Financial modelling is a method of predicting future economic outcomes using complex mathematical formulas that can provide estimates of economic growth, inflation, and unemployment.
It’s a technique used in finance for forecasting financial assets, for example, for investments or the value of stocks.
The modelling method has become popular with academics who use it to try to predict future inflation, but the models used to make predictions of economic performance have also been widely criticised for their reliance on simplistic and unrealistic predictions, according to new research published today in the Journal of Financial Economics.
Researchers from the Australian National University’s Centre for Financial Systems (CFSS) and the University of New South Wales (UNSW) used financial modelling to forecast the impact of a new national financial management system, called the Australian Savings Guarantee Scheme (ASG).
The scheme was announced by the Treasurer, Scott Morrison, in October 2017, with the aim of boosting savings for Australians.
“The ASG is a step forward in reducing the impact on Australia of future changes to the GST and the changes to our tax system,” said Professor Paul Williams, one of the authors of the paper.
However, the modelers say the model used in the paper, based on data from an earlier survey, did not adequately reflect the current state of the scheme, which is being phased in over the next two years.
A lot of the modelling assumptions used to predict the impact and performance of the ASG are based on assumptions that are not based on the reality of the current situation, the authors say.
For example, the ASg assumes that savings rates will stay at the same level or even increase, with only modest rises to take account of inflation and the fact that most Australians do not receive savings, or that inflation will remain low for longer periods.
As well, the models suggest that the ASGs expected savings to be “quite high” when they are actually lower.
Professor Williams and co-author Dr Jody MacGillivray, an associate professor in the department of economics at UNSW, found the modelling approach was flawed because the model assumed that savings would stay high and inflation would stay low for more than two years before reaching their projected levels of 6 per cent per annum.
What they found instead was that the predicted savings were “quite low”, they found, as they expected the scheme to increase spending over the period from July 1, 2021 to April 30, 2022, with inflation expected to remain low.
Dr Williams said the results showed the modelling method was “a poor choice of data source” because the models did not capture what the savings would actually be.
But the researchers also found that the modelling process was flawed when it came to forecasting the future growth of the Australian economy, since the ASGS “generally assumes a flat growth rate over the coming decades”.
“We think that this will be a result of a very low level of income growth for most households and that a relatively high level of investment will lead to a more balanced economy,” Dr Williams said.
To understand how the modelling results could be misleading, Dr Williams and Dr MacGills looked at the ASgs own projections.
They found that a significant portion of the models assumed that the government would not raise its tax rates until 2034, the date when the ASGB is due to end, which was also the date of the government’s own forecast.
In fact, Dr MacGs paper found that when the government raised its tax rate from 2034 to 2043, its forecasts of future income growth were “significantly lower” than the modelling of the economists.
Moreover, Dr McGills and Dr Williams found that modelling that assumed the government was not going to raise its income tax rates in 2022 was “significantly different” to modelling that expected the ASgb to have a “flat” growth rate.
When the ASgowans assumptions were factored in, the modelling predicted a flat, un-expanded economy, Dr. Williams said, and a “much more balanced” economy.
“[The models] showed that there was a significant reduction in the economic growth potential of the economy and in the expectation for future income, which would be a much more robust economic performance,” he said.
Professor Williams said that the fact the ASgg had a lower level of growth than the economic modelling would “not necessarily” mean that the economic models were “completely wrong”.
However he said the fact it had a smaller number of forecasts was “certainly a reason” for caution in modelling the future.
He said the modelling methodology could have a major impact on how the ASgr scheme is administered, and the government has not said whether it will adopt it.
While the modelling methods have been criticised for being too simplistic and inaccurate, they are still “a relatively good example of how economics is not a science”, Professor Williams said in a statement.
Read more on financial modelling