The Australian government is preparing to release the sixth iteration of the intergenerational report (IGR) this week, which will include forecasts about the economy and budget over the next 40 years.
But what are the pitfalls or potential utility of making projections over that timeframe? We take a look.
How useful are forecasts?
Many people have discounted the value of forecasting.
“Prediction is very difficult, especially if it’s about the future,” was the view of Danish Nobel physicist Niels Bohr.
That quip, of course, doesn’t stop the prognosticators. Every Reserve Bank meeting brings dozens of expert predictions of an interest rate rise or hold, with little price beyond bruised egos for an erroneous call.
One economist told Guardian Australia his bank tried to be as accurate as possible for clients about one year out. (His bank miscalled the past three RBA monthly meetings.)
The IGR will project out four decades, extrapolating in part from the May budget, which had forecasts running out to 2033-34.
“Three years of the forward estimates is probably about as far as you want to go for a realistic sort of forecasting,” says Ben Phillips, a modelling expert at the Australian National University. “So 10 years is pushing it and 40 years is really what I would just call an interesting projection into the future.”
Should I take the results literally?
This year’s IGR comes just 26 months after the release of the last one, by the then treasurer, Josh Frydenberg. Labor has tended to release them every three years or so when in office, compared with five-year breaks or longer for Coalition governments.
We’re told to expect a 300-page missive, about 50% bigger than the 2021 effort. A whole section on “climate change” will swamp the two mentions made in Peter Costello’s inaugural IGR (all of 100 pages long) two decades ago and the 31 mentions two years ago.
Each day of late, the government has released another dollop of this year’s IGR’s findings, such as how spending on defence and aged care is forecast to swell in coming decades or how population will grow at the slowest pace since federation.
John Hawkins, a former Treasury and RBA economist now at the University of Canberra, noted in a 2019 Treasury report that official forecasts really took off after the second world war. Australia’s first macroeconomic forecasts were prepared in about 1945.
Forecasts, he says, are typically better for macroeconomic variables such as inflation and GDP than for exchange rates or stock market prices. They also “tend to have significant difficulties with large turning points”.
“Despite the significant margins of error, there are some aspects [of IGRs] such as population ageing where we can make reasonable projections, eg projecting the number of centenarians in 2063 based on the number of 60-year-olds now,” Hawkins says.
Danielle Wood, chief executive of the Grattan Institute, agrees that the “pretty ugly” demographic challenges facing Australia are important to discuss even if the precise numbers turn out to be off. Not all developments break in a bad way, though, such as the increase in women’s participation in the workforce that the 2002 IGR didn’t pick.
“The benefits are in the conversations,” Wood says.
How is the data used?
Anthony Scott, a professor in Monash University’s Centre for Health Economics, says debate can also spur action to ensure the worst “doom and gloom” don’t eventuate.
“It’s really about how this information is used, particularly by government, and also by everybody else because it affects the whole economy and all of our society,” Scott says.
For instance, if tax revenues must increase to meet rising aged and disability care demands, should stage-three tax cuts proceed? Some observers think the Albanese government had intended to use an early IGR to shift public support, allowing them to trim the most lucrative tax cuts for big earners. That impetus, though, has gone.
Scott notes technological advances – including artificial intelligence – offer the promise of improved health and reducing need for care. Population trajectories, too, can change with incentives for migration or more babies – or the reverse.
Climate change doesn’t have to be as bad as 2063 might look, if Australia and the rest of the world got serious about cutting greenhouse gas emissions, says Andy Pitman, director of the ARC Centre of Excellence for Climate Extremes.
As Pitman has noted previously, institutions are kidding themselves if they think they can forecast with precision how a heating planet will affect us. Last year’s record floods in Lismore, for instance, would have been far less damaging had the rainfall centre been only a few kilometres different.
“We know broadly what’s coming,” he says, adding “it’s the extremes that have the big impact on the economy”.
“We have a wicked problem of building climate change into our forecasts,” Pitman says.
