Federal pendulum

elections AT


two decades of Newspolls

state votes at federal elections

Votes and seat
1949 - 2001


Newspoll &
Morgan graphs

preferential voting

federal election 2001

NSW election 2003

Vic election 2002

How important are preferences?

a long-winded aside

A paper last year on the 2001 election prepared by the Federal Parliament House Library - where I got the preference distribution numbers used here - gave rise to several misinterpretations and some correct ones.

One misapprehension is that Democrat and Green preferences didn't decide any seats. Rubbish. They decided plenty; any seat that wasn't won on primary votes can be said to have been decided on preferences, including Democrat and Green ones.

What is true is that it appears that in no seats where the Democrats or Greens "allocated" preferences to Labor (meaning they advised their voters to put Labor ahead of the Coalition on the ballot slip) does it seem that that "allocation" decided the result. That's mainly because that "allocation" makes little difference to the actual flow of preferences. Green voters in particular tend to decide for themselves.

The paper also showed that the number of seats where preferences made a net difference to the result - that is, where the candidate leading on primaries was overtaken by another with the help of preferences - was historically low. This has also been interpreted as pointing to preferences making little difference to the 2001 election result.

In fact, Green and Democrat preferences were vital to the ALP staving off a whopping defeat. In most elections you get a mix of seats fitting this category: between those where the ALP candidate came from behind on primaries to overtake the Coalition with the help of preferences, and the Coalition candidate doing the same to Labor.

In 2001 preferences changed results in only one direction: Labor's. In no seats (apart from 'three cornered contests', where Liberal and National candidates split the conservative primary vote but naturally preference each other) did the Coalition candidate win after trailing on primaries.

Now this was unusual; it pointed to the large number of Dem/Green preferences preventing Coalition candidates winning from behind with the help of others' preferences.

Make no mistake, Green and Democrat preferences made the difference in plenty of seats in 2001. In some ways, however, it's a semantic point, as many were disaffected Labor voters who, under compulsory preferential voting, were always going to give their vote back to Labor further down the ballot slip anyway.

Optional preferential voting (where you don't have to 'number every square') used to be an article of faith for the ALP (and is used in some states) but it would be deadly for federal Labor in the current climate.

Using minor party preference flows from the 1999 NSW election (where optional preferential is practiced) would have increased the election 2001 Coalition two party preferred lead from 50.9 to 49.1 to 51.9 to 48.1.

return to Newspoll's preferences

external links

Qld WA


Green (ABC)


NSW Parl (links)

Adam Carr



WA Uni
election database