Will we get a credible forecast?

Eric Lobbecke

KEVIN Rudd has at last stopped the boats: the boats that export coal from Australia.

His decision to yet again postpone deciding on a vitally needed expansion of Queensland's Abbot Point coal port says it all about his political courage; and on top of deferring environmental approval for a major coalmine in the Galilee Basin, it is a further blow to plummeting business confidence.

Yet his government's economic statement has revenues from the minerals resource rent tax nearly trebling to 2016-17. Whether those estimates survive into the Pre-Election Fiscal and Economic Outlook, due tomorrow, will be an acid test of that document's integrity.

Unlike budgets and budget updates, PEFO is not a government document; it is issued on the personal responsibility of the secretaries of the departments of Treasury and Finance.

But that has never stopped governments trying to influence it, notably by tabling an economic update immediately before an election. The secretaries are then placed in an awkward position: if PEFO deviates substantially from the update, difficult questions will be asked.

The result is that PEFOs generally reflect the views already expressed by the government of the day. But this time deserves to be different.

To begin with, the departments' reputations have been hammered by the 2010 PEFO. Far from projecting a credible, independent assessment, that document was even more bullish about Labor's economic prospects than the update the government had released just days earlier.

Thanks to a virtually unprecedented surge in revenues, the 2010 PEFO forecast a $30 billion improvement in the cash balance between 2010-11 and 2011-12, with the balance moving into the black by 2012-13.

That forecast was completely implausible; but once it had been issued under the secretaries' imprimatur, Wayne Swan waved the PEFO like a magic wand, while repeating Labor's mantra of a "return to surplus in three years' time, three years ahead of schedule".

The 2010 PEFO therefore served Labor well: it not only accepted the errors in the government's update but compounded them, invariably in a direction that favoured the government.

Repeating that performance would be a serious mistake, all the more so given how flawed Chris Bowen's update statement obviously is.

Its account of the economy is difficult to understand, as the inflation, employment and gross domestic product forecasts appear to be internally inconsistent. And similar difficulties arise for the individual components.

Typically, for example, lower employment growth leads to a fall in participation, cushioning the impact on the unemployment rate. So it takes a 1 per cent fall in employment growth to cause an 0.5 per cent increase in the unemployment rate.

In the statement, however, a 0.3 per cent decrease in employment growth in a single year causes an 0.65 per cent increase in the unemployment rate for two years, suggesting the decline in employment growth is understated, the rise in the unemployment rate overstated, or both. And then the unemployment rate suddenly falls, boosting government revenues.

Those defects may simply reflect Treasury's forecasting, which seems to lack a coherent conceptual framework. But even were those weaknesses to persist into the PEFO, there are three issues PEFO simply cannot ignore.

The first is the carbon tax. According to Bowen's update, permit prices will rise by over 40 per cent annually from 2014-15 to 2019-20. On that assumption, a dollar invested in European permits will earn an annual return more than 20 times current European bond rates; if so, why haven't European investors thrown every centime they can borrow at this money pump? Indeed, why hasn't the Labor government itself?

The reality is that contract prices in European electricity markets (which reflect expected permit values) show no sign whatsoever of such a leap in permit costs; nor do prices in futures markets for the permits themselves. Accepting Bowen's carbon price estimates would therefore be fatal to PEFO's credibility, as it would greatly understate the likely budget shortfall.

The second is the MRRT. The 2012-13 budget's forecast that the MRRT would raise more than $13 billion in a four-year period has now been reduced by more than 70 per cent.

But even with that revision, Bowen's update has MRRT revenues increasing from $600 million in 2013-14 to $1.7bn in 2016-17. Yet the latest analysts' reports on major mining stocks suggest no such increase. This revenue boost too must therefore be abandoned in PEFO.

Third and last, the National Broadband Network. The government's shares in NBN Co are plainly no longer worth what was spent to acquire them: consistent with public sector accounting standards, and as would happen in the private sector, they must therefore be written down immediately, with the loss being brought to book. Given the amounts already invested, and those committed in the latest budget, that could involve recognising a loss of $9.6bn.

At the same time, the likely cost to taxpayers of the government's guarantees to NBN Co needs to be properly valued and disclosed, as the accounting standards also require. Those guarantees amount to $26bn, with a current taxpayer liability that is likely to be around $5bn.

All these are, of course, mere estimates, and Finance would be well-placed to generate more accurate amounts. Sure, even those may ultimately prove incorrect: after all, virtually by definition, to forecast is to err. To expect omniscience from PEFO would be absurd.

But it is reasonable to expect PEFO to be intellectually rigorous, including by properly explaining and quantifying uncertainties.

That test was failed in 2010; tomorrow, PEFO will be put to the test again. And this time the reputations, and future, of Treasury and Finance will be on the line with it.