic trajectory.

Economists are also exceedingly valuable in assessing the present. By critically appraising vast reams of otherwise impenetrable data they can supply us all with vital guidance on the current rhythms of the economy.

So when you think about it, economists are like the financial equivalents of medical practitioners. They can generally do a pretty good job of categorising what ailments you have been afflicted by. They also offer a useful account of themselves in judging your current state of health. Where they depart from doctors is in their attempt to reach beyond their intrinsic abilities by divining the course of the future. In this respect, they are no better than Nostradamus.

Imagine how absurd it would sound if a doctor produced a monthly report that predicted what maladies you were going to contract in the coming months! Imagine if you then medicated (read invested) on the basis of this advice. While this might at first blush appear marvellous, you would surely twig to the fact that the good doctor’s prognostications have no more value than if you had asked your 80 year old grandmother to manufacture the same output.

Perhaps, therefore, the profession would be well served by refraining from inundating us with meaningless long-term forecasts, which in the beguiling nature of all predictions tend to induce unfounded confidence. Perhaps we would all be better off if they stuck to unearthing analytical learnings from the past and appraising the present in an attempt to cautiously prepare for the future.

One policy opportunity here would be to force analysts, strategists, and economists to disclose the expected errors associated with their projections based on their past performance. I think you would quickly find that all but the most highly probable estimates would disappear.

With that aside in mind it is worthwhile dwelling for a moment on a recent innovation in the Westpac-Melbourne Institute consumer confidence survey that provides insights into household views on the residential property market.

In the May, July and October surveys a new question was added querying consumers about their expectations for house price growth over the next 12 months. This research implied that 73% of respondents expect house prices to rise over the next year with all subsidiary cohorts also positive (see Table 6 extracted from the subscriber-only RP Data-Rismark Monthly).

Fascinatingly, the most bearish households are to be found in NSW despite the fact that house prices have increased strongly in this state during the year-to-date. Equally interesting was the finding that the most bullish consumers are located in Western Australia, which while rebounding in 2009 has performed very poorly since September 2007. Westpac’s Matthew Hassan comments:
“Some 73% of respondents expect prices to increase over the next 12 months with 15.9% expecting no change and 9.9% expecting a decline. The proportion expecting an increase compares with 53% in July and 32% – a minority – in May. Overall, those expecting rises outnumbered those expecting falls by a substantial 63.7%, another big jump on the net 34.5% recorded in July and net 0.6% in May.

Consumers across all states are now strongly bullish on prices. Consumers were marginally more upbeat in WA (optimists outnumbering pessimists by 71.2%) and Victoria (optimists outnumbering by 68.4%) and slightly more restrained in NSW (optimists ahead by a net 54.9%).

The biggest turnaround in expectations continues to come in the resource states, with the net percentage of optimists minus pessimists rising 44pts in WA and 35pts in Qld between July and October. Both of these markets were notably weaker than the rest of the nation through 2008, especially in terms of prices.

The degree of consensus is particularly striking across age groups with all now expecting price rises and a fairly uniform swing from month to month. The breakdown by income category shows more variation with a notably larger swing towards optimism in the $61-90k pa income ranges between July and October – groups that were the most pessimistic back in May. The only sub-group across all demographic breakdowns to report a decline in price expectations between May and October was those on incomes of $21-30k pa. “Interestingly, the same survey recorded a pull-back in responses to the question: “Is now a good time to buy a dwelling?”, albeit to still optimistic territory.”