As economists sat down to frame their predictions for BusinessDay's latest economic survey, they would have known that their most important forecast, the level of the Australian dollar, was also the most difficult to get right.
Treasury doesn't even attempt to predict where the dollar will go in the budget papers. Like budgets before it, Wayne Swan's latest simply loaded in the Australian dollar's value at the time, $US1.03.
It's a strategy borne out by experience. In my three decades-plus of reporting on business, the economy and the markets, currency predictions have been a persistent corporate fault-line - a source of wildcat losses when hedging strategies go wrong that are on occasion (Pasminco, for example) company-killers.
Treasury isn't alone in avoiding the dark art of currency forecasting. The Reserve Bank does, too, and BHP Billiton, Rio Tinto and the other big miners base themselves in US dollars, the currency used to price the commodities they sell. They accept exchange rate fluctuations in the various countries that they operate in, however they fall.
In the funds management industry, fixed interest funds always hedge foreign exposure because they need to offer an assured yield.
About 30 per cent or $150 billion of the $500 billion that is invested by super funds in shares is invested overseas, however, and about half of it is deliberately unhedged. The unhedged portion is steadily rising as equity managers decide that the cost of continually hedging their exposure is too high given that it basically produces a smoother ride to a similar long-term return as an unhedged strategy.
It may be difficult to forecast, but the Australian dollar is the big wildcard in this year's economic forecasts.
After bottoming out at just over US61¢ in 2008 during the market phase of the global financial crisis, the Australian dollar climbed in two distinct legs, first to just over US90¢ in the second half of 2009 and the first half of 2010, and then to parity with the US dollar by the end of 2010.
Between then and April this year, it traded in a channel on either side of $US1.04 but the descent that began in April has disconnected it from its parity-plus trading range, and pushed it back to where it was after the first leg of its climb out of the global crisis.
Swan's May budget is already out of date. It relied on the-then exchange rate of $US1.03, a stable trade-weighted index of currencies and a 9.8 per cent decline in the relative strength of export and import prices (the terms of trade) to predict nominal economic growth of 5 per cent in the next two financial years.
An earlier budget forecast of tax revenue in the next four years was slashed by $61 billion and the budget papers said the high Australian dollar was pressuring corporate profits, "with firms absorbing costs by squeezing profit margins at the same time as they pass on lower import prices to consumers".
On Friday afternoon, however, the dollar was changing hands at US91.4¢, 11.2 per cent lower than the budget assumed, and 11.8 per cent below last year's average of US103.6¢.
Now let's look at what BusinessDay's stable of economists predict.
Firstly, they predict that, on average, the Australian dollar will be worth US91¢ at the end of this year. The outlier on the low side is University of Western Sydney's Steve Keen, who tips US70¢. The highest is JP Morgan's Tom Kennedy, at US100¢.
Second, they predict that, on average, the Australian dollar will be slightly lower at US89¢ by June 30 next year. Keen is again at the bottom, again at US70¢. Kennedy is at the top, this time at $US1.10. St George Bank's Hans Kunnen is the only other economist in the survey above $US1, at $US1.02.
The average forecast of US91¢ for December 31 this year underlines how uncertain currency forecasting is. Six months ago, BusinessDay also asked the economists where the Australian dollar would be on December 31. The average was $US1.02, almost 11 per cent higher than the new forecast, and 11 per cent above the dollar's value on Friday.
To say currency forecasts need to be taken with a grain of salt is not to say the value of the dollar is not absolutely crucial to the outlook for the economy and the markets over the next year, however.
If the currency stays where it is or goes lower, there will be less pressure on corporate profits than last May's budget predicted.
Revenue and earnings will be higher, share prices will be stronger - even if share earnings multiples do not expand - tax revenue will rise, and the budget deficit will narrow more quickly than expected. The prospects for employment and consumer and business spending will also improve.
The result will be less positive if sliding commodity prices drag the terms of trade down by more than expected. Resources sector profits would be lower than expected, offsetting the bounce a lower Australian dollar delivers.
There's also the US Fed's plan to begin cutting back on quantitative easing sometime later this year and eliminate it by mid next year for the economists to consider. It is pushing interest rates higher overseas, making the "carry trade" into Australian-dollar fixed interest securities less attractive.
The Bank of Japan's own QE program that - at $US75 billion ($82 billion) a month - is only $US10 billion smaller than the Fed's is pushing up on our dollar, and the outlook for global growth is another factor.
Slower global growth would pull commodity prices and the Australian dollar down. Something worse would push both ways as Australia's terms of trade deteriorated but its attraction as a safe harbour was reinforced. It's a dice roll, in other words. In the currency markets, it always is.