Running into elections, the finding was the performance of these indicators varies and is impacted by the time periods being compared.
"This widely held belief is not grounded in fact.”
The clearest example of data slowing into the election is employment growth. But this is not the case in all instances. In particular, when comparing the three months leading into the election with the period four to seven months earlier, employment growth has tended to accelerate.
Despite this, ANZ Research thinks it reasonable to conclude employment growth slows in the immediate pre-election period.
This does not come at the expense of higher unemployment. Since 2000, the unemployment rate has had a slight tendency to be unchanged or lower in the three months before an election; but the difference is modest.
There is also little difference in monthly retail sales growth. For the sentiment indicators, consumer confidence has a tendency to rise while business conditions are mixed. The differences are small.
So this widely held belief is not grounded in fact.