Peter Martin* says an economic model swung the last two elections – and it now points to a coalition victory.
This election will be won by the Coalition and Prime Minister Scott Morrison if the economic models work as expected – and they usually do.
A model refined in 2000 by University of Melbourne economists Lisa Cameron and Mark Crosby found that most federal election outcomes in records dating back to 1901 can be predicted fairly well by just two economic indicators.
And these are not the indicators one would expect.
Growth in real wages in the year preceding the election does not appear to have an effect on the chances of the ruling party returning to power.
(Which is just as good for the Coalition, as the purchasing power of salaries has diminished.)
Similarly, GDP (which is shorthand for gross domestic product, the measure meant to encompass almost everything known about the state of the economy) is found to be “not strongly correlated” with government support. in place in Australia, although it is in the United States.
The only two economic variables that matter, and they seem to matter a lot, are the rate of inflation and the rate of unemployment, each in a different way.
For inflation, the higher it is, the more the incumbent suffers, as one might expect.
For unemployment, what is important is not the rate itself.
High rates and low rates do not seem to be covered by the ruling party.
What is registered with us, for the most part, is the rate change.
Voters reward lower unemployment
A government accused of reducing the unemployment rate is rewarded, while a government accused of raising the rate is punished.
Cameron and Crosby find that a one percentage point increase in the unemployment rate reduces a government’s vote share by 0.58 percentage points.
And they find a wrinkle.
In swing seats, coalition governments are likely to be punished if unemployment rises, while Labor governments are likely to be rewarded.
They say their findings are “consistent with voters’ perception that Labor is more committed to reducing unemployment”.
In 2005, economists Andrew Leigh (who later became a Labor politician) and Justin Wolfers applied a slightly different model to the 2004 election.
They found the result to be correct, but underestimated the size of the coalition victory.
The model generally performs well
In the latest edition of the Australian Economic Review, University of Queensland economist Hamish Greenop-Roberts applied Cameron and Crosby’s model to the last four elections, the one Labor won in 2010 and those won by the Coalition in 2013, 2016 and 2019.
He found he picked the outcome three out of four times, putting it on par with the polls and betting odds, which also picked the correct outcome three out of four times.
The crucial difference is that the economic model has performed well in each of the last two elections – something the others have clearly failed to do.
Asked this week what the economic model would predict for the current election, Greenop-Roberts notes that on the one hand unemployment is much lower than it was at the start of this government’s term (and much lower than which was planned), which, according to the model, should help him get re-elected.
On the other hand, inflation is abnormally high, which the model says would hurt.
What matters in predicting the outcome is the size of each move and how large the size of each move has been in the past.
And it’s not a contest.
The effect of the dramatic reduction in the unemployment rate (from 5.2% to 4%) is so significant that it greatly exceeds the effect of the inflation rate of 3.5%, “paving the way for the return of Coalition “.
Unemployment beats inflation
What happened to unemployment is so significant that Greenop-Roberts says an inflation rate of at least 8-9% would be needed to reverse the prediction.
Whatever Australia’s official inflation rate is ahead of Election Day (there will be an update next Wednesday), it will most likely be above its current level of 3.5%, but still well off 8. at 9%.
Or maybe the model will be wrong when it comes to inflation.
Greenop-Roberts points out that since the early 1990s, a whole generation of voters has entered adult life without experiencing severe inflation, and may either be alarmed by it or misunderstand the concept.
This election could serve as a test.
And it may be one of the few elections where the state of the economy does not predict the outcome.
Opinion polls have done poorly in the last election, but they could pick up and they point to a Labor victory.
The betting markets have also turned sour and only hint at a Labor victory.
Polls, experts and even the model can be wrong
Experts are often wrong.
Greenop-Roberts points to a poll of 13 experts released two days before the 2019 election.
Twelve predicted a Labor victory.
The only expert who had not predicted that the coalition would be forced to govern jointly with independents, a prediction well short of the outcome, which was an overall coalition victory.
The reality is that this election will go seat by seat, and Greenop-Roberts has identified a new metric that could help predict those outcomes.
His Australian Economic Review article compares the voter-by-voter results of the 2017 same-sex marriage poll with the Coalition’s voter-by-voter move in 2019.
He finds that the voters who swung the most to the Coalition in 2019 (see below) were the most opposed to same-sex marriage.
The statistically significant link predicted voting intention better than income, education, or unemployment.
That may still be the case, or we may not have perfected the science of predicting what’s to come yet, which could be just as well.
What happens in this election is up to all of us.
*Peter Martin is a visiting scholar at the Crawford School of Public Policy at the Australian National University.
This article first appeared on theconversation.com.