Australia’s system of preferential voting gives rise to the concept of the two-party-preferred vote.
The two-party-preferred vote is the total number of votes received by the Australian Labor Party (ALP) versus the Coalition (Liberal & National Parties) after all preferences have been distributed.
These totals are a combination of the primary votes (also known as “first preference” votes) and the preferences distributed from other candidates.
For example, suppose there are 3 candidates standing for election to a seat in the House of Representatives. They receive the following first preference votes:
Taking these figures, we say that Smith and Jones each received 45% of the primary vote.
An absolute majority (50% + 1) of votes is required to win under preferential voting. No candidate has secured that number, so the candidate with the least votes, Brown, is eliminated and the second preferences of Brown’s 10 primary votes are allocated between Smith and Jones.
Suppose 6 of those votes went to Smith and 4 went to Jones. The new tally would be:
Smith would now be declared the winner with 51% of the two-party-preferred vote. The 51 votes comprise 45 primary votes and 6 preference votes.
Historically, the two-party-preferred statistics are always calculated between the ALP and the Coalition because the overwhelming number of seats in the House of Representatives are won by those two groups. Sometimes the calculation may be between the ALP vs Greens, ALP vs Independent, Coalition vs Independent, Liberal vs National, etc.
The two-party-preferred figures are the only valid way to compare electorates and the swing required for them to be won or lost. For this reason, the Australian Electoral Commission calculates the two-party-preferred result in all electorates, even those where a candidate secured an absolute majority of the primary vote.
In the above example, Smith holds the seat with a margin of 1%.
The two-party-preferred statistics are also used to show the overall level of support for each of the main political groups. Two-party-preferred statistics for all federal elections since 1949 are available here.
The statistics show that it is possible for a party to win government without winning a majority of the two-party-preferred vote. This occurred in 1998, 1990, 1969, 1961 and 1954. The reason for this is that voter support is unevenly distributed amongst the House of Representatives electorates. The major parties can build up large majorities in individual seats which increases their overall vote but doesn’t win them extra seats.